MULTITASKING, COGNITIVE COORDINATION AND COGNITIVE SHIFTS DURING WEB SEARCHING

By

Jia Tina Du B.S., M.L.I.S

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Written under the direction of

Professor Amanda Spink

and approved by

________________________

________________________

________________________

________________________

Faculty of Science and Technology Queensland University of Technology

Brisbane, Australia

2010

Copyright © by Jia Tina Du All rights reserved

2010

SUPERVISORY PANEL

Principal Supervisor

Professor Amanda Spink Research Capacity Building Professor of Information Science

Vice-President - Queensland Academy of Arts & Sciences Faculty of Science and Technology

Queensland University of Technology

Associate Supervisor

Dr. Dian Tjondronegoro Faculty of Science and Technology

Queensland University of Technology

i

ii

STATEMENT OF ORIGINAL AUTHORSHIP The work contained in this thesis has not been previously submitted to meet requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made. ………………………………………………………….. Signature Jia Tina Du ………………………………… Date

iii

iv

ACKNOWLEDGEMENTS

My sincere appreciation is given most to my principal supervisor, Professor

Amanda Spink. Thank you Amanda for your continued support, guidance and

encouragement, both intellectually and socially. During the years of working with

you, I was challenged, encouraged, tested, and most importantly helped. Your

emphasis and attention to rigour have helped me grow towards being a capable

researcher. Great appreciation is also given to my associate supervisor, Dr. Dian

Tjondronegoro, for his efforts with my thesis, his confidence in me and his warm

friendship. I am grateful for the teaching opportunities that he offered me. I would

also like to thank my thesis final seminar panel members, Associate Professors

Sylvia Edwards, Yuefeng Li and Shlomo Geva, and two anonymous external

examiners, for their valuable contributions and advice.

Warm thanks are given to my fellow Ph.D. colleagues, Bhuva Narayan, Awadh

Alharbi and Kinley Kinley, for their generous help no matter whether in my research

or my life. Their company has made my Ph.D. journey a very pleasant and

memorable one.

Thanks to Assistant Professor Jingfeng Xia, from Indiana University, and Professor

Qinghua Zhu, from Nanjing University, for their insightful and challenging

comments on my thesis at various workshops. I would further like to thank the

group researchers, Associate Professor Yuefeng Li, Drs. Daniel Tao and Susan

Zhou, and Lifeng Ai, for their support and help during the years that I have studied

in the information science group.

Thanks to the School administrators Agatha Nucifora, Sara Thomas, Matt Williams,

Ilana Bolingford, and Therese Currell, who have assisted me with many of the

administrative aspects of this thesis and my Ph.D. study at QUT.

Appreciation is given to my husband Roger Yongjian Chen for accompanying me

through periods, whether happy or tough. Your optimism and persistence have

helped me to concentrate on completing this work. This is a cherished memory in

our continuing life.

Finally, I want to thank QUT for the financial support provided for this research. My

special thanks are due to the forty-two anonymous study participants for their time

v

and efforts. Without their participation and cooperation, this project would have not

been possible.

vi

DEDICATION

This dissertation is dedicated to my father, Weicheng Du, my mother, Huiling

Chen, and my husband, Roger Yongjian Chen, for their love and support.

vii

ABSTRACT

Multitasking, Cognitive Coordination and Cognitive Shifts During Web Searching

by Jia Tina Du

PhD Thesis Supervisor:

Professor Amanda Spink

As Web searching becomes more prolific for information access worldwide, we

need to better understand users’ Web searching behaviour and develop better

models of their interaction with Web search systems. Web search modelling is a

significant and important area of Web research. Searching on the Web is an

integral element of information behaviour and human–computer interaction. Web

searching includes multitasking processes, the allocation of cognitive resources

among several tasks, and shifts in cognitive, problem and knowledge states. In

addition to multitasking, cognitive coordination and cognitive shifts are also

important, but are under-explored aspects of Web searching. During the Web

searching process, beyond physical actions, users experience various cognitive

activities. Interactive Web searching involves many users’ cognitive shifts at

different information behaviour levels. Cognitive coordination allows users to trade

off the dependences among multiple information tasks and the resources available.

Much research has been conducted into Web searching. However, few studies

have modelled the nature of and relationship between multitasking, cognitive

coordination and cognitive shifts in the Web search context. Modelling how Web

users interact with Web search systems is vital for the development of more

effective Web IR systems. This study aims to model the relationship between

multitasking, cognitive coordination and cognitive shifts during Web searching. A

preliminary theoretical model is presented based on previous studies.

viii

The research is designed to validate the preliminary model. Forty-two study

participants were involved in the empirical study. A combination of data collection

instruments, including pre- and post-questionnaires, think-aloud protocols, search

logs, observations and interviews were employed to obtain users’ comprehensive

data during Web search interactions. Based on the grounded theory approach,

qualitative analysis methods including content analysis and verbal protocol analysis

were used to analyse the data. The findings were inferred through an analysis of

questionnaires, a transcription of think-aloud protocols, the Web search logs, and

notes on observations and interviews.

Five key findings emerged.

(1) Multitasking during Web searching was demonstrated as a two-dimensional

behaviour. The first dimension was represented as multiple information problems

searching by task switching. Users’ Web searching behaviour was a process of

multiple tasks switching, that is, from searching on one information problem to

searching another. The second dimension of multitasking behaviour was

represented as an information problem searching within multiple Web search

sessions. Users usually conducted Web searching on a complex information

problem by submitting multiple queries, using several Web search systems and

opening multiple windows/tabs.

(2) Cognitive shifts were the brain’s internal response to external stimuli. Cognitive

shifts were found as an essential element of searching interactions and users’ Web

searching behaviour. The study revealed two kinds of cognitive shifts. The first kind,

the holistic shift, included users’ perception on the information problem and overall

information evaluation before and after Web searching. The second kind, the state

shift, reflected users’ changes in focus between the different cognitive states during

ix

the course of Web searching. Cognitive states included users’ focus on the states

of topic, strategy, evaluation, view and overview.

(3) Three levels of cognitive coordination behaviour were identified: the information

task coordination level, the coordination mechanism level, and the strategy

coordination level. The three levels of cognitive coordination behaviour interplayed

to support multiple information tasks switching.

(4) An important relationship existed between multitasking, cognitive coordination

and cognitive shifts during Web searching. Cognitive coordination as a

management mechanism bound together other cognitive processes, including

multitasking and cognitive shifts, in order to move through users’ Web searching

process.

(5) Web search interaction was shown to be a multitasking process which included

information problems ordering, task switching and task and mental coordinating;

also, at a deeper level, cognitive shifts took place. Cognitive coordination was the

hinge behaviour linking multitasking and cognitive shifts. Without cognitive

coordination, neither multitasking Web searching behaviour nor the complicated

mental process of cognitive shifting could occur.

The preliminary model was revisited with these empirical findings. A revised

theoretical model (MCC Model) was built to illustrate the relationship between

multitasking, cognitive coordination and cognitive shifts during Web searching.

Implications and limitations of the study are also discussed, along with future

research work.

x

KEYWORDS

Interactive information retrieval (IR), Web search interaction, Web search behaviour,

Multitasking, Cognitive shifts, Cognitive coordination, Cognition, Web search

modelling, Human information behaviour, Human–computer interaction

xi

TABLE OF CONTENTS SUPERVISORY PANEL ............................................................................................ i STATEMENT OF ORIGINAL AUTHORSHIP .......................................................... iii ACKNOWLEDGEMENTS ......................................................................................... v DEDICATION...........................................................................................................vii ABSTRACT.............................................................................................................viii KEYWORDS ............................................................................................................ xi TABLE OF CONTENTS ...........................................................................................xii LIST OF TABLES ...................................................................................................xvii LIST OF FIGURES .................................................................................................xix Chapter 1 Introduction.....................................................................................1

1.1 Problem Statement .........................................................................1 1.2 Aims of Research ...........................................................................4 1.3 Research Questions .......................................................................4 1.4 Contributions and Significances .....................................................5 1.5 Structure of the Work......................................................................7

Chapter 2 Literature Review............................................................................9

2.1 Introduction.....................................................................................9 2.2 Interactive Information Retrieval (IR)............................................10

2.2.1 Overview.......................................................................................10 2.2.2 Interactive IR Model...................................................................... 11

2.2.2.1 Bates (1989) Berry-picking Model ................................................11 2.2.2.2 Ingwersen (1992, 1996) Cognitive IR interaction Model.............. 12 2.2.2.3 Saracevic (1996) Stratified Interactive IR Model ......................... 13 2.2.2.4 Belkin (1996) Episodic Model of IR Interaction............................ 14 2.2.2.5 Spink (1997) Interactive Feedback and Search Process Model.. 14

xii

2.3 Web Search ................................................................................. 16

2.3.1 Overview ...................................................................................... 16 2.3.2 Web Search Model....................................................................... 19

2.3.2.1 Choo et al. (2000b) Web Behavioural Model ...............................20 2.3.2.2 Wang et al. (2000) Multidimensional User–Web Interaction Model

.....................................................................................................20 2.3.2.3 Ford et al. (2001, 2005) Individual User Differences Web IR model

.....................................................................................................21 2.3.2.4 Knight & Spink (2008) Web IR Model...........................................21 2.3.2.5 Park (2008) Prioritising and Coordinating Information Behaviour

Model ...........................................................................................22 2.4 Multitasking .................................................................................. 24

2.4.1 Multitasking Research in Cognitive Science and Psychology...... 24 2.4.2 Multitasking Studies in Human Information Behaviour................. 25 2.4.3 Multitasking Studies in Web and Information Retrieval ................ 27 2.4.4 Coordination Viewpoint in Multitasking Research ........................ 29

2.5 Cognitive Coordination................................................................. 31

2.5.1 Research on Coordination ........................................................... 31 2.5.2 Cognitive Coordination Mechanism in Psychology ...................... 34 2.5.3 Psychological Concepts in Support of Coordinating Multiple Tasks. ..................................................................................................... 36 2.5.4 Elements within Coordinating Web Searching Process ............... 38

2.6 Cognitive Shifts ............................................................................ 41

2.6.1 Cognitive Shifts in Interactive IR .................................................. 41 2.6.2 Variables Affecting Cognitive Shifts.............................................. 43

2.7 Relevant Dissertations and Theses ............................................. 46 2.8 Theoretical Model ........................................................................ 52 2.9 Chapter Summary........................................................................ 55

xiii

Chapter 3 Research Design..........................................................................57

3.1 Introduction...................................................................................57 3.2 Data Collection .............................................................................58

3.2.1 Study Participants.........................................................................58 3.2.2 Research Setting ..........................................................................59 3.2.3 Information Problem in Web Searching Context ..........................60 3.2.4 Web Browser and Web Search System .......................................62 3.2.5 Time Constraints...........................................................................63 3.2.6 Instruments...................................................................................64 3.2.7 Procedures ...................................................................................71

3.3 Data Analysis................................................................................73

3.3.1 Overview of Methods....................................................................73 3.3.2 Identification of Variables..............................................................75 3.3.3 Classification of Variables.............................................................78 3.3.4 Open Coding ................................................................................80

3.4 Verification of Methodology ..........................................................83

3.4.1 Credibility......................................................................................83 3.4.2 Transferability ...............................................................................84 3.4.3 Dependability................................................................................84

3.5 Chapter Summary ........................................................................86

Chapter 4 Results..........................................................................................87

4.1 Introduction...................................................................................87 4.2 Demographic Data........................................................................88 4.3 Web Using Experience .................................................................92 4.4 Multitasking Behaviour during Web Searching.............................98

4.4.1 Multiple Information Problems (IP) ...............................................99

xiv

4.4.2 Factors Affecting Information Problem Search Ordering.............111 4.4.2.1 Information Problem Search Ordering ....................................... 111 4.4.2.2 Reasons for Information Problem Search Ordering................... 113

4.4.3 Evolving Information Problem .................................................... 123 4.4.4 Information Problem Searching Task Switching......................... 130

4.4.4.1 Types of Information Problem Searching Task...........................130 4.4.4.2 Task Switching Pattern...............................................................136 4.4.4.3 Reasons for Information Problem Searching Task Switching ....140

4.4.5 Multiple Web Search Sessions .................................................. 148 4.4.6 Summary.................................................................................... 152

4.5 Cognitive Shifts during Web Searching...................................... 154

4.5.1 Holistic Cognitive Shifts.............................................................. 155 4.5.2 Cognitive State Shifts ................................................................. 164

4.5.2.1 Types of Cognitive State ............................................................165 4.5.2.2 Shifts of Cognitive State .............................................................167

4.5.3 Summary.................................................................................... 170

4.6 Cognitive Coordination during Web Search ............................... 171

4.6.1 Cognitive Coordination Results Overview.................................. 172 4.6.2 Level One: Information Task Coordination (TC)......................... 174 4.6.3 Level Two: Cognitive Coordination Mechanism (CM) ................ 175 4.6.4 Level Three: Cognitive Strategy Coordination (SC) ................... 177 4.6.5 Cognitive Coordination Behaviour on Three Levels................... 182

4.6.5.1 Frequency of Occurrences.........................................................186 4.6.5.2 Transition Analysis of Cognitive Coordination Levels ................198

4.6.6 Summary.................................................................................... 210

4.7 Chapter Summary.......................................................................211

xv

Chapter 5 Discussion ..................................................................................213 5.1 Key Findings of the Study...........................................................213

5.1.1 Multitasking during Web Search.................................................213 5.1.2 Cognitive Shifts during Web Search...........................................219 5.1.3 Cognitive Coordination during Web Search ...............................224

5.2 Revised Relationship Model .......................................................231 5.3 Theoretical Implications..............................................................235

5.3.1 Implications for Multiple Search Sessions Model .......................235 5.3.2 Implications for Cognitive IR Model: The Role of Cognitive Coordination ...............................................................................236

Chapter 6 Conclusion and Further Research..............................................239

6.1 Summary of the Study ................................................................239 6.2 Significance of the Study ............................................................241 6.3 Contributions of the Study ..........................................................241 6.4 Limitations ..................................................................................243 6.5 Further Research........................................................................244

Appendix A. Participation Information and Consent Form ....................................247 Appendix B. Pre-Web Search Questionnaire........................................................251 Appendix C. Post-Web Search Questionnaire ......................................................257 Appendix D. Semi-structure Interview Questions .................................................265 Appendix E. Web Searching Process as Flowchart (Examples)...........................268 Appendix F. Steps of Transition between Cognitive Coordination Behaviours......283 Appendix G. Glossary ...........................................................................................315 Bibliography ..........................................................................................................317 Curriculum Vita......................................................................................................337

xvi

LIST OF TABLES Table 2-1. Search terms and amount ..................................................................... 46 Table 2-2. Statistics on dissertations and theses.................................................... 47 Table 2-3. The annual number of published dissertation and thesis related to this study....................................................................................................................... 47 Table 3-1. Coding scheme...................................................................................... 79 Table 4-1. Study participant profiles........................................................................ 89 Table 4-2. Number of study participants in each gender category.......................... 90 Table 4-3. Number of study participants in each age category............................... 90 Table 4-4. Number of study participants in each academic status category........... 90 Table 4-5. Number of study participants in each faculty category .......................... 91 Table 4-6. Years of Web use by study participants ................................................. 92 Table 4-7. Number of study participants: frequently used Web browsers............... 92 Table 4-8. Number of employed Web search systems ........................................... 93 Table 4-9. The employed Web search systems during the Web searches............. 94 Table 4-10. Study participants’ information problems ........................................... 100 Table 4-11. Information problem topic area .......................................................... 108 Table 4-12. Status of original information problems prior to the searching........... 109 Table 4-13. Number of study participants with related or unrelated information problems searching ...............................................................................................110 Table 4-14. Factors affecting information problem search ordering ......................114 Table 4-15. Summary of the factors affecting multiple information problems search ordering .................................................................................................................117 Table 4-16. Evolving information problems generated per study participant ........ 125 Table 4-17. Number of study participants with evolving or non-evolving information problems developed ............................................................................................. 128 Table 4-18. Information problem searching task switching pattern....................... 137 Table 4-19. The percentage of each information problem searching task pattern 140 Table 4-20. Reasons for information problem searching task switching............... 141

xvii

Table 4-21. Summary of reasons for searching tasks switching ...........................143 Table 4-22. Number of search sessions conducted during the current Web searching ..............................................................................................................149 Table 4-23. Mean queries, Web search systems, and windows/tabs for an information problem searching per study participant ............................................150 Table 4-24. Holistic cognitive shifts data ...............................................................156 Table 4-25. Cognitive state data............................................................................166 Table 4-26. Shifts between cognitive states ..........................................................168 Table 4-27.Summary of cognitive state shifts occurrence.....................................169 Table 4-28. Overall results ....................................................................................173 Table 4-29. Time allocation between multiple information problems searching (Global Strategy) ...................................................................................................179 Table 4-30. Cognitive coordination behaviour data...............................................187 Table 4-31. Number and type of cognitive coordination occurrences per study participant .............................................................................................................195 Table 4-32. Summary of cognitive coordination types and occurrences...............197 Table 4-33. Summary of cognitive coordination transition steps per study participant..............................................................................................................................199 Table 4-34. Sequence on cognitive coordination levels per study participant (Study Participants 1 to 21) ..............................................................................................201 Table 4-35. Sequence analysis on cognitive coordination levels per study participant (Study Participants 22 to 42) ..............................................................204 Table 4-36. Summary of each type of cognitive coordination level sequences.....208 Table 4-37. Summary of cognitive coordination transition on three levels ............209 Table 5-1. Cognitive State Shift (this study) vs. Information Problem Shift (Robins, 2000).....................................................................................................................221 Table 5-2. Cognitive coordination mechanism vs. feedback mechanism.............225

xviii

LIST OF FIGURES Figure 2-1. Year-Number (from 1970 to 2008) ....................................................... 48 Figure 2-2. A conceptual model of multitasking, cognitive coordination and cognitive shifts during Web searching ................................................................................... 52 Figure 2-3. Dynamic Web search interactions........................................................ 53 Figure 3-1. An example of an open coding outcome .............................................. 81 Figure 4-1. Ordering of information problems........................................................112 Figure 4-2. Forty-two study participants’ information problem searching task switching............................................................................................................... 132 Figure 5-1. Two-dimensional Multitasking Web search behaviour ....................... 214 Figure 5-2. Holistic cognitive shifts and cognitive state shifts............................... 223 Figure 5-3. Flowchart example of interplay between the three coordination levels (Study Participant 36) ........................................................................................... 228 Figure 5-4. Shifts between the three coordination levels (Study Participant 36) .. 229 Figure 5-5. Interplay between three cognitive coordination levels........................ 230 Figure 5-6. Multitasking, Cognitive Coordination and Cognitive Shifts (MCC) Model.............................................................................................................................. 232

xix

xx

Chapter 1 Introduction

This chapter provides an overview of the dissertation research, including the

problem statement, the aims of research, the research questions, the significance

of the study, and the contribution to research. Lastly, the structure of the document

is outlined.

1.1 Problem Statement

Humans have been seeking, organising and using information as they learned and

evolved their patterns of information behaviour while resolving problems for survival,

work and everyday life (Case, 2002). Information behaviour studies have become a

key research area within the field of information science. Researchers have sought

to study the human behaviour related to seeking, searching, foraging, retrieving,

organising and using information. Information seeking studies focus on purposive

information behaviour, while information searching is a sub-set of information

behaviour, and is associated with the human–system interaction process.

In recent years, there has been an explosive growth of information on the Web.

Web searching is a predominant tool and channel for people to acquire information.

Web searching has the characteristics of both information seeking and information

searching. Web searching is represented as a series of actions between logging on

and logging off a Web search system.

Research shows that users’ actions are engaged in multitasking information

behaviour during Web searching episodes (Ozmutlu, Ozmutlu & Spink, 2003b;

Spink & Jansen, 2004; Spink, Ozmutlu & Ozmutlu, 2002; Spink, Park, Jansen &

Pedersen, 2006; Spink, Park & Koshman, 2006). Web searching behaviour is

described as a multitasking process which includes searching for information

1

related to one information task, and then switching to search for items on another

information task (Spink, Cole & Waller, 2008). Multitasking is a prevalent

phenomenon in the Web searching context. Multitasking has been recognised as

important user behaviour during Web searching (Ozmutlu, Ozmutlu & Spink, 2003a,

b; Spink, Bateman & Greisdorf, 1999; Spink, Park, Jansen & Pedersen, 2006;

Spink, Park & Koshman, 2006). Web users prefer to search for multiple information

problems concurrently during single or multiple Web search sessions.

User studies can tell us a lot about the actual and typical characteristics of the Web

searching process. Web search studies are concerned with how people search the

Web, especially the cognitive processes involved in Web search activities. Recently,

efforts have been made to build Web search models with an emphasis on

illustrating the dynamic interaction between the information problem, the user, and

the information environment, and on the iterative effect on user search strategies,

processes and outcomes. Ford, Miller and Moss (2001, 2005) modelled how users’

individual differences have an effect on Web search performance. Originally, Spink

and her colleagues developed a model of Web search as multitasking (Spink &

Park, 2005; Spink, Park & Koshman, 2006), showing that the user may pool

together more than one related or unrelated topic when searching on the Web.

There is still a significant gap between the Web search model and the real users’

dynamic Web searching process. Web searching is proposed as an important

element of interactive information retrieval which includes multitasking processes,

and the allocation of cognitive resources among several tasks, and shifts in

cognitive, problem and knowledge states (Du & Spink, 2009). Multitasking involves

cognitive shifts in task focus. During the Web searching process, users experience

various cognitive, emotional and physical reactions when they identify a gap in

knowledge that needs to be filled with the information they are searching for (Spink

& Dee, 2007). Interactive Web searching involves many human shifts on cognition

2

at different levels of information behaviour (Du & Spink, 2009; Robins, 2000; Spink,

2002; Spink & Dee, 2007). Cognitive shifts are an important but under-explored

research area for understanding the cognitive processes associated with Web

searching. The identification of types of cognitive shifts may be meaningful in

understanding the outcomes of user–Web interaction.

Wickens and Gopher (1977) argued that one of the important insights into people’s

ability to dual-task was that while there was some interference between the two

tasks that were being performed people could consciously trade off performing one

task for the other. The key point of such trading off stems from the person’s

coordination capability. Cognitive coordination allows humans to manage

dependences among information tasks and the resources available. A key issue for

cognitive coordination research in the Web search context concerns those

coordination mechanisms that move users through a multitasking Web search,

while experiencing various shifts in cognition. An exploration of this issue may be

significant in developing a comprehensive, deep understanding of Web searching

behaviour. Spink and Du (2007) proposed that humans must cognitively coordinate

a number of elements, both internal (cognitive) and external (environmental), into a

coherent Web search process. However, it is not clear what kinds of elements are

cognitively coordinated, or how these elements interplay in order to achieve a

coherent Web search process.

As Web searching becomes the predominant form of information access worldwide,

we need to develop better user interaction models of a Web search. Much research

has been conducted into Web searching. However, few studies have modelled the

relationship between multitasking, cognitive coordination and cognitive shifts in a

Web search context. And this is something that needs to be explored. Cognitive

coordination, in conjunction with multitasking and cognitive shifting, may form a

3

theoretical framework for understanding how Web searching behaviour is

constructed.

1.2 Aims of Research

Broadly, the objective of this research is to develop a framework for theory building

and research work in Web searching involving multitasking, cognitive coordination

and cognitive shifts. Specifically, this study aims to model the relationship between

multitasking, cognitive coordination and cognitive shifts during a Web search. The

model aims to depict how users conduct multitasking information behaviours and

how they perform cognitive coordination between multiple information problems

and the resources available. The study also investigates the various types of

cognitive shifts and how these shifts occur during Web searching. Finally, the

research is expected to discover how multitasking, cognitive coordination and

cognitive shifts interplay to influence users’ behaviours during Web search

interactions.

1.3 Research Questions

The major research problem underpinning this study is:

What is the relationship between multitasking, cognitive coordination, and cognitive

shifts during Web searching?

Four minor research questions addressed in this study are:

(1) How do users conduct their Web searches on multiple information problems?

(2) What types of cognitive shifts occur during Web searching?

(3) What levels of cognitive coordination occur during Web searching?

4

(4) How do multitasking, cognitive shifts and cognitive coordination interplay during

Web searching?

1.4 Contributions and Significances

Saracevic, Kantor, Chamis and Trivison (1988) suggested that the key to the future

of information systems and searching processes lay not in the increased

sophistication of technology, but in the increased understanding of human

involvement in relation to information. Users play a predominant role during Web

searching interactions. These conclusions form the rationale of this study and

indicate why this rationale is significant.

1) This dissertation contributes theoretically to interactive IR research. By providing

a comprehensive picture of a Web searching interaction, the exploration of

multitasking, cognitive coordination and cognitive shifting extends the user Web

search model to include cognitive mechanisms within searching interactions.

2) This study contributes to the formal characterisation and better understanding of

elements and processes involved in the Web searching process in relation to

cognitive IR interactions. Such an understanding is fundamental to basic research

directed toward theories, models, and practices on human information behaviour

involving Web IR systems, and to applied development directed toward

improvement of human–computer interaction involved in Web searching.

3) This research analyses the users’ Web searching as they attempt to solve their

information problems, and looks for patterns involving multitasking, cognitive

coordination and cognitive shifts. Comprehensive patterns lead to the development

of a theoretical model depicting the inter-relationship between multitasking,

cognitive coordination and cognitive shifts. This study theoretically underpins

5

interactive Web searching studies, and it might also expand the theoretical basis of

multitasking and coordination theory in cognitive science.

4) In practical terms, the results of the study will not only show how people search

information which might help users understand their own Web searching, but will

also provide insights into the design of Web technologies. Modelling how users

interact with Web search engines from different cognitive perspectives is important

in the development of more intelligent and effective Web based IR systems.

5) Users’ multitasking, coordination and management of different information

search tasks is little understood or supported by current search technologies (Spink,

Park & Koshman, 2006). Modelling users’ cognitive Web searching process that

integrates multitasking, cognitive coordination and cognitive shifts could impact on

the development of technologies that support coherent Web searching and that

lead to improvements in the performance of search technology. Web search

systems must evolve out of the users’ needs and the common characteristics of

Web searching, rather than from expecting the user to adapt to their singularities.

6) The Web search patterns and models proposed in this study allow the possibility

of predicting users’ Web searching. Web search engine companies could benefit

from knowing how their products are used, which may help designers to redesign

and reconstruct search engines in order to attract visitors to their sites. The efforts

to model Web searching behaviour are important for the design of Web search

systems and for information providers in general. If a system is familiar with users’

behavioural patterns, it may more easily adapt and personalise users’ interactive

process.

6

1.5 Structure of the Work

The thesis comprises six main chapters. Chapter 1 provides an introduction to the

thesis research. Chapter 2 presents a review of the literature, with a focus on the

behavioural characteristics of multitasking, cognitive coordination, and cognitive

shifts research. In particular, the research is contextualised information retrieval.

Chapter 3 describes the research design, including data collection instruments,

selection of study participants, data analysis techniques, and justification of the

methodology. Results are reported in Chapter 4. Chapter 5 contains a discussion of

the key findings. The terms of these findings are guided by the research questions.

Finally, in Chapter 6, a conclusion is made which outlines how this research

improves our knowledge of Web search behaviour, the research’s limitations and

some further research directions.

7

8

Chapter 2 Literature Review

2.1 Introduction

This study is concerned with users’ actual information search behaviours and

processes within the context of searching the Web and focuses on the underlying

coordination mechanisms which may be involved in users’ Web search interactions.

This chapter contains a review of interactive information retrieval and Web search

literature; a discussion of representative interactive IR models and recent Web

search models; and research about multitasking, cognitive coordination and

cognitive shifts in general, and in the context of interactive information retrieval in

particular. Investigation into relevant state of the art dissertations and theses is then

presented. This is followed by an outline of the research gap identified from the

critical analysis of the previous studies, and an argument for the research proposed

will be put forward. The last section will explore the theoretical model which has

emerged: this is based on the analysis of the studies reviewed and forms the

foundation for the theoretical framework in this study.

9

2.2 Interactive Information Retrieval (IR)

2.2.1 Overview

Information retrieval (IR) is a fundamental component of information behaviour

(Ruthven, 2006). Information retrieval is defined as the process involved in

representation, storage, searching, finding, filtering and presentation of potential

information perceived to be relevant to a requirement of information desired by a

user in context (Ingwersen & Jarvelin, 2005). Interactive information retrieval is

viewed as information acquisition via formal channels and in organised knowledge

sources such as information systems like the Internet (Ingwersen & Jarvelin, 2005).

Research on interactive information retrieval originated in the early 1980s, when a

theoretical direction was generated to address the dynamic of the searcher/user

and system interaction. In the last two decades, research on interactive IR has

mainly concentrated on understanding the ways and processes that

searchers/users iteratively search IR systems, modelling user and search

intermediary behaviours and using information behaviour models to design

automated intermediary devices and IR systems which train and educate

information professionals on information user literacy.

Searcher/user-oriented IR research has been aimed at understanding the

searchers/users’ behaviour and the dynamic and evolving search characteristics

(Bates, 1989; Fidel, 1985). Efforts have been made to recognise the complex

dynamics and interactive processes involved in this communication and to look for

a more accurate way of capturing the cognitive developments involved in the

user/searcher’s information problems. Past attempts to build user and search

intermediary interactive IR models were carried out in more natural situations

instead of simulated laboratory settings (Belkin, 1984; Saracevic & Kantor, 1988).

10

The recording of “talk aloud” discourse and interviews became principal tools of

experimental methodology. The scope of the study participant group was extended

to include users from all levels of society, going beyond the experienced

researchers normally analysed by traditional IR studies.

Subsequently, a cognitive focus took place in the interactive IR research. Saracevic

and Kantor (1988) correlated users’ and searchers’ cognitive styles to their search

results. Ingwersen (1992) described the impact of cognitive science on IR research,

analysing IR as a process involving cognitive states and complex interactions. In

1996, Ingwersen (1996) proposed a first formulation of a cognitive theory for

information retrieval interaction. Later, these models led to a more sophisticated

model of the cognitive communication system (Ingwersen, 2001). In 2005,

Ingwersen and Jarvelin built an integrated cognitive information seeking and

retrieval (IS&R) research framework involving the concepts of “cognitive actor” and

“context” (Ingwersen & Jarvelin, 2005, p. 19).

2.2.2 Interactive IR Model

This section presents a discussion about several representative interactive IR

models. The interactive nature of the users’ information searching behaviour has

become a primary focus of the interactive IR models that have been developed

since late 1980s. The IR models have an emphasis on the dynamic interaction

between the searcher/user, the information need, and the information environment.

2.2.2.1 Bates (1989) Berry-picking Model Bates (1989) described the online search as an evolving berry-picking process in

which the changes in the search strategy were due to the experience of a variety of

sources and were the result of new information encountered which provided new

ideas and directions to the original search query. Her model illustrates that the

11

results of each search query would provoke cognitive thoughts in the user to make

continual judgments regarding the relevance and interoperability as the information

was sought and used.

Based on the basic implication that the information searcher is also the information

user, the berry-picking model more closely discovers users’ actual information

behaviour than previous traditional linear IR models, in that it considers the users’

dynamic and continuous cognitive responses during IR interaction and their effects

on the follow-up search queries. Yet Bates’ ideas on the model have never been

empirically validated. Thus, whether the users’ cognitive thought is the factor

resulting in changes to the search query still remains a question. Additionally, the

forms of the users’ cognitive response have not yet been investigated. A major aim

of this dissertation is to investigate the interplay of multiple cognitive aspects on the

users’ information search behaviour.

2.2.2.2 Ingwersen (1992, 1996) Cognitive IR interaction Model As one of the earliest IR interaction models, Ingwersen’s (1992, 1996) model was

the first to illustrate that a dynamic interaction process occurred at multiple levels

within the “cognitive space” of the user and the “information space” of the IR system.

The multiple levels of interaction were said to occur not only between the user and

the IR system, but also between the user and the information objects within the

system. The model focused more on understanding the actual information system

being used during the interactive cognitive processes. Ingwersen’s model provides

a way to understand the process of how information is being retrieved and

ultimately used.

Ingwersen (1996) claimed that a wide ranging influence of factors should be

considered in IR research, such as social environment, IR system, information

12

objects, search intermediary and user. He incorporated all these variables into the

notion of polyrepresentation. His models have presented a reasonable synthesis of

studies regarding IR interactions with empirical evidence. As Robins (2000) pointed

out, however, the problem with Ingwersen’s model lies in determining the way to get

input from the users’ cognitive space into the request model builder, since the

differences among the four components of user cognitive space put forth by

Ingwersen are subtle. The study presented in this dissertation attempts to better

understand the components of the users’ cognitive space in Ingwersen’s model.

2.2.2.3 Saracevic (1996) Stratified Interactive IR Model Conceptually borrowed from human–computer interaction, Saracevic’s (1996)

theoretical model described the IR interaction between the user, the IR system and

the information objects through the system. His model was also based on the

assumption that users interacted with IR systems in order to use information. He

originally emphasized that understanding the reason why a user sought out

information was an important part of discerning the influencing factors during that

interaction.

The stratified model involved three strata of IR interaction:

• a surface level—the interactions between the user and the interface of the

IR system.

• a cognitive level—user-made judgments regarding the results given by the

system. Both the users’ thinking and system’s information objects were

identified as cognitive entities.

• a situational level—a context-driven interaction, influenced by the need for

original information and how the user or system might categorise, or even

iteratively change the need.

13

In terms of a theoretical framework, it was a comprehensive model covering all of

the three IR interaction levels. However, the details on what and how the changes

occurred as a process of IR interaction were not fully established.

2.2.2.4 Belkin (1996) Episodic Model of IR Interaction Belkin is another pioneer who advanced the interactive viewpoint in information

retrieval. Belkin’s (1996) episodic model was based on his anomalous states of

knowledge (ASK) hypothesis (Belkin, 1980), which modelled a user who turned to

the information system with a high level of cognitive uncertainty. The users’ state of

knowledge was anomalous. As such, he/she could not adequately present his/her

information need to the information system. Belkin considered that the real problem

in IR was how to represent the users’ anomalous state of knowledge.

Interaction with the information system led to the users’ altered state of knowledge,

which enabled him/her to define, reformulate and re-focus the information need,

and eventually to contribute to his/her underlying information problem solving.

Although the model provides a research framework for interactive IR, it lacks a

treatment of the social/contextual facets of the user information problem and the

corresponding effects on IR interactions.

2.2.2.5 Spink (1997) Interactive Feedback and Search Process Model

While focusing on understanding how the interactive process actually took place,

Spink’s search process model (1997) was developed based on the empirical

research. User judgments, search strategies and the interactive feedback loops

within the search process were presented. The model reflected that a variety of

feedback mechanisms were the major influencing factors in the interactive IR

process.

14

Importantly, the model demonstrated that a users’ interaction with the system could

consist of multiple feedback transactions, leading to additional inputs or queries

which could in turn result in different feedback and new inputs. The strength of

Spink’s model is that it observed IR from an interactive point of view and, in

particular, that it provided a complete investigation of the feedback mechanisms.

Yet the feedback loops were identified within the discussions between the user and

the search intermediary. That is to say, in her model, a feedback loop was incurred

when one of the participants gave feedback to the other, followed by a judgment or

an action taken. The recognition of feedback was based on analysis of the

discourses between the two participants. Apparently, it was not applicable under a

Web searching context in which the searcher is the actual user. Another weakness

of the model is that it lacks appropriate explanations about the underlying cognitive

changes during the occurrence of feedback loops. The present study seeks to

address both of these issues.

The above interactive IR models have laid the foundation for the development of

later Web search models, which are discussed in the following section.

15

2.3 Web Search

As special forms of IR systems, Web search engines are designed specifically for

the hypermedia environment of the Web and are considered to be the major portals

for users of the Web, with 71% of Web users accessing Web search engines to

locate other Web sites in 2000 (Spink & Jansen, 2004). Thus it is critical to have an

understanding of how people search information on Web search systems. Web

searching is the context for the present research project. The study of Web

searching focuses mainly on users’ information behaviour while interacting with

Web search systems.

2.3.1 Overview

Web searching refers to users' actions during the logged-on to logged-off period on

a Web information system (Spink, Ozmutlu & Ozmutlu, 2002). The current state of

Web search studies can be summarised within the following categories: (1) Web

search behaviour; (2) single Web site search studies; (3) information foraging

studies; (4) children’s Web search behaviour; (5) Web search training and learning;

and (6) Web search evaluation (Spink & Jansen, 2004).

Amongst the above categories, Web search behaviour studies are becoming a

flourishing research area which is concerned with why and how people search the

Web and the process of their interaction with Web search tools. Researchers have

argued that Web searching behaviour is separated into information searching,

seeking and retrieving behaviour on the Web. A significant number of Web users

begin their interactions with the Web with searching and retrieving type behaviours,

such as a query formulation, and then shift to seeking type moves, such as

16

scanning or browsing the returned results by the search engine (Choo, Detlor &

Turnbull, 2000b; Choo & Marton, 2003).

Research has been conducted in order to understand the elements and patterns of

the users’ Web search. For instance, Catledge and Pitkow (1995), as well as

Montgomery and Faloutsos (2001), investigated Web search browsing patterns of

adult Web searchers and concluded that they relied mainly on hyperlink structures;

Tauscher and Greenberg (1997) revealed that people repeatedly returned to the

same Web pages and conducted short searches; Hawk and Wang (1999) identified

people’s Web searching patterns with ten strategies, including exploring, link

following, back and forward movements, engine seeking and using, shortcut

seeking and surveying or scanning Web pages; Rieh and Xie (2001) examined

patterns and sequences of query reformulation during users’ interactions with the

Excite Web search engine; White and Iivonen (2002) pointed out that users regard

closed or predictable topics as easy to search and open or unpredictable topics as

difficult to search.

The users’ tasks and goals are mentioned and investigated in the studies. Rose

(2006) investigated many of the characteristics of users' Web search behaviour,

including the variety of information seeking goals, the cultural and situational

context of the search and the iterative nature of the search task. Kim (2008) found

that, to experienced Web users, the effects of emotional control and search tasks

were significant on Web search behaviour, but not on the search performance, and

that the effects of users' emotional control on the search behaviour varied

depending on search tasks.

Some of the individual differences which are viewed as important factors affecting

the Web users’ searching are personality (Amichai-Hamburger, 2002), age and

gender (Roy, Taylor & Chi, 2004; Tillotson, Cherry & Clinton, 1995), and level of

17

experience (Hill & Hannafin, 1997; Hoelscher & Strube, 2000; Kellogg & Richards,

1995; Navarro-Prieto, Scaife & Rogers, 1999). Ford, Miller and Moss (2001, 2002,

2003 & 2005) explored the relationship between Web search strategies and human

individual differences from multi-perspectives of cognitive and demographic factors,

Internet attitudes and approaches to study. Morgan (2008) explored Web searching

behaviour in correlation with individual differences within the context of health

information searching. This evaluated the ways in which individual differences

influence user Web search behaviour, determining that differences such as race,

gender, age, socio economic class and geographic location all influence searching

behaviour. However, it was found that those factors work together to influence

behaviour, rather than independently.

In particular, research identifies a link between individual differences in cognitive

styles and their Web search outcomes. Cognitive styles are characteristic ways in

which different individuals engage in information processing and representation. As

to the holist/analytic cognitive style, in relation to Web searching, Ford and Miller

(1996) found relationships between a holist cognitive style and a preference for

browsing over keyword searching, a broad-based approach to Internet exploration

and lower levels of reported distraction by irrelevant material. Wang, Hawk and

Tenopir (2000) linked the holist cognitive style to greater levels of difficulty and

confusion experienced during Web searching. Palmquist and Kim (2000) studied

the effects of cognitive style on search performance and found that less time was

taken and fewer nodes were traversed in locating information by analytic novice

searchers.

Kim and Allen (2002) found that the cognitive individual differences during Web

searching consisted of holist/analytic and imager/verbaliser cognitive styles and

cognitive complexity. In terms of a verbaliser/imager cognitive style, links were

found between an imager cognitive style and low levels of reported disorientation,

18

information overload, avoidance of unplanned Internet browsing and poor retrieval

effectiveness in a study of Web searching that focused on retrieval performance

rather than strategy and the use of the internet when required to rather than

through intrinsic interest (Ford & Miller, 1996; Ford, Miller & Moss, 2001).

Modelling the mutual impact of users’ information behaviour and Web searching

process is another important research topic that several studies have pursued.

Choo, Detlor and Turnbull (2000a) developed one of the first behavioural models of

Web interaction which depicted how users translated their information needs into

search strategies. Wang, Hawk and Tenopir (2000) designed a multi-dimensional

model of user–Web interaction, consisting of users, the interface and the Web.

Hodkindon and Kiel (2003) modelled consumer Web search behaviour, including

personal demographic, behavioural, use and experience variables. Lau and Coiera

(2006) developed a Bayesian model for predicting the impact of Web searching on

human decision making. Aula and Nordhausen (2006) chose to model Web search

success with the concept of task completion speed (TCS) and the results showed

that the variables related to Web experience had the expected effects on TCS.

2.3.2 Web Search Model

Web search models should be different from previous information seeking and

retrieving models, which partly derive from the distinct users under the two contexts.

Before the advent of the Web, IR system searchers comprised information

professionals, including intermediaries and educated professionals, who usually

had formal training in developing appropriate search queries or retrieval strategies.

The enormous growth of the Web has since provided an environment for a whole

new user group with a vast computational capacity to search for information instead

of asking librarians for help related to how to refine a query or improve a search

result. New dynamic variables of different users’ interactions have to be considered

19

that involve the users’ cognitive ability, personality, information task, and search

outcomes, all of which are key to developing sound Web searching models (Knight

& Spink, 2008).

This section provides a discussion of more recent Web search models. It seeks to

trace their contribution to theoretical construction on information seeking, including

Web searching behaviour research.

2.3.2.1 Choo et al. (2000b) Web Behavioural Model As one of the earliest behavioural models of Web interaction, Choo, Detlor and

Turnbull’s model (2000b) depicted how knowledge workers used the Web to seek

information. The behavioural model of information seeking on the Web combined

and extended Aguilar and Francis' (1967) modes of scanning and Ellis et al.'s

seeking behaviours (Ellis, 1989; Ellis, Cox & Hall, 1993), and consisted of the four

main modes of undirected viewing, conditioned viewing, informal search and formal

search. For each mode, information seeking activities or moves occurred frequently

and included starting, chaining, browsing, differentiating, monitoring and extracting.

The strength of this model is that it related motivations (the strategies and reasons

for viewing and searching) and moves (the tactics used to find and use information)

and this was helpful in analyzing Web-based information seeking.

2.3.2.2 Wang et al. (2000) Multidimensional User–Web Interaction Model Taking a holistic approach, Wang, Hawk and Tenopir (2000) proposed a multi-

dimensional user–Web interaction model encompassing users, the interface and

the Web. The user is the first and foremost element in this model; the Web space is

what the user interacts with to obtain wanted information between the user and the

Web and there is an interface which has been designed to mediate communication

20

between the two. User–Web interaction is viewed as a communication process

facilitated through an interface.

The model related users’ behaviour to deficiencies in the design of interfaces and

the Web. It designated that Web browser designers needed to understand the

users’ mental models. An effective interface must provide great affordance and

facilitate correct mental model development by presenting appropriate

messages/clues and providing context-sensitive help. Whilst pointing to the

important role of the users’ mental model in the provision of information, Wang et al.

(2000) did not elaborate the users’ cognitive processes in detail.

2.3.2.3 Ford et al. (2001, 2005) Individual User Differences Web IR model Ford and his colleagues made continual efforts to study the impact of individual

characteristics on Web search strategies and search performance (Ford, Miller &

Moss, 2001, 2005). Key characteristic differences were identified in the users’

cognitive styles, study approaches, prior experience, internet perceptions, gender

and age which were linked to users’ internet-based information seeking.

What the model of Ford and his colleagues lacks, however, is an understanding of

why users choose their search strategies in a given way and to what extent

personal attributes affect the Web search process. Nevertheless, this research

provides an approach for studies about individual differences. The technology

acceptance model was later integrated into an interdisciplinary investigation of the

impact of user perceptions of information quality on IR strategies.

2.3.2.4 Knight & Spink (2008) Web IR Model Different to Saracevic’s model, which suggested the reason why a user sought out

information was an important part of discerning the influencing factors during the

21

interaction, Knight and Spink’s (2008) Web IR model contends that the users’ pre-

existing cognitive style is seen as influencing the Web search strategies and is

followed by two types of system interaction, browse-seek and search-seek.

Knight and Spink’s (2008) model hopefully provides a better understanding of the

impact of the users’ pre-existing cognitive style on the users’ Web search strategies,

such as the adoption of certain search engines and the perception of the value of

the search engine’s results to their query. A weakness of the model is its lack of

accounting for the influence of current knowledge and a dynamic cognitive state on

the users’ Web searching performance. In addition, it stays only on a theoretical

hypothesis level.

2.3.2.5 Park (2008) Prioritising and Coordinating Information Behaviour Model

Empirically validated, Park’s (2008) model describes the processes that individuals

engage in to manage multiple information task Web searching under time pressure.

The inner level of the model indicates that self-regulating individuals engage in

information task perceptions, emotional, mental and temporal reactions. The initial

processes at the inner level then send out a signal to the outer level to prioritise

and coordinate multiple information tasks.

Park’s (2008) model illustrated dynamic internal and external processes that users

employ in order to efficiently and effectively deal with multiple information tasks

while interacting with Web search systems. The strength of her model is that it

examines Web searching behaviour from a multitasking point of view. Park

acknowledged that individuals monitor and coordinate their internal (i.e., emotion,

effort, and time) and external (i.e., task performance) activities through continuous

self-feedback. The self-feedback mechanism, however, was not clearly examined

or presented.

22

The examples outlined above indicated that current Web search models

concentrate more on the users’ cognitive efforts involved in a single topic Web

search interaction. Few studies have examined how users’ cognitive efforts are

made during a multiple topics’ Web searching process. Additionally, current Web

technologies and interface design are generally based on the assumption that most

users are engaging in single Web searches on a single topic. Yet an increasing

number of recent studies show that most people engage in multiple search

activities. Multitasking is observed as an important attribute of Web searching

behaviour (Spink, Ozmutlu & Ozmutlu, 2002). A comprehensive multitasking Web

search model may provide implications for the development of adaptive Web

technologies.

Literature about multitasking, including up-to-date research about multitasking in a

Web search context, is reviewed in the following section.

23

2.4 Multitasking

2.4.1 Multitasking Research in Cognitive Science and Psychology

The research on multitasking has a decades-long history in the field of cognitive

sciences and psychology. Cognitive psychologists have provided extensive

research literature on multitasking, concurrent information processing, task

switching (Burgess, 2000; Pashler, 2000) and sequential actions (Carlson & Sohn,

2000).

Multitasking is the ability of humans to handle the demands of multiple tasks

concurrently through task switching or interleaving if necessary (Just, Carpenter,

Keller, Emery, Zajac & Thulborn, 2001; Lee & Taatgen, 2002). Lee and Taatgen

(2002) argued that multitasking behaviours are a product of skill acquisition.

Multiple task situations are faced frequently in daily life. For example, having

snacks when watching a TV program, or using multiple information systems

simultaneously, for example, checking emails while chatting with friends online.

Neuro-cognitive psychologists have investigated the human brain’s activation

mechanism which is associated with multitasking behaviour. They found that when

humans conduct multiple tasks at the same time, the activation volume in the

cortical systems underlying the execution of tasks decreased compared to that in

single task conditions. That is, the cognitive limitation of multiple task performance

causes a brain activation decline. They concluded that the productivity level of

multitasking performance is accordingly reduced (Just, Carpenter, Keller, Emery,

Zajac & Thulborn, 2001). Engineering psychologists found, however, that people

may adopt strategies of time sharing or time swapping to manage their multitasking

situations effectively. Time sharing is for performing multiple tasks simultaneously,

while time swapping is for performing multiple tasks sequentially (Wickens, 1989,

24

1991). Hunt and Joslyn (2000) identified the characteristics of individuals who did

well in multitasking and decision making situations under considerable time

pressured conditions.

Task switching has been recognised as an important element of multitasking.

Monsell (2003) reviewed the notion of task switching in cognitive science research

in which multitasking was considered as switching behaviour from one task to

another in rapid succession. The costs to the individual of switching tasks

compared to non-switch or task repetition trials are a focus of task switching

research in cognitive science. To explain how such multiple tasks and task

switching are performed, experimental psychologists have proposed that cognitive

executive control systems govern processes including the selection, initiation,

execution, and termination of each task (Rubinstein, Meyer, & Evans, 2001). The

cognitive executive control system provides a supervisory function controlling other

perceptual/motor and cognitive processes when switching from one task to another.

2.4.2 Multitasking Studies in Human Information Behaviour

Spink, Cole and Waller (2008) elucidated information behaviour as a multitasking

process. Multitasking information behaviour is emerging as an important

information behaviour research area.

In the realm of human information behaviour, the process of seeking information

concurrently over time in relation to more than one, possibly evolving, set of

information tasks, including shifts in beliefs, cognitive, affective and/or situational

states, is called multitasking information behaviour (Spink, Ozmutlu & Ozmutlu,

2002). Multitasking information behaviour includes searching for information related

to one information task and then switching to search for items on another

information task (Spink, Cole & Waller, 2008).

25

Information scientists have observed the phenomenon of multitasking information

behaviour by employing different methodology in a number of different

environments, such as library use, database searching and the work environment.

Spink (2004) reported results from a case study exploring the multitasking

information behaviour by one information seeker in a public library. The results

showed that people engaged in multitasking information behaviours consisting of

electronic search, physical library search, serendipity browsing and information task

switching in libraries as they seek and search for information on more than one

information task. A process of seventeen information task switches over two library

visits was identified. Spink, Alvarado-Albertorio, Narayanan, Brumfield and Park

(2007) investigated the multitasking information behaviours of ninety-six public

library users through diary questionnaires and found that 63.5% of the 96 library

users sought information on multiple topics and engaged in multitasking behaviours.

Waller (1997) examined how air crew work groups managed multiple tasks under

dynamic and deadline conditions through two field studies. Her model suggested

that work groups engaged in information gathering, task prioritisation and resource

allocation activities in order to perform multiple tasks. The task characteristics

linked to multiple task switching were revealed: (1) the familiarity of the task and its

relative difficulty; (2) the source of the task; (3) the task deadline; (4) the status of

the task in terms of its potential completion; and (5) the sequence of the task in

terms of any interdependence among the tasks being prioritised.

González and Mark (2004) presented results of fieldwork observation of information

workers in three different roles: analysts, software developers and managers. They

introduced the concept of working spheres to explain the inherent way in which

individuals conceptualize and organise their basic units of work. People worked in

an average of ten different working spheres and they spent about twelve minutes in

26

a working sphere before they switched to another. Spink and Park (2005)

conducted a study investigating business consultants’ multitasking information and

non-information task switching and their interplay. They found that information

seeking tasks occurred within multitasking and task switching sequences with non-

information tasks, including computing and communication tasks. The execution of

information seeking tasks often supported or responded to communication or

computing tasks.

Spink and Cole (2005) proposed a model of multitasking and task switching

information behaviour. They argued that information behaviour may involve a

combination of cognitive and physical actions on multiple tasks concurrently or

sequentially, including switching between different information tasks. Information

seekers have to coordinate a number of factors, including their cognitive state, level

of knowledge and understanding of their information problem, into coherent

processes of human information seeking, searching, retrieving and usage

behaviours. Spink, Park & Cole (2006) argued that multitasking is an essential

element of the information behaviour process that must be closely examined,

allowed for and facilitated in the design of IR systems.

2.4.3 Multitasking Studies in Web and Information Retrieval

This section investigates how multitasking research has been conducted in

information retrieval and Web search studies.

Limited studies have shed light on multitasking information behaviour during Web

search and IR sessions. Spink, Ozmutlu and Ozmutlu (2002) originally suggested

that IR searches often include multiple topics during a single search session or

multitasking search. They found that multitask searching is common human

information retrieval behaviour as many IR system users conduct information

27

searching on multiple related or unrelated topics and also switching among the

topics. In addition, IR multitasking search sessions are longer than single topic

sessions, with a mean of 2.1 topic changes per search session. Their study

represents advances in providing a research framework for multitasking during

Web and information retrieval.

Subsequently, multitask searching was examined on the Excite and AlltheWeb.com

Web search engines. Ozmutlu, Ozmutlu and Spink (2003b) found that multitasking

Web searches are noticeable user behaviours, as one tenth of Excite users and

one third of AlltheWeb.com users conducted multitasking searches. Additionally,

multitasking search sessions with a broad variety of search topics are longer than

regular search sessions in terms of queries per session and duration. Koshman,

Spink and Jansen (2006) also reported that 11.1% of search sessions over the

Vivisimo search engine were multitask searches, including a broad variety of

search topics in multitask search sessions.

In 2006, Spink, Park, Jansen and Pedersen (2006) conducted separate studies of

two-query search sessions and three or more query search sessions on the

AltaVista Web search engine. The degree of multitasking search and information

task switching was examined, based on these two sets of search sessions.

Findings included (1) a high degree of multiple topics existed in both two-query

sessions (81%) and three or more query sessions (91%); (2) three or more query

sessions sometimes contained frequent topic changes; and (3) multitasking was

found to be an un-ignored but growing element in Web searching.

Further, Spink, Park and Koshman (2006) investigated multiple information

problems ordering which is engaged in Web search by observing forty study

participants’ Web searching behaviour. They concluded that information task

ordering was affected by the following factors: personal interest, problem

28

knowledge, perceived level of information available on the Web, ease of finding

information, level of importance and seeking information on information problems in

order from general to specific. Personal interest and problem knowledge were the

two major factors affecting multiple information problem prioritisations.

Park (2008) specifically studied human prioritising and coordinating information

behaviour among multiple information tasks in a Web information seeking and

retrieval context. She found that human prioritising behaviour was affected by

multiple factors, such as task attributes, emotions and time. Users’ dynamic Web

search interactions existed among the components of their prioritising and

coordinating information behaviour. Multiple tasks prioritisation and coordination

was operationalised as being composed of the level of each task, cognitive,

affective, and temporal and behaviour dimension. The affective factor was

considered as an important factor affecting task performance. She stated that how

we manage our emotions ultimately yielded successful performance. Her

discussion of coordination information behaviour was more on the task level, such

as coordinating activities of task switching, and tabbed browsing.

2.4.4 Coordination Viewpoint in Multitasking Research

According to Wickens (1989), multitasking research includes both task

characteristics and coordination processes. Multitasking information behaviour is

conceptualized as a binding process that works with human coordination

behaviours to construct an information behaviour process. It provides a framework

for coordinating and integrating the different levels within information behaviour.

Information science researchers such as Spink, Park and Cole (2006) and Spink

and Du (2007) incorporated the concept of coordination into multitasking

information behaviour. They discussed the role of multitasking and coordination as

conceptualizing and binding elements in the integrated information behaviour

29

framework. People coordinate the translation of their information problem(s) by

performing search term selection tasks, tactic and strategy tasks, search engine

interaction tasks, and relevance judgments. Effective interactive IR must be a

successful process of coordinating the switching between related or unrelated tasks.

Multitasking processes, in general, involve a person’s allocation of his/her own

scarce cognitive resources among several tasks and the moderating impact of task

elements, task processes, and task resources on multiple-task performance. The

task coordination research concerns how people coordinate their activities to

perform tasks, in particular, decision-making and problem-solving tasks (Waller,

1997). Iani and Wickens (2004) pointed out that the performance of multiple tasks

was controlled by cognitive executive processes that enable humans to choose and

prioritise tasks, and monitor, interrupt and adjust task performance. Hence, it is

necessary to identify how such cognitive executive control processes establish

priorities among multiple information tasks and allocate resources to them, thus

allowing efficient multiple-task performance.

The following section discusses in detail the research on coordination and its theory,

mechanism of coordination, especially the coordination in information retrieval, and

ends up with recent studies on cognitive coordination within Web searching context.

30

2.5 Cognitive Coordination

2.5.1 Research on Coordination

Coordination Theory

Malone and Crowston (1994, p. 88) adopted the term “coordination theory” to refer

to theories examining how coordination can occur in diverse systems: coordination

in parallel and distributed computer systems, in human systems, and in many

cases, coordination in complex systems including both people and computers

(Clark, 1985, 1992, 1996; Malone & Crowston, 1994).

Coordination theory focuses on understanding the principles underlying how people

collaborate effectively and productively in order to achieve coordination.

Coordination theory provides an understanding of the nature of various kinds of

dependencies among activities and how people can manage to avoid problems,

such as unsatisfied pre-requisites for important tasks (Crowston, 1997).

Coordination in Multiple Disciplines

Researchers viewed coordination as the process of managing dependencies

among activities, or conflicts between goals, tasks, and resources of various agents

(Kling, Kraemer, Allen, Bakros, Gurbaxani & Elliott, 2001; Malone & Crowston,

1994; Rapoport & Fuller, 1998).

Coordination has specific meanings in various disciplines, ranging from

organisation theory to computer science, from economics to linguistics, and from

health to psychology, to name just a few. In the area of management information

systems (MIS), coordination is a central problem in information system

development (ISD) projects. During the execution of project tasks, coordination

31

mechanisms can be classified into two categories: vertical and horizontal.

Horizontal and vertical coordination affect the performance of IS projects (Parolia,

Goodman, Li & Jiang, 2007). Here, coordination mechanisms are more like the

information exchange system between IS team members and stakeholders or

users.

In the area of industry and business, various types of coordination mechanisms

have been used in supply chain coordination, such as quantity discount, credit

option, buy back/return policies, quantity flexibility, and commitment of purchase

quantity. For example, Sarmah, Acharya and Goyal (2007) developed a

coordination mechanism through credit option such that both manufacturers and

buyers could divide the surplus equitably after satisfying their own profit targets.

In the field of health and exercise sciences, motor research on individual

differences views coordination abilities as major determinants of motor

achievements in sports (Busch & Strauss, 2005). A number of studies have also

demonstrated high force coordination of the hand grip force (GF) and load force (LF)

in a variety of manipulation tasks. When manipulating a hand-held object, the

exertion of excessive grip can cause muscle fatigue or crash fragile objects; thus

appropriate adjustments of GF should be implemented simultaneously with, or even

in advance of, changes in LF associated with object manipulation (Freitas, Krishnan

& Jaric, 2007). During the process of studying force coordination in grasping and

manipulation tasks, the underlying neural control mechanisms could be revealed.

In the research area of natural language communication, Clark and his colleagues

have conducted a series of fruitful studies which investigated the role of

coordination during people/agents conversation communications (Clark, 1985,

1992, 1996; Clark & Carlson, 1981; Clark & Marshall, 1981). They found that

linguistic communication was a type of joint action performed by two

32

participants/agents in the communication through certain shared common ground.

Common ground was an important discovery and basis in their theories. Common

ground, referring to the mutual knowledge of conversational participants, formed a

related context against which the coordination took place. Two participants could

coordinate and communicate successfully only if they monitored what was mutually

known about a situation and used the knowledge effectively. The common ground

was inferred from three kinds of information: linguistic evidence, perceptual

evidence, and community membership. Based on the findings, they suggested that

coordination could extend its application to other forms of communication, including

human–computer communication.

Coordination in IR Interaction

In information retrieval, the concept of coordination has been applied to design

“blackboard architecture” in distributed expert IR systems (Belkin, Brooks & Daniels,

1987). Belkin and his colleagues developed the blackboard architecture to support

coordination among subsystems, including plan, agenda, user model, request

model, and I/O requests.

Recently, Ma (2008) established an IR coordination model in which she examined

IR as a communication process between users and IR systems based on Clark’s

theory framework of language communication coordination analysis (Clark, 1985).

In her study, coordination was taken as an intrinsic and fundamental process to IR

interaction. Users and the IR system need to coordinate such intricate aspects as

are routinely coordinated in natural language use, as descried in Clark’s studies. IR

is a process of coordination involving the dynamic identification and construction of

common ground between users and IR systems. She suggested that common

ground consisted of a good knowledge of the matching mechanism, of requests,

search strategies and tactics, and of the search objectives. Such common ground

33

was dynamically constructed, during IR interaction, from three kinds of information,

linguistic evidence, perceptual evidence, and community membership, through

exploiting the search logs. Her IR coordination model provides a useful tool with

which to analyse IR interactions and improve system design. However, her IR

model of coordination research emphasizes the communication point of view

between users and IR systems. The deep level of coordination mechanism was not

touched.

2.5.2 Cognitive Coordination Mechanism in Psychology

The following psychological studies of cognitive coordination mechanism may

provide us with some clues for exploring coordination mechanisms in information

retrieval and Web searches.

Glass (1996) adopted the psychological refractory period (PRP) procedure to

examine dual-task performance and used executive-process/interactive-control

(EPIC) architecture to yield a detailed, computational explanation of PRP data. The

PRP procedure and the EPIC architecture can be used to fruitfully study aging and

dual-task processing. The architecture of EPIC was employed to analyse the

performance of both single and multiple tasks in detail and to separate differences

in ability (such as information processing speed) from differences in task-

coordinating strategies. Research demonstrates that the Strategic Response

Deferment (SRD) model has been successful in modelling the results of PRP

experiments with response-selection overlap (Meyer, Kieras, Lauber, Schumacher,

Glass, Zurbriggen, Gmeindl & Apfelblat, 1995).

Blume and Gneezy (2000) presented an experimental investigation of optimal

learning in repeated coordination games. The games were two-player pure

coordination games that were repeated twice. These researchers thought of the

34

focus on optimal attainable strategies (OAS) as an expression of players' strategic

uncertainty.

Phillips and Silverstein (2003) emphasized the important role of cognitive

coordination in schizophrenia, and proposed that a failure of such coordination may

occur during schizophrenia as a result of reduced ion-flow through NMDA

glutamate receptors. Authors argued that cognitive coordination, NMDA-activity,

and psychosis were related. Reducing the activity of NMDA-receptor channels

could produce the cognitive disorganisation. Cognitive disorganisation and thought

disorder are more related than any other symptoms to each of the major aspects of

cognitive coordination. All the impairments with schizophrenia in perception, pre-

attentive sensory gating, selective attention, working memory, and long-term

memory, as well as other cognitive impairments more obviously interpreted in terms

of context, involved cognitive coordination. Cognitive coordination was referred to

as a special kind of modulation.

Kushleyeva, Salvucci and Lee (2005) proposed an adaptation of the ACT-R

(Adaptive Control of Thought-Rational) cognitive architecture that incorporated a

notion of elapsed time for the current goal and used time to determine when to

switch away from the current task. They demonstrated how the ACT-R model could

account for various aspects of human subjects’ switching behaviour when applying

this mechanism to a dynamic, time-critical dual search task.

Salvucci, Taatgen and Kushleyeva (2006) performed a study of discrete driving in

which participants used a keyboard to steer a vehicle while entering navigation

information as a secondary task. They described a cognitive model of multitasking

and interleaving for extended concurrent tasks, specifically focusing on learning

when to switch from one task to another and how task interleaving may evolve in

changing task conditions.

35

Still about the driving task, Brumby, Salvucci and Howes (2007) conducted an

experiment that investigated dual-task trade-offs while driving and dialing. Their

results demonstrated that under dual-task conditions people could adjust their

strategies depending on varying task objectives, namely, people’s strategy shifts

occur in order to adapt to certain objectives.

Yeung, Nystrom, Aronson and Cohen (2006) proposed a cognitive control hierarchy

comprising regions responsible for maintaining task-specific information about rules

or goals, and regions involved in the coordination of these goals. They used

functional magnetic resonance imaging (fMRI) to examine the nature of interactions

between brain regions and the role of the prefrontal cortex (PFC) in cognitive

control as subjects switched between simple face and word categorisation tasks.

Findings which support the theory that between-task competition is a critical

determinant of behaviour include that activity in brain regions selective for the

currently irrelevant task predicted the behavioural cost associated with switching

tasks.

2.5.3 Psychological Concepts in Support of Coordinating Multiple Tasks

The concepts listed below are of importance in the support of multiple tasks coordinating. • Cognitive Strategy Cognitive strategies are based on past experiences and the knowledge bases

developed from these experiences as one's understanding of the relationships

between the environment, tasks, and oneself grows. A cognitive strategy is a

mental plan that guides how one uses sensory input, detects and corrects errors

and directs motor actions (Sangster, Beninger, Polatajko & Mandich, 2005).

36

• Cognitive Style Cognitive style, intentionally underlying search strategies, expertise and other

cognitive factors in interactive information retrieval, is an individual's characteristic

and self-consistent modes of functioning in cognitive activities, such as problem

perception and problem solving. An individual with a certain cognitive style

tendency may find certain tasks easier than others (Wang, Hawk & Tenopir, 2000).

Cognitive styles are investigated during Web search sessions (Ford, 2000; Ford,

Miller & Moss, 2002, 2003; Ford, Wood & Walsh, 1994).

• Conscious Control and Subconscious Control People's ability to overlap the information processing for multiple tasks is more

extensive than previously thought, and the amount of overlapped information

processing is under strategic control (Glass, 1996). The large body of psychological

research on attention develops a two-system approximation to the control of

behaviour. One system, conscious control, has limited resources; the other system,

subconscious control, seems to develop specialized procedures for tasks that are

relatively independent of one another. As a result, subconsciously performed tasks

can be treated as resource unlimited, so that several can be done simultaneously.

Only well-learned, routine tasks can be done subconsciously (Miyata & Norman,

1986).

• Planning Planning helps discover the situations in which some tasks must wait until

prerequisite conditions are satisfied while other tasks should be postponed when

more important work should be done instead. In a word, the planning stage helps

arrange the order of tasks execution appropriately. One of the major difficulties in

planning comes about because of the limitations of human processing resources.

37

The limits on conscious resources and working memory capacity mean that in-

depth planning is often not possible without external aid (Miyata & Norman, 1986).

Planning is largely seen as a deliberate and conscious process. In the context of

multitasking, one can clearly imagine people reasoning about the structure of the

multiple tasks in which they must engage, and explicitly devising a plan to

interleave the tasks (Lee & Taatgen, 2002).

• Interruptions Interruptions can be both external and internal. External interruptions result from

events in the environment. Internal interruptions come from our own thought

processes, normally, new ideas that draw attention away from the current activity

(Miyata & Norman, 1986).

• Immediate Action and Deferred Action Tasks are run in immediate mode and deferred mode respectively. Immediate

mode allows task processing to run to completion without delays; deferred mode

lets task processing proceed without delay until the task lockout point is reached.

Decrements in dual-task performance are attributed to competition for peripheral

processors, for example, hands or eyes, and to strategic control of task processing

to comply with instructions or other priorities dictated by the task environment

(Glass, 1996).

2.5.4 Elements within Coordinating Web Searching Process

Web searching is a type of complex activity involving users’ cognitive and

behavioural efforts. Motivation, intention, cognition, emotion, and actions of people

need to be coordinated when conducting Web search interactions.

38

Miyata and Norman’s (1986) research showed that humans have different levels of

cognitive coordination. Spink, Park, Jansen and Pedersen (2006) pointed out that

people must perform a series of tasks, such as search term selection task, tactic

and strategy tasks, search engine interaction task, relevance judgments task, in

order to recognise, cognitively articulate and make sense of an information problem

or a gap in their knowledge. Interactive IR occurs as a series of cognitively

coordinated task actions.

To achieve successful interactive IR, people cognitively coordinate a number of

elements, including their cognitive state, their level of domain knowledge, and their

understanding of their information problem, into a coherent information retrieval

process (Spink & Du, 2007).

Park (2008), in her dissertation study, operationalised coordinating information

behaviour as being composed of the level of each dimension of a task: cognitive,

affective, temporal, and behavioural. The coordinating activities in Web information

seeking and retrieving context included task switching, tabbed browsing, strategic

search planning, and information evaluation.

During Web searching, humans cognitively coordinate relevant elements in order to

achieve a coherent Web searching process and thus a comfortable and successful

searching experience. Users' information problem may or may not stem from his or

her level of satisfaction or dissatisfaction with previously acquired information. We

suppose that the user has a certain cognitive strategy under conscious control at

the very beginning of Web searching. Such a cognitive strategy is a mental plan

that guides how to select search term input, evaluate system outcomes, modify or

change queries, and switch between information problems until satisfied with

search results.

39

Adopting the above mentioned concepts, cited from psychology and cognitive

science, we propose that the elements that need to be coordinated during Web

searching interaction could be divided into two categories, internal (cognitive)

factors and external (environmental) factors, which have a mutual effect on each

other. Internal factors include identification of the information problem, level of

domain knowledge, understanding of the information problem, planning of

information problems, level of experience on Web search engine, evaluation of

system output, self-learning and modification, cognitive strategy, internal

interruptions, immediate action and deferred action, conscious control and

subconscious control. Alternatively, visional serendipity, external interruptions, as

well as time constraints belong to external factors. For instance, visional serendipity

may cause users to suspend ongoing topic searching and start a new Web

searching task. These assumptions are to be validated in this study.

Effective Web searching interaction must be a successful process of coordinating

internal and external elements. Furthermore, how internal coordination factors and

external factors interplay with each other in order to make a coherent process of

human Web searching behaviour is an important issue that deserves special

attention. As people do multiple tasks and coordinate among these tasks, cognitive

shifts in their mind must occur during such processes. The following section gives a

discussion of the studies related to cognitive shifts.

40

2.6 Cognitive Shifts

Studies on cognitive shifts are found in a variety of fields of psychology, cognitive

science, human factors, and human–computer interaction research (Allport, Styles

& Hsieh, 1994; Ashby, Isen & Turken, 1999; Diaper & Stanton, 2003; Iani &

Wickens, 2004). Cognitive shifts are a type of psychological phenomenon

experienced by individuals. Simon (1981) argued that a cognitive shift, the unique

ability of humans, belongs to higher human cognitive processes. Cognitive shift is a

human ability to handle the demands of complex and often multiple tasks resulting

from changes due to external forces (Spink & Dee, 2007).

2.6.1 Cognitive Shifts in Interactive IR

The phenomenon of cognitive shift is an important element of human interaction

with IR systems. Spink and Dee’s (2007) study concludes that if an IR system user

does not experience some type of shift in information problem processing,

represented by shifts in cognitive, problem and knowledge states, then the IR

system interaction has not been effective.

Various forms of cognitive shifts that take place during information seeking and

retrieving processes have been identified and modelled. The information problem

stage (Spink, 2002; Spink, Wilson, Ford, Foster & Ellis, 2002b) and information

seeking stage (Kuhlthau, 1993) were identified as two types of cognitive shifts at

information seeking level. Researchers examined shifts in search strategies (Xie,

2000), information problem (Robins, 2000), search stages (Santon, 2003),

uncertainty (Ingwersen, 1996; Wilson, Ford, Ellis, Foster & Spink, 2002), and query

reformulation (Rieh & Xie, 2006) as types of cognitive shifts during the interactive

IR process.

41

Saracevic’s (1996) stratified model of IR interaction provided a theoretical

explication of the possible levels and shifts between those levels within an IR

interaction. Xie (2000) used an information-seeking strategy to describe users’

shifts from one state to another in their attempts to achieve their information-

seeking goals. Results showed that there exist two dimensions of information-

seeking strategies: methods and resources; and three levels of shifts: (1) current

search goal shifts, (2) interactive intention shifts, and (3) information-seeking

strategy shifts.

Robins’ (2000) interaction shifts were defined as any change in focus of the

conversation between a user and a search intermediary with respect to the users’

information problem. Shifting of focus was denoted by a change in some topical

aspect of the conversation, or by some shift to a non-topical aspect of the

information problem. Ten categories of focus were classified for the analysis of

information problem focus shifts: DOC (focus on documents availability and format,

etc), EVAL (focus on judgments of system output), I (indiscernible passage), SNSR

(focus on the discussion of social issues not related to the search), ST (focus on

the discussion related to the experimental itself), STRAT (concerned with the

search strategies), SYS (focus on explanations, preparations, or problems with the

IR system), TECH (focus on the discussion of technical issues), TOPIC (focus on

the subject area guiding the search), and USER (focus on users’ background

including searching experience). The shifts within those foci occurred during the

discourse between a user and a search intermediary within IR interactions. Spink

(2002) assessed and operationalised the users’ information problem shift by

measuring the change in an information-seeker’s information problem stage before

and after their interaction with an IR system.

42

Santon (2003) investigated users’ moves/shifts between different cognitive search

stages in an iterative information retrieval process. Search stages included search

intension, prior experience, database selection, query formulation, results reviewing,

search progress reviewing, physical delivery of results, technical related problems,

and so on. Although some of the stages are represented as physical actions, they

reflect the users’ evolving cognitive and strategic states.

2.6.2 Variables Affecting Cognitive Shifts

Quite a few information seeking and retrieving studies have included individual

differences as variables affecting shifts on users’ cognition. User-based variables

include cognitive styles, internet perceptions, study approaches, levels of prior

experience, personality traits, and demographic variables (Ford, Miller & Moss,

2001; Heinström, 2003; Palmquist & Kim, 2000; Wang, Hawk & Tenopir, 2000).

In the 1980s, Dervin and Nilan (1986) argued that information was a subjective

construct. An individual's need for information shaped what that person saw in the

environment and what information that person would construct from the

environment, rather than by the objective nature of the information being sought.

Miyata and Norman (1986) stated that there existed two styles of human

information processing, task-driven processing and interrupt-driven processing.

Individual differences played a role in deciding whether a person was in a state of

task or had interrupted driven processing. They found that some people were more

easily controlled by task-driven structures, while others tended to be distractible by

extraneous events or thoughts.

Recent studies have shown that different individuals seek and process information

using very different strategies which may be more or less effective for different

people in different contexts. Individuals may, to some extent, be characterised by

43

cognitive styles which are a consistent tendency to adopt one or other type of

information processing strategy (Ford, Miller & Moss, 2001).

Wang, Hawk and Tenopir (2000) investigated how users search for factual

information on the web with a focus on individual differences. Their results

suggested that (1) differences in cognitive styles affected the search process; (2)

field-dependent individuals probably had greater difficulty in the web environment

and became confused more easily than field-independent individuals. Palmquist

and Kim’s (2000) study shows that there was a relationship between cognitive style

and experience on Web searching. Field-dependent novice searchers took longer

and traversed more nodes in locating relevant information than field-independent

novices.

Ford, Miller and Moss (2001) investigated the role of individual differences in

Internet searching, modelling the relationships between individual difference factors

and retrieval effectiveness of internet-based information seeking. Results showed

that the interaction between self-efficacy, gender, and cognitive styles may be an

important factor in models of human–computer interaction. Ford, Wilson, Foster,

Ellis and Spink’s (2002) study reported on the relationship between cognitive styles

and problem solving and associated information-seeking. They concluded that field-

independent researchers were more analytical and active than their field-dependent

counterparts. Furthermore, holists engaged more in exploratory and serendipitous

behaviour, and were more idiosyncratic in their communication than serialists.

Heinström (2003) studied individual differences in information behaviour, with a

particular focus on how and why personality traits influenced information strategies.

It was claimed that each individual had a unique way of seeking information. Spink

and Dee’s (2007) research showed that (1) all study participants reported some

level of cognitive shifts in their information problem, information seeking, and

44

personal knowledge due to their search interaction; and (2) different study

participants reported different levels of shifts on various criteria.

Different people deal with information problems using different information

processing styles and cognitive styles. Cognitive styles play a role in creating

knowledge with an influence on information behaviour (Ford, Wilson, Foster, Ellis &

Spink, 2002). The process of cognitive shifts varies depending on individuals’

cognitive styles. Different people inherently have different cognitive styles that

decide different patterns of their cognitive shifts and the undergoing of different

cognitive states (Ingwersen, 1996). However, in addition to the inbuilt cognitive

styles, cognitive shifts may be influenced by users’ active Web search interactions.

We plan in the present study to explore the impact of dynamic searching process

on users’ cognitive shifts.

45

2.7 Relevant Dissertations and Theses

This section describes a bibliometrics analysis of the dissertations and theses

associated with the current research project. Dissertations and theses have been

retrieved from ProQuest Dissertations and Theses Database via QUT library link

http://proquest.umi.com.ezp01.library.qut.edu.au/pqdweb. The database includes

Doctoral dissertations from 1861 onwards, and Masters theses from 1962 to the

present. As of the publication year 2008, sixty-six dissertations and theses have

been retrieved by using “document title” as the search item.

Table 2-1 provides a summary of retrieval results, including the search terms and

number of each corresponding dissertations and theses.

Table 2-1. Search terms and amount Search Terms Number % Web search* 33 50 multitasking 28 42 shift*and cognition 3 5 cognitive shift* 1 1.5 coordinat* and cognition 1 1.5 Total 66 100 The number of Web search/searching and multitasking research papers accounts

for 92% of the total retrieved papers. Few dissertations or theses talk about

research on the topics of cognitive coordination and cognitive shifts in the field of

interactive IR.

46

Table 2-2 shows the number of Doctoral dissertations and Masters theses,

respectively. It demonstrates that more doctoral dissertations than Masters theses

are related to this research.

Table 2-2. Statistics on dissertations and theses Categories Number % Doctoral dissertation 41 62 Masters thesis 25 38 Total 66 100 Table 2-3 provides a summary of the published dissertations and theses annually,

from 1970 to 2008.

Table 2-3. The annual number of published dissertation and thesis related to this study Year Number 1970 1 1973 1 1979 1 1984 1 1985 1 1988 2 1990 1 1992 2 1993 1 1995 1 1996 2 1997 2 1998 3 1999 2 2000 2 2001 2 2002 3 2003 2 2004 3 2005 5 2006 15 2007 3 2008 10

47

Figure 2-1 further indicates the trend of the yearly number of published studies.

Figure 2-1. Year-Number (from 1970 to 2008)

1 1 1 1 12

12

1 12 2

32 2 2

32

3

5

15

3

10

0

2

4

6

8

10

12

14

16

1970 73 79 84 85 88 90 92 93 95 96 97 98 99 2000 01 02 03 04 05 06 07 08

Year

Num

ber

Number

In general, the number of dissertations and theses on Web searching, multitasking,

cognitive coordination, and cognitive shifts is not huge. From Figure 2-1, however,

we can see the increasing trends year by year. Before 1990, there was only one

published PhD or Masters Thesis per year; after 1990, the mean number reached

two or three papers each year. In 2006, the number rose to fifteen. The number of

dissertations for year 2007 dropped to three, but went up to ten in 2008. This

suggests that the topic presently studied is becoming an important one and is

attracting more and more interest from researchers.

48

Topical Analysis

Topics of the retrieved dissertations and theses are analysed according to the

relevance level to the current study. Amongst the retrieved degree documents,

most of them refer to:

• The Entities of Web Search Engines and Web Search Strategies These include, for example, implicit-feedback based ranking methodology for Web

search engines (Adya, 2005), collaborative autonomous interface agent for

personalised Web search (Al Nazer, 2006), effects of query ambiguity and results

sorting method (Aurelio, 2002), psychological fidelity of Web search engines

(Dudziak, 2000), efficient query processing in large Web search engines (Long,

2006), using the genetic algorithm to optimize Web search (Nguyen, 2006), web

search strategies in technical environments (Sran, 1999), and contextual Web

search based on semantic relationships (Zhang, 2006a).

• Multitasking in Multiple Fields Studies have discussed multitasking interfering effects of multitasking on muscle

activity (Au, 2005), cognitive multitasking in situated medical reasoning (Farand,

1996), media multitasking among American youth (Foehr, 2006), optimizing the

multitasking of workers in just-in-time systems (Horng, 1996), real-time multitasking

kernel for the IBM personal computer (Ju, 1988), multitasking control environment

for flexible manufacturing system (Judt, 1988), social multitasking and nonverbal

decoding (Lieberman, 1998), effects of multitasking on quality inspection in

manufacturing systems (Pesante-Santana, 1997), real-time, multitasking approach

to automatic monitoring and control of industrial processes (Pezas, 1992),

multitasking on social networks (Swamy, 2005), multitasking in work groups (Waller,

49

1995), and effects of monochronicity and polychronicity on multitasking strategy

and performance (Zhang, 2006b).

• Cognitive Shifts and Coordination in Psychology and Language Learning Keeling (1973) conducted a study on hypnosis and adaptive regression using

Wild's cognitive shift measure; Cheng (1998) examined the effects of cognition–

emotion processes on shifts in conflict management strategies; Hochstadt (2004)

adopted eye-tracking during sentence-picture matching to link deficits in language

comprehension and cognition in Parkinson's disease; Fuse (2006) investigated

flexible coordination of spatial cognition and language.

Even so, a few dissertations have shared similarities, to some extent, with this

study which tries to identify rules, principles, relationships and coordination

mechanisms of cognitive and multitasking information behaviour during the Web

searching process.

• The model in Waller's (1995) dissertation examined how work groups

managed multiple tasks under dynamic and deadline conditions and

suggested that work groups engage in information gathering, task

prioritisation, and resource allocation activities in order to perform multiple

tasks. It provided the implications for theories of self-regulation and

performance feedback effects while conducting multitasking behaviours, but

not under Web searching context.

• Santon (2003) shed light on users’ information seeking behaviour in

mediated information retrieval interaction. She explored the search stages

users pursued as they proceeded to solve their information problems. The

shifting patterns of the transitions between search stages were observed

and identified, leading to the new model of information seeking behaviour

50

that represented a non-linear process of multiple re-iterative cycles

describing multiple search stages, aspects and dimensions of information

seeking interaction. But there remains limited focus on how users

coordinate these shifts between multiple search stages.

• Drawing on working memory theory, Tao (2006) examined cognitive

processing during a Web search, whether the number of relevant search

results returned during a Web search would increase cognitive load, and

whether the increase would augment the processing of peripheral and

irrelevant advertisements. However, this dissertation considered few

multitasking characteristics during Web searching.

• Katzeff (1989) attempted to view problems of HCI in relation to theories in

cognitive psychology. The thesis investigated adults' reasoning when

learning to use database systems, with the aim of identifying cognitive steps

involved in query writing and of examining the significance of the system

image for users' mental models.

• Jong (1991) developed a HCI framework to describe the seven stages of

HCI activities, classified as knowledge-based, rule-based, or skill-based

behaviours and the underlying cognitive processes.

However, the above dissertations and theses are mainly concerned with cognitive

processing or steps as users search on the Web or conduct HCI, paying limited

attention to integrated characteristics of Web searches involving the aspects of

multitasking, cognitive coordination, and cognitive shifts.

51

2.8 Theoretical Model

Based on the analysis of previous studies reviewed above, a preliminary

conceptual model has emerged to present a proposed relationship between

multitasking, cognitive coordination and cognitive shifts during information problem

oriented Web searches (Figure 2-2).

Figure 2-2. A conceptual model of multitasking, cognitive coordination and cognitive shifts during Web searching

52

Figure 2-2 outlines an integrated cognitive process of Web searching. It shows how

a user interacts with Web search systems, such as a Web search engine,

incorporating multitasking, cognitive coordination, and cognitive shifts as primary

behavioural and cognitive mechanisms. This is a basic model which will be

validated by empirical data derived from this study. In this model, Web searching is

understood in the context of the information problems of the user, the inner or

cognitive processes of the user, and the environmental factors relating to the

information. These factors have an iterative effect on the users’ way of responding

to the information problem.

A dynamic Web searching interaction exists between users, information problems

and information environment (such as a Web search engine) that produces an

iterative affect on users’ search strategies, processes and outcomes, as shown in

Figure 2-3.

Figure 2-3. Dynamic Web search interactions

People generate information problems and conduct Web searching by coordinating

the translation of their information problems through physical and mental activities

53

including query formulation, search strategies selection, search execution, and

system outcomes evaluation (relevance judgments).

The evaluation process may lead to the identification of a brand new information

problem or the redefinition of an information problem, or simply the reformulation of

a new query as a result of cognitive shifts occurring. During Web searching, people

multitask, switching back and forth between different information problems because

of the factors shown in Figure 2-3. Establishing and sustaining effective Web

searching requires humans to coherently coordinate and multitask their information

problems, cognitive activities, and interactive search tasks.

54

2.9 Chapter Summary

Web searching can be portrayed as a high-level and complex multitasking and

coordinating activity. At a basic level, this may involve executing multiple

perceptual-motor actions concurrently, such as reviewing search results and

clicking an associated link by using the mouse. At a more complex level, this may

involve interleaving the steps of many cognitive and interactive tasks, such as

judging relevance while reformulating the query and starting a new-run of searching.

Web searching is an important element of human–computer interaction which

includes multitasking processes and the allocation of cognitive resources among

several tasks, as well as shifts in cognitive, problem and knowledge states (Spink &

Dee, 2007; Spink & Du, 2007). Multitasking is now considered to be an important

feature of Web searching behaviour. However, a greater understanding is needed

of how cognitive executive control processes establish priorities among multiple

information tasks and allocate resources to them, thus allowing efficient multiple-

task performance.

In addition to multitasking, both cognitive shifts and cognitive coordination are also

important but under-explored aspects of Web search behaviour. Cognitive shifting

is the ability of humans to use a higher mental process of cognitive processes that

is triggered by the brain's response and change due to some external force.

Different levels of cognitive shifts that take place during Web searching have been

identified and modelled (Spink & Dee, 2007).

Cognitive coordination is an important cognitive process. Internal and external

factors have an effect on cognitive coordination mechanisms, allowing humans to

manage dependences among information tasks and the resources available.

55

Human cognitive information coordination facilitates the moving through an

information behaviour process.

However, little research has been done on what and how elements are cognitively

coordinated in order to achieve a coherent Web search process. Limited attention

has been paid to modelling the relationship between multitasking, cognitive

coordination and cognitive shifts during a Web search. Cognitive coordination, in

conjunction with multitasking and cognitive shifting, may form a theoretical

framework for understanding how Web searching behaviour is constructed. An

integrated model of Web searching as a multitasking, cognitive coordination and

cognitive shifts process is needed.

56

Chapter 3 Research Design

3.1 Introduction

Research design refers to an overall approach with regard to data collection and

data analysis within the study. The study was designed to explore multidimensional

cognitive aspects of the interaction between users and Web search systems. The

research goals were outlined with the intention of modelling the inter-relationship

between users’ multitasking, cognitive coordination and cognitive shifts during Web

search interactions through a detailed analysis of empirical data, and drawing

conclusions which are applicable to system design.

This chapter describes the methodology of data collection and data analysis.

Importantly, it discusses the recruitment and selection of study participants, and the

various instruments used to collect the data of multitasking, cognitive coordination

and cognitive shifts. This is followed by a description of micro-analytic methods

through a grounded theory approach.

In order to pretest and refine the data collection and analysis techniques, a pilot

study was conducted with two student participants during the first months of the

study. The results of the pilot study were reported in a published conference paper

(Du & Spink, 2009).

57

3.2 Data Collection

The data were derived from the Web search interactions between the study

participants, their real information problems, and the Web search systems,

collected by the author. The following description of the data collection methods

refers to the recruitment and selection of study participants, the research setting

which includes the Web searching tools, and the instruments and procedures used

for collecting the data.

3.2.1 Study Participants

A total of 42 postgraduate students from the Queensland University of Technology

(QUT) in Brisbane, Australia participated in this study. They were recruited via e-

mail groups under the guidelines of the QUT Research Ethics Committee.

The bulk of the emails were sent to all QUT postgraduate students. In the email,

the study was introduced and a clear mention was made of the purpose of the

research and what was expected of the study participants. In exchange for their

time and efforts involved, an honorarium was presented to each study participant in

the form of a QUT bookshop voucher worth AUD $25. Research shows that paying

respondents in research surveys increases response rates and the value they feel

(Thompson, 1996).

If a contacted person was willing to participate in the study, he or she emailed back

to the given email address. They were chosen according to the principle of “first

come, first serve”, along with an attempt to select study participants from different

programs in order to allow the study participants to be as heterogeneous as

possible in their Web searching interests and activities. The study participants were

then scheduled to perform the searches at their convenience. Finally, forty-two

58

postgraduate student participants were involved in this study. The relatively small

number of study participants was recruited because many studies relating to Web

IR are conducted using similar numbers and users (Spink, 2002; Spink, Park &

Koshman, 2006; Spink & Dee, 2007). Further demographic characteristics of the

study participants are presented in Chapter 4 Results.

Ahead of participation, each study participant completed a consent form (see

Appendix A) regarding their participation, confidentiality, and rights and protections

under the guidelines established by the QUT Research Ethics Committee.

3.2.2 Research Setting

The study was conducted in a laboratory setting. The venue was Professor

Amanda Spink's office, which was located on Level 3 of the Faculty of Science and

Technology Building at the Queensland University of Technology, Brisbane,

Australia. The study was run individually and took around 1.5 to 2 hours per study

participant. The period of data collection lasted nearly two months from September

to November in 2008.

One laptop was provided at the office for use by study participants. The laptop was

a TOSHIBA 1.40GHz Intel Pentium with 512 MB of RAM and a 60 GB hard disk

running Microsoft Windows XP. The laptop was equipped with Camtasia Studio

software for recording and capturing search logs and verbal data. The software

records audio and video streams by capturing any activities, such as keystrokes

and screen actions on the Windows screen. In particular, this software was used to

record those URLs visited by the study participants and the continuous screen

shots (actions), while the think-aloud utterances were made on a disk. The logs

with timelines and verbal reports were recorded throughout the entire process.

59

A headset with microphone was provided as necessary peripheral hardware for

imputing study participants’ audio data into the laptop. A pen and a piece of A4 size

paper were also given to the study participants so that they could write down

anything that came to mind.

An MP3 player was used to record the interviews between the researcher and

study participants at the end of their participations. The MP3 player was Samsung

YP-U3 2GB model, with 1-inch OLED display and built-in USB plug for uploading

and downloading documents and materials to and from the laptop.

3.2.3 Information Problem in Web Searching Context

Problem Orientation For a long time, information seeking and information retrieving models were often

based on the concept of information need (Saracevic, Kantor, Chamis & Trivison,

1988). Slowly, however, modelling has changed to include problem orientation and

viewing the information problem as new central to information seeking and

retrieving context. Representatively, Belkin and Vickery (1985) argued that

information needs implied that there was specific information for which a user had a

need, and for which he/she was seeking. Using the term information need would

then restrict any given situation in which the information sought was known to exist

within a well-defined context.

The problem-oriented view is greatly affected by cognitive science (Saracevic,

Kantor, Chamis & Trivison, 1988). In cognitive science, problem solving research

holds that a problem is said to exist when it is at a given state, and it is desired to

be at another state, but there is no clear way to get from the given state to another

state (Meyer, 1977). An individual is faced with a problematic situation, in everyday

life, or the world of work tasks. Problem–solving is the underlying motivation for

60

information seeking and retrieval. An information problem is treated as dynamic,

not static, and hence may change over time, probably due to learning and cognition

in context during IR interaction (Ingwersen & Jarvelin, 2005).

An information problem implies a broader range of factors affecting a search

strategy, such as query formulation, relevance judgments, and information use and

seeking behaviour (Robins, 2000). As described, information problems are of

importance to interactive IR because they provide a guide by which we proceed in

a search, and also provide a means by which we measure the success of our

searches. For example, in interactive Web searching situations, the statement of

users’ information problem is provided by the query entered into a Web search

engine. The more specific term information problem is used to describe the task a

user needs to carry out during Web searching. Information problems can evolve

and change over time (Spink, Park & Koshman, 2006).

Information Problem in This Study

This study focuses on how users search for information on the Web. The

information problems were user-initiated rather than imposed by the researcher, in

order to better simulate Web searching reality. Subsequent data collection and data

analysis relied on the context of information problems provided by the study

participants.

In the recruitment email, study participants were asked to prepare three information

problems related to their individual research or work or everyday life in advance

and bring them to the study session. There was no treatment or control on how the

study participants interacted with the Web to find solutions to their information

problems. The requirement to investigate three information problems allowed the

researcher to analyse each user’s Web searching behaviour. This can be

61

characterised as multitasking, a method which was also adopted in the Spink, Park

and Koshman (2006) research of multitasking Web searches.

The way to analyse the study participants’ information problems is usually to

capture the context of the problems by asking them to describe their problems; this

method has normally been employed in previous studies (Spink, Albertorio,

Narayanan, Brumfield & Park, 2007; Spink, Wilson, Ford, Foster & Ellis, 2002a). In

our study, study participants were required to write down the descriptions of their

three information problems on the pre-Web search questionnaire. Detailed analysis

of the information problems is presented in the Results section in Chapter 4.

3.2.4 Web Browser and Web Search System

Windows Internet Explorer (IE) was set up as the default Web browser installed

beforehand in the laptop used in this study. The default page in the Web browser

was set for the University's (QUT) homepage.

In order to make the situation as naturalistic as possible, for example, getting the

study participants instead of the researcher to select searching problems, and there

were no restrictions on the selection of Web search systems. Study participants

were free to using search systems to start their searching and using their selected

Web search systems during the process. Every effort was made to preserve the

real situations of interaction between users and Web search systems. Each study

participant answered the following question in the post-Web search questionnaire

to indicate the used Web search systems:

• Which Web search systems were employed in your current searches?

The analysis of the employed Web search systems is described in Chapter 4

Results.

62

3.2.5 Time Constraints

Time has an important impact on the information seeking and retrieving process

that influences users’ expectations, predictions, choices they make, and outcomes

they achieve (Kuhlthau, 2004). The process functions within the time frame of

information problems to be solved. Generally, people alter their decision strategies

due to time constraints or time pressures. Time constraints give people a sense of

control over the pace and process of problem solving. On the Web, they attempt to

trade off between the cost which is measured in time and the value of the

information pursued (Slone, 2007).

In this study, time constraints were assumed as important factors reflecting the

study participants’ search strategies. Each study participant was given one hour in

total for their Web searching activity and they determined the allocation of the one-

hour searching period between multiple information problems. Different users may

have different reactions to time pressure. At the end of the study, study participants

were asked about their feelings on time pressure due to the rate or pace at which

information problems solved through post-Web search questionnaire. A five-point

Likert scale was used to measure the users’ feeling:

1—--2—--3—--4—--5

Low High time pressure time pressure

1 - Low time pressure

5 - High time pressure

If the study participants felt time pressure was low or they often had spare time,

they should tick the number ‘1’; if the study participants felt time pressure was high

or they almost never had spare time, they should tick the number ‘5’.

63

Likert Scale A Likert scale is a type of psychometric response scale often used in

questionnaires. The Likert scale is the sum of responses on several Likert items. A

Likert item is simply a statement in which the respondent is asked to evaluate

according to any kind of subjective or objective criteria. Generally the level of

agreement or disagreement is measured. A five-point Likert scale (five-level Likert

items included) is often used. For example, Saracevic, Kantor, Chamis and Trivison

(1988) utilised a five-point Likert scale to obtain an indication of the information

seeking context from users. Another case is Kuhlthau, Spink and Cool’s (1992)

study, which adopted a five-point Likert scale to investigate the nature of the users’

problem definition and work stage in the search process. A five-point Likert scale

was also used in Ford, Miller and Moss’ (2005) research to measure users’

cognitive complexity.

In this study, a series of five-point Likert scales were utilised to obtain study

participants’ data on many aspects, including their changes on Information problem

perception/ understanding, personal knowledge, judgment of contribution to the

resolution of the information problem, and time pressure.

3.2.6 Instruments

The combination of multiple data collection instruments facilitates the capture of a

comprehensive and intensive picture of complex phenomena. Even if each data

collection method delivers only partial evidence on the phenomenon, several

methods together could cover multiple aspects. In addition, multiple methods allow

for cross-checking the results of each, thus increasing the reliability and validity of

the data: giving rise to the term triangulation (Ingwersen & Jarvelin, 2005). People’s

64

Web search interaction is a sort of complicated activity involved with a lot of

cognitive and behaviour efforts. The method of interview or survey could provide

only a partial picture of the interaction because users may not be aware of or able

to recollect what they did during the searching process. Triangulation through

multiple methods revealed a more complete picture of users’ Web searching

behaviour.

Saracevic, Mokros and Su (1990) used videotaping, search logging and

observation to collect data on the nature of online searching interaction between

users and intermediaries. Bystrom (1999) used multiple data collection methods

including theme interviews, observation and diaries in the study on civil servants’

information seeking behaviour. In addition to a pre-search questionnaire, Wang,

Hawk & Tenopir (2000) employed a process-tracing technique for recording

individual user's processes and behaviours, specifically the transactions with

timestamps and verbal reports as they interacted with the Web when searching for

information. These studies provided good examples of using a variety of methods

to collect data in order to explore users’ information behaviour.

In this study, a combination of multiple data collection techniques, including pre-

and post-search questionnaires, think-aloud protocols, observations, Web search

logs, and post-search interviews, were used to gather data. Multiple methods

provided in-depth and comprehensive insights into the users’ Web searching

behaviour.

Pre- and Post-Web Search Questionnaires A questionnaire requires less effort but can be an effective technique to gather

information directly from study participants.

65

Basic information of study participants was gathered through a pre-search

questionnaire, including variables of age, gender, affiliation, student status, and

Web using experiences. The following questions partly based on the questions

raised in Spink, Park & Koshman’s (2006) study were designed to indicate each

study participant’s Web using experiences:

• How long have you been using the Web to look for information (one year–

five years; six years–ten years; eleven years and over)?

• Which Web browser is used most frequently for information browsing (Ex.

Internet Explorer, Netscape, Mozilla, Maxthon, MyIE, and FireFox)?

• Which Web search engines do you use most frequently for information?

Both a post-Web search questionnaire and a pre-Web search questionnaire were

used to capture the transition of information users’ state before and after the search

(Spink & Dee, 2007). This allowed the measurement of cognitive shifts of study

participants that resulted from their Web search.

The pre-and post-Web search questionnaires are included in Appendix B and

Appendix C, respectively.

Think-aloud Protocols Think-aloud or verbalization of a person's thoughts while undertaking a cognitive

activity has been used for years as an assessment tool by psychologists and

reading researchers to investigate the mental processes applied to various kinds of

thinking, problem-solving, and reading comprehension tasks (Smith, 2006).

The advantages of think-aloud and recording lie in the real-time attribute that

everything is recorded in real time, and that the often complicated cognitive tasks,

66

which take place over long periods of time, can be analysed. In addition, thinking

aloud could produce valid verbal accounts of cognitive processes (Ericsson &

Simon, 1993). The disadvantage of the think-aloud method originates from its

obtrusiveness, which may lead to a validity problem—there is uncertainty about the

degree to which the protocols reflect actual thoughts or intentionality of actions, and

bring forward reliable data (Ingwersen, 1992). In a comprehensive summary,

Ericsson and Simon (1993) argued that the think-aloud method is an accurate and

representative measurement of cognitive processes, particularly when subjects are

reporting memory traces that are already in verbal form before they begin the

process of verbalizing about them. This condition is presumably met in Web

searching study because individuals will verbalise while solving information

problems.

Yang (1997) used think-aloud and protocol analysis in the study on information

seeking and retrieving in hypertext. The verbal data were recorded and analysed to

find the goal-oriented and information seeking patterns in the subjects’ problem

solving. Spink, Park, and Koshman (2006) asked subjects to think aloud as they

searched and were encouraged to express the reasons for their Web search

actions.

Think-aloud protocols provided rich data in the study, including both operational

steps and cognitive moves. Study participants were required to verbalise everything

related to the searches, including thoughts, motivations/reasons, and actions, as

they were searching on the Web. The verbal stream was recorded through

Camtasia Studio software. These think-aloud data uncovered important cognitive

processes surrounding the Web searching actions and decision making processes.

Cognitive elements identified through study participants’ verbal reports included

information problem identification, clarification, and query formulation. Verbal

67

reports also reflected how the study participants moved between cognitive states,

for example, from relevance judgment to query formulation.

The recorded think-aloud utterances offered an excellent vehicle to the researcher

to access the study participant’s thoughts and an opportunity to explain and confirm

the Web search logs which logged their actions of interacting with Web search

systems.

Observation

Observation allows real time data collection. The value of this method is that it

permits researchers to study people in their native environment in order to

understand "things" from their perspective. In the process of observation, the

researcher uses all of his/her senses to gather information about the phenomena

under study (Adler & Adler, 1994). While it may affect the search process because

of obtrusiveness, it may also provide a rich data set for analysis. Non-participant

observation requires a longer period of accommodation by those being observed to

return to natural behaviour, but it also allows the observer to concentrate on the

observation process (Krathwohl, 1997). Spink (2004) used observation as one of

the data collection methods to explore from a distance the multitasking information

behaviour by an information seeker in a public library during library visits. On-site

observation was also used in Spink, Park and Cole’s (2006) research to examine

how a business consultant conducted multitasking work-task behaviours.

Non-participant observation was used in this study to familiarise the researcher with

the users’ Web searching processes by generating notes to supplement thinking

aloud audio data. The observations were carried out as unobtrusively as possible.

Everything was done as it would have proceeded without the observation. The

researcher interrupted study participants only when reminding them about how

68

much time was left or when asking the study participants to speak loudly as doing

the searching. The observation notes and comments were manually recorded by

the researcher at the same time. The observation notes and comments helped the

researcher to construct certain interview questions for each study participant

afterwards.

Web Search Logs During Web searching, search logs can be collected unobtrusively and

economically. Web search engine logs analysis has been employed successfully to

reveal information about users’ search strategies, the products of users’ minds

(Hider, 2007). Many search engine log studies have been performed since the late

1990s (Jansen & Pooch, 2001). For instance, Spink, Park, Jansen and Pedersen

(2006) used transaction logs of AltaVista Web search engine to analyse two-query

and three-or-more-query search sessions respectively, in order to examine the

characteristics of users' task switching and multitasking information behaviour

during Web searching sessions. Nevertheless, the data are poor if this method is

used alone, since they lack most traces of a searcher’s intentions and thoughts

during Web interaction (Ingwersen & Jarvelin, 2005).

It is understandable that study participants could verbalise only a subset of the

thoughts occurring during the interactions because some thoughts were difficult to

verbalise (Wang, Hawk & Tenopir, 2000). The method of Web search logs captured

all of the screens and the moves that study participants made, including those

which study participants were not aware of or were not able to verbalise. In this

study, characteristics of actual searching behaviours of study participants and the

duration of search sessions were recorded by the search logs. The Web search

logs which recorded each study participant’s Web searching interactions were

captured and saved by Camtasia Studio software. Analysis of the Web search logs

69

along with the verbal reports provided insight into the study participants’ Web

searching behaviour and thoughts.

Post-search Interview

Interviewing requires more effort than questionnaires but provides the possibility for

more thorough analysis of Web search behaviour. It may effectively augment and

validate data collected through previous methods, such as observation, which may

be incomplete and/or given incorrect interpretations. Therefore, a clarifying

interview used as triangulation of data may be very helpful (Ingwersen & Jarvelin,

2005). Dervin (1992) and Schamber (2000) developed a micro-moment time-line

interview technique for the sense-making approach, which involved asking the

respondent in detail what happened in a problematic situation step by step. Spink,

Park and Cole (2006) interviewed a business consultant about his multitasking

behaviours. The goal of the open-ended interview technique was to discuss in

detail the consultant’s information and non-information tasks and processes.

In the study, a post-Web search interview was conducted with each study

participant, immediately after their Web search interactions, in order to allow further

elaboration of their actions and underlying thoughts. The interview with study

participants made the acquisition of retrospective data possible. We could seek

evidence or further explanation of what was found in the think-aloud utterances.

Interview questions were formulated on the semi-structured interview outline and

the individual-based observation notes. The researcher modified questions or

added additional ones as deemed appropriate. The interviews were recorded using

an MP3 player at the same time. The semi-structured interview questions are

included in Appendix D.

70

3.2.7 Procedures

Even though the data were collected under laboratory conditions, the data

collection still involved real users (study participants shared the characteristics of

Web search system users) with real problems (study participants searched for their

personal information problems) conducting real Web searching activities (study

participants solved their information problems by interacting with their preferred

Web search systems).

Each study participant was asked to prepare their own three information problems

(topics) related to their research or work or everyday life and to bring them to the

study session. They conducted Web searching activities based on the own three

topics. No restrictions were imposed on the Web search systems to be employed

and the information problems to be searched for.

Ahead of participation, each study participant completed a consent form under the

guidelines provided by the QUT Research Ethics Committee. After signing a

consent form, each study participant followed the same data collection procedures,

descried as follows:

1) The study participant was given instructions specifying what were offered and

required regarding the research process;

2) Before Web searching, the study participant filled out a pre-Web search

questionnaire;

3) Wearing a headset with microphone, the study participant opened the Web

search system he or she preferred to use and conducted Web searching;

4) At the same time as searching on the Web, the study participant performed a

think-aloud activity related to his or her Web search interactions. A sheet of

71

instructions for thinking aloud was distributed in advance. The Web search logs

and think-aloud data were recorded simultaneously by Camtasia Studio

software;

5) Non-participant observation was employed by the researcher as the study

participant was conducting Web search interaction. The researcher kept written

observation notes and observer comments while observing;

6) The recorded Web search logs and think-aloud data were saved into the laptop

hard disk drive after the study participant’s Web search sessions was over;

7) The study participant then completed a post-Web search questionnaire;

8) At the end of the Web search, an interview was conducted by the researcher

with the study participant, and the talks were recorded using an MP3 player;

9) In conclusion, the researcher collected all the documents and expressed

appreciation to the study participant.

72

3.3 Data Analysis

3.3.1 Overview of Methods

Qualitative research can provide in-depth descriptions and interpretations of people

and their settings (Gay & Airasian, 2003). A qualitative analysis approach, the main

methodology in this study, can meet the requirements of deep understanding of

users’ Web searching behaviours. Qualitative methods can be used to uncover and

understand what lies behind any phenomenon about which little is yet known, or

which is difficult to convey with quantitative methods (Strauss & Corbin, 1990).

Adopting a qualitative research approach also depends on the nature of the

research questions in this exploratory study.

A grounded theory approach (Henwood & Pidgeon, 2006; Strauss & Corbin, 1990),

content analysis (Schamber, 2000), and verbal protocol analysis (Ericsson & Simon,

1993) were three main qualitative analysis methods employed in this study. These

analysis techniques have been widely used in prior user-Web/IR system interaction

studies (Spink & Dee, 2007; Spink, Park & Koshman, 2006; Spink, Wilson, Ford,

Foster & Ellis, 2002a; Wang, Hawk & Tenopir, 2000).

Grounded Theory Approach The qualitative methods used in this study were based on grounded theory (Strauss

& Corbin, 1990). Developed by Glasser and Strauss (1967), grounded theory

approach is often referred to the methodology adopted in qualitative studies. It is

characterised by developing theories that are close to the data through interplay

between theory and data. In this way, an emerging theory is grounded in a real-life

situation. The development of theory is an inductive process where newly emergent

data are continuously integrated with previously recorded data until associative

73

patterns emerge (Glaser & Strauss, 1967). The inductive, theory-discovery

methodology “allows the researcher to develop a theoretical account of the general

features of a topic while simultaneously grounding the account in empirical

observations or data” (Martin & Turner, 1986, p. 141). An emerging pattern, concept

or proposition is discovered and integrated into the emergent theory. Therefore, it

derives theory from data rather than verifying existing theory (Ingwersen & Jarvelin,

2005).

Grounded theory is especially applicable to an exploratory study, such as the study

described in this thesis. The data were coded and analysed manually using the

grounded theory approach.

Content Analysis Content analysis is appropriate in studies based on a grounded theory approach

(Ingwersen & Jarvelin, 2005). Content analysis is defined as a research method

which uses a set of procedures to make valid inferences from texts (Weber, 1990).

The key point of applying the content analysis method is to classify many words of

texts into much fewer categories. In order to code for the content, the basic units of

the texts need to be classified, including each word, word sense, sentence, and

theme (Weber, 1990).

Qualitative research typically generates very large amounts of text for analysis

(Ellis, 1993). Web searching behaviour study was no exception. In this study, the

content analysis of pre- and post-questionnaires, observation notes, search logs,

and interview responses developed preliminary taxonomies of multitasking Web

search behaviours and cognitive shifting reported by study participants.

74

Verbal Protocol Analysis Verbal protocol analysis is a method mostly used by cognitive psychologists and

human–computer interaction researchers in order to understand the thoughts of

users at the same time as they are performing the assigned tasks (Ericsson &

Simon, 1993). Protocol analysis is another name for content analysis. It is a major

analysis method for user-oriented and cognitive IR research, particularly when the

data are transcribed think-aloud accounts of cognitive processes (Ingwersen &

Jarvelin, 2005).

In this study, thinking aloud verbal protocols provided rich data for cognitive

coordination analysis. Users’ cognitive coordination behaviours during Web search

interaction were captured from the thinking aloud utterances. The researcher

transcribed the verbal protocols, focusing on the different levels of cognitive

coordination that each study participant experienced.

3.3.2 Identification of Variables

The data included pre- and post-Web search questionnaires, transcribed think-

aloud utterances, observation notes, Web search logs, and transcribed post-Web

search interview notes. Amongst multiple data sources, the transcribed think-aloud

data and the Web search logs constituted two principle sources of analysis in this

study. The analysis of think-aloud data and Web search logs was regarded as

utterance-search segments analysis.

As mentioned, the research aimed to explore the relationship between multitasking,

cognitive coordination and cognitive shifts during Web searching. The study

examined the nature of users’ multitasking, cognitive coordination and cognitive

shifts during Web searching through the identification of variables including

75

multitasking sessions, cognitive coordination occurrences and types of cognitive

shifts in the data.

Multitasking Sessions Multitasking sessions were observed through the analysis of Web search logs

recorded by the screen-capture software Camtasia Studio. A multitasking session

was demonstrated as behaviours, including the ordering and switching between

multiple information problems (tasks), and the generation of evolving information

problems including serendipity browsing activities. Study participants’ responses on

the post-Web search questionnaire, associated with the following questions, were

considered as self-explanations of their recorded logs of multitasking behaviours.

• How and why did you order your multiple information problems?

• How and why did you switch information problems (tasks) from one to

another?

• Were there any evolving information problems generated during your Web

searching? What were they?

Cognitive Coordination Occurrences

Cognitive coordination occurrences were identified through the analysis of think-

aloud transcripts and Web search logs, that is, verbal protocol analysis of the

utterance-search segments. Utterance-search segments were recorded by

Camtasia Studio software. Each study participant was required to verbalise their

thoughts, actions and underlying reasons at the same time as conducting the Web

search. Each study participant’s utterance related to the Web searching activities

was transcribed onto the section of their search logs which recorded those activities.

76

The transcripts and search logs were segmented into cognitive coordination

occurrences. A cognitive coordination occurrence was demonstrated as iterative

sequence behaviour including:

• Information problems identification;

• Search terms (re)selection and queries (re)formulation;

• System output in response to the search strategies;

• Study participants’ utterances including relevance and magnitude judgment

feedback, and sense-making process relating to the systems output;

• Study participants’ actions relating to the relevant items retrieved.

Cognitive coordination occurrences were then coded and classified into the three

different levels that are reported in the Results in Chapter 4.

Types of Cognitive Shifts Types of cognitive shifts were identified mainly through the analysis of pre- and

post-Web search questionnaires, particularly the data largely reported on Likert

scales. A cognitive shift was indicated as a change of study participants’ perception

in terms of information problem understanding, information problem stage,

information seeking stage, personal knowledge, and contribution to the information

problem resolution prior to and after Web searching. Content analysis and

descriptive statistics revealed study participants’ shifts in cognition before and after

the Web search.

Study participants’ cognitive state shifts during the searching interactions were

captured by the Web search logs. The analysis of the utterance-search segments

77

provided the details of study participants’ cognitive states and the changes between

the states.

A unified coding scheme was created based on the investigation of the integrated

data from the transcribed verbal reports and search logs. The coding scheme was

used to analyse a possible underlying relationship between the behaviours of

multitasking, cognitive coordination and cognitive shifts during Web searching.

3.3.3 Classification of Variables

After identifying the variables of multitasking, cognitive coordination, and cognitive

shifting during the course of Web search interaction, the next step toward

categorising these variables was to develop a unified coding scheme.

Coding Scheme

The aim of this study is to understand the interplay between multitasking, cognitive

coordination, and cognitive shifting during Web search interactions. The principal

data sources were the transcribed think-aloud data and the Web search logs,

namely, utterance-search transcripts. Therefore, it is necessary to derive cognitive

Web searching behaviours level coding from the utterance-search level coding.

After identifying and analysing each of forty-two utterance-search segments, the

coding scheme, consisting of 13 categories, was derived for the analysis of Web

searching behaviours involving multitasking, cognitive coordination, and cognitive

shifting. Table 3-1 presents the coding scheme for analysing the utterance-search

transcripts. Detailed interpretations of each code along with the instances from the

study participants are reported in Chapter 4 Results.

78

Table 3-1. Coding scheme Code Description CRF Content relevance feedback occurs when a user makes relevance

judgments regarding the retrieved results displayed. EVA Focus on judgments regarding the relevance, magnitude, etc. of system

output (the list of entries). GS Global strategy is used to balance resource (e.g. time) allocation between

the multiple information tasks. IT Information task identification. MF Magnitude feedback occurs when a user makes a magnitude judgment

on the size of the number of system returned results. OVE Focus on the overall search outcome and the change between

information tasks. PSS Problem specific strategy is the tactics on each searching information

problem solving. SLR Self-learning and regulating (sense-making process) occurs based on the

past searching experience and current domain knowledge. STR Concerned with the search strategies adoption and adjustment, e.g., term

selection, query (re)formulation, and Web search system selection. TCF Tactical review feedback occurs when a user makes a strategy-related

judgment to adjust the search strategy. TOP Focus on the specific subject area/information problem guiding the

search. TRF Term relevance feedback occurs when a user identifies a term (terms)

within the retrieved results subsequently used to modify the search strategy.

VIE Focus on viewing and examining a specific opened search result link against the information problem searching aim, such as availability, format, useful or not.

79

3.3.4 Open Coding

The coding paradigm represents a high-level data analytical device (Ellis, 1993).

Open coding is described as the process of breaking down, examining, comparing,

conceptualizing, and categorising data (Strauss & Corbin, 1990). Open coding was

the key part of grounded theory analysis which pertained specifically to the naming

and categorising of phenomena through a close examination of data. Without open

coding, the rest of the analysis that followed could not have taken place.

For the purpose of utterance-search segments analysis in this study, the recorded

think-aloud audio stream and Web search activities were transcribed. Open coding

was employed to code the data. Coding the transcribed utterance-search data was

carried out according to the coding scheme. Utterance-search transcripts were

coded in the order of occurrence. Each variable was coded into an appropriate and

corresponding category described in the coding scheme.

80

An Example of Open Coding The following segment is an example of open coding (see Figure 3-1). The

example demonstrates how each code in the coding scheme was used.

Figure 3-1. An example of an open coding outcome

The occurrence of utterance-search segments was illustrated in the form of a

flowchart which combined the study participants’ think-aloud data and associated

Web search logs.

81

Web Searching Process as Flowchart All the forty-two study participants’ cognitive Web searching processes were coded

according to the coding scheme and were fully illustrated as flowcharts. Each

flowchart was mapping to a study participant’s Web searching process,

incorporated the behaviours of multitasking, cognitive coordination and cognitive

shifts. Finally, forty-two flowcharts were presented. Appendix E collects the

flowchart examples of Study Participants 2, 6, 11 and 30.

The flowcharts of forty-two study participants are available from the link:

http://picasaweb.google.com.au/dujiarainy/DuPhD_ThesisWeb_Searching_Process

_as_FlowchartStudy_Participant_14203#

.

82

3.4 Verification of Methodology

Validity, reliability and generalisability are the notions used to evaluate the quality of

quantitative research which is in the positivist research paradigm (Creswell, 1994).

The criteria for evaluating qualitative studies which are in the

naturalistic/interpretive research paradigm should differ from the way for

quantitative studies (Lincoln & Guba, 1985). As qualitative researchers and writers,

Lincoln and Guba (1985) developed certain languages to distance themselves from

the positivist paradigms. They proposed four criteria for evaluating qualitative

studies: credibility, transferability, dependability and confirmability. Since the

present study is qualitative research on investigations of users' behaviours,

credibility, transferability, and dependability were thus adopted to assess the

methodology used and the accuracy of findings concluded.

3.4.1 Credibility

A set of activities were recommended to strengthen the credibility of a qualitative

study, including persistent observation, triangulation, negative case analysis,

checking interpretations against raw data, peer debriefing, and member checking

(Lincoln & Guba, 1985). Creswell (1994, 2003) also suggested triangulation and

member-checking as two frequently used and easy strategies that help improve the

credibility. The employment of triangulation refers to collecting data from multiple

sources. In the present study, data were collected through multiple instruments

including questionnaires, think-aloud protocols, observations and interviews. The

triangulation of data enhances the validity of the data collected and subsequently

corroborated the findings. Member-checking refers to seeking feedback from

participants. In this study, the researcher's observation data and comments were

presented back to the study participants during post-search interview and

83

discussed with them whether the data and comments were accurate. The member-

checking activity largely ensures the truth value of the data and the researcher's

interpretations. In addition, all the study participants were under the same setting,

with the same equipments of the laptop, the headset with microphone, the MP3

player, and Camtasia Studio software, and with the same data collection

procedures and standardized instruments such as the pre- and post- Web search

questionnaires. All of these efforts were made in order to improve the credibility of

the findings reported in this study.

3.4.2 Transferability

The purpose of qualitative research is not to generalise findings, but to form a

unique interpretation of events (Creswell, 1994). The notion of transferability in

qualitative research paradigm requires rich and thick descriptions to convey the

findings so that other researchers are able to make judgments about the findings'

transferability to different settings or contexts (Zhang & Wildemuth, 2009). In the

current study, descriptions and interpretations of data collection settings and

procedures, data analysis strategies including coding processes, and results were

reported in detail so that other researchers can adopt the methods and compare

the results in different contexts.

3.4.3 Dependability

Dependability refers to checking the consistency of the study processes.

Dependability addresses the issue of consistency in the application of the

methodology, or the degree to which an instrument measures the same way each

time it is used under the identical condition with the same subjects. The major

technique for establishing dependability is through external audits of the research

processes (Lincoln & Guba, 1985).

84

The question of dependability pertains to the consistency with which the coding

scheme developed in this study can be applied by different coders. In this study,

the researcher transcribed thinking aloud data, Web search logs, and observations

and interviews data. The researcher then did all coding tasks. To ensure the

reliability of the coding scheme and solve the dependability issue of the consistency

of content analysis, a second coder was invited to code two search transcripts

(forty-two search transcripts in total). The samples of transcripts were selected

randomly. The second coder was experienced in qualitative research methods. She

was given the data transcripts along with the coding scheme and the instruction

used in the study. The consistency came from an inter-coder agreement. The inter-

coder agreement was calculated using Holsti’s (1969) reliability formula:

21

2..NN

MRC+

=

In the formula, M is the number of coding decisions agreed upon by the

researcher and the coder, and and refer to the number of coding decisions

made by the researcher and the coder, respectively. The inter-coder agreement

reached the level of 0.81, which indicates an acceptable level for drawing

conclusions in qualitative research (Krippendorff, 1980).

1N 2N

85

3.5 Chapter Summary

This chapter detailed the overall research design and explained the methodologies

involved in the data collection and analysis. The researcher justified how multiple

data collection techniques were integrated in the capture of users’ complex and

dynamic Web searching behaviours. The application of qualitative analysis

approaches based on grounded theory successfully obtained insights into users’

cognitive activities during the Web search interactions. The next chapter of this

dissertation makes use of all of the data sources, measures, and data analysis

methods discussed above in the investigation of users’ behaviours of multitasking,

cognitive coordination, and cognitive shifts in the Web searching context.

86

Chapter 4 Results

4.1 Introduction

This chapter reports the results from the Web searching study in which forty-two

study participants were involved. The results were obtained through the qualitative

analysis of pre-and post-Web search questionnaires, think-aloud transcripts, Web

search logs, and observation and interview notes.

The results have addressed the research questions which are:

1) How do users conduct their Web searches on multiple information problems?

2) What types of cognitive shifts occur during Web searching?

3) What levels of cognitive coordination occur during Web searching?

4) How do multitasking, cognitive shifts and cognitive coordination interplay during

Web searching?

This chapter contains the following five sections: Section 4.2 provides the study

participants’ demographic characteristics, Section 4.3 reports their Web using

experience, Section 4.4 presents the results of multitasking behaviour, Section 4.5

presents the results of cognitive shifts, and Section 4.6 provides the results of

cognitive coordination behaviour.

87

4.2 Demographic Data

Forty-two postgraduate students from Queensland University of Technology (QUT)

in Brisbane, Australia participated in the study. Table 4-1 summarises the

demographic information of the study participants.

88

Table 4-1. Study participant profiles

Study Partic-ipant

Gender Age Academic Status Faculty Web Use (yrs)

1 M 30–39 Doctoral Science and Technology 6–10

2 M 20–29 Doctoral Built Environment and Engineering 6–10

3 M 20–29 Doctoral Built Environment and Engineering 6–10

4 M 20–29 Doctoral Science and Technology >11 5 M 20–29 Masters Business 6–10 6 F 30–39 Doctoral Science and Technology 6–10

7 F 20–29 Doctoral Built Environment and Engineering 1–5

8 M 20–29 Doctoral Built Environment and Engineering 1–5

9 M 20–29 Masters Science and Technology 1–5 10 M 50–59 Doctoral Education 6–10 11 F 20–29 Masters Business >11 12 F 20–29 Masters Education 6–10

13 M 30–39 Doctoral Built Environment and Engineering 6–10

14 M 20–29 Doctoral Business 6–10 15 M 20–29 Graduate Diploma Education >11 16 M 30–39 Masters Science and Technology >11 17 M 50–59 Doctoral Creative Industries >11

18 M 20–29 Doctoral Built Environment and Engineering 6–10

19 F 20–29 Doctoral Creative Industries 6–10

20 M 20–29 Doctoral Built Environment and Engineering 6–10

21 M 20–29 Doctoral Science and Technology 6–10 22 M 30–39 Doctoral Science and Technology 6–10 23 M 20–29 Graduate Certificate Business >11 24 M 30–39 MBA Business >11 25 F 20–29 Masters Education 6–10 26 F 20–29 Doctoral Health 6–10 27 F 30–39 Doctoral Creative Industries 6–10 28 F 30–39 Masters Education >11 29 M 20–29 Masters Business 1–5 30 M 20–29 Masters Science and Technology 6–10 31 F 20–29 Doctoral Science and Technology 6–10 32 F 20–29 Doctoral Business 6–10 33 M 20–29 Doctoral Science and Technology 6–10 34 M 20–29 Masters Business 6–10 35 F 20–29 Doctoral Business 6–10 36 F 30–39 Doctoral Creative Industries 6–10 37 F 40–49 Doctoral Health 1–5 38 F 40–49 Doctoral Science and Technology >11 39 F 20–29 MBA Business >11 40 M 30–39 Doctoral Science and Technology >11 41 F 20–29 Doctoral Health 6–10 42 F 30–39 Doctoral Health 6–10

89

Study participants included twenty-four (57%) male students and eighteen (43%)

female students, as shown in Table 4-2.

Table 4-2. Number of study participants in each gender category

Study Participants Gender Number % Male (M) 24 57 Female (F) 18 43 Total 42 100 Table 4-3 demonstrates age distribution of the study participants.

Table 4-3. Number of study participants in each age category

Study Participants Age Number % 20–29 27 64 30–39 11 26 40–49 2 5 50–59 2 5 Total 42 100 Most study participants were in their 20s (64%) and 30s (26%). Two study

participants are in their 40s and two in their 50s.

Table 4-4 shows the number of study participants in each academic category.

Table 4-4. Number of study participants in each academic status category

Study Participants Academic Status Number (mode) % Doctoral 28 (full-time) 67 Masters 10 (8 full-time+2 part-time) 24(19+5) MBA 2 (full-time) 5 Graduate Diploma 1 (full-time) 2 Graduate Certificate 1 (part-time) 2 Total 42 100 The study participants consisted of 28 PhD students, 10 Masters students, 2 MBA

students, 1 Graduate Diploma student, and 1 Graduate Certificate student. Most of

them (93%) were full-time students.

90

Table 4-5 shows the study participants’ different academic discipline background.

Table 4-5. Number of study participants in each faculty category

Study Participants Faculty Number % Science and Technology 12 28 Business 10 24 Built Environment and Engineering 7 16 Education 5 12 Creative Industries 4 10 Health 4 10 Total 42 100

• The 28% of study participants were from the Faculty of Science and

Technology, 24% from the Faculty of Business, and 16% from the Faculty of

Built Environment and Engineering.

• The remaining 32% study participants were composed of the postgraduate

students in Education, Creative Industries, and Health.

91

4.3 Web Using Experience

Tables 4-6 through 4-9 provide information about the study participants’ Web using

experience, including years of using the Web for information, frequently used Web

browsers, and the Web search systems which were employed during the current

Web searches.

Table 4-6 shows the years of Web use by study participants.

Table 4-6. Years of Web use by study participants

Study Participants Years of Web Use Number % One year–five years 5 12 Six years–ten years 26 62 Eleven years and over 11 26 Total 42 100 Most study participants have been using the Web for information for over six years,

representing 88% of the total.

Table 4-7 provides a summary of the number of study participants who frequently

have employed certain browsers for Web searching.

Table 4-7. Number of study participants: frequently used Web browsers

Study Participants Frequently Used Web Browsers Number % IE only 22 53 Firefox only 6 15 IE & Firefox 6 15 Maxthon only 2 5 IE & Maxthon 1 2 IE & Opera 1 2 Safari only 1 2 Chrome & Firefox 1 2 Firefox, Safari & iPhone Browser 1 2 Navigator, IE & Firefox 1 2 Total 42 100

92

• More than half of the study participants preferred Internet Explorer (IE) as

their frequently used Web browser.

• Another 15% study participants chose both IE and Firefox. Firefox was the

second-most popular browser in current use, after IE.

• Other frequently used Web browsers included Maxthon, Opera, Safari, and

Chrome.

Table 4-8 shows the number of Web search systems adopted by forty-two study

participants during the current one-hour Web searching.

Table 4-8. Number of employed Web search systems Number of Employed Web Search System

Number of Study Participants

%

Four 11 26 Two 9 21 One 7 17 Six 5 12 Three 4 10 Five 2 5 Seven 2 5 Eight 1 2 Ten 1 2 Total 42 100 The number of employed Web search systems ranged from one to ten. At the least,

one Web search system was employed; at most, ten Web search systems were

adopted during a study participant’s Web searches. Most of the study participants

(74%) chose one to four Web search systems.

Table 4-9 provides further details on the Web search systems employed by the

study participants.

93

Table 4-9. The employed Web search systems during the Web searches

Employed Web Search System No. Name and Combo

Google only 1 Google Scholar only Google & Google Scholar Google & QUT library database Google & Website with searching feature Google & Baidu

2

Google Scholar & QUT library database Google, Google Scholar & Website with searching feature Google & two QUT library databases Google, QUT library database & Website with searching feature

3

Google & two Websites with searching feature Google & three Websites with searching feature Google, Google Books, Google Scholar & QUT library database Google, Google Groups, QUT library database & Website with searching feature Google, Google Maps, Google Scholar & Website with searching feature Google, Google Maps, QUT library database & Website with searching feature Google, Google Maps & two Websites with searching feature Google, Google Scholar & two QUT library databases Google, Google Scholar & two Websites with searching feature Google, Google Scholar Yahoo, & QUT library database Google, QUT library database & two Websites with searching feature

4

Google, Yahoo & two Websites with searching feature Google, Cuil & three Websites with searching feature 5 Google, Live Search, Cuil & two Websites with searching feature Google, Baidu, & four Websites with searching feature Google, CiteSeer, Yahoo, two QUT library databases & Website with searching feature Google, Google Answers, Yahoo Answers & three Websites with searching feature Google, Google Scholar, Live Search, Yahoo & two Websites with searching feature

6

Google, Baidu, Baidu Baike, Live Search, QUT library database & Website with searching feature Google, Google Translate & five Websites with searching feature 7 Google, & six Websites with searching feature

8 Google, Baidu, four QUT library databases & two Websites with searching feature 10 Google, Live Search, six Websites with searching feature & two QUT library

databases

94

1) The Web search systems used by study participants fall into three categories:

Web search engine, professional database (via QUT Library), and Website with

searching feature. The number of employed Web search systems during each

study participant search ranged from one to ten.

2) The Web search engines used included Google, Google Scholar, Google Books,

Google Groups, Google Maps, Google Answers, Google Translate, Yahoo,

Yahoo Answers, Cuil, Live Search, CiteSeer, Baidu (Chinese Web search

engine), and Baidu Baike.

3) Google (for searching web pages) and its search products (providing services

for searching academic papers, images, maps, etc.) were the indispensable

choice for the study participants when searching for information on the Web,

even though Google was not necessarily the first choice.

4) In terms of the reasons for using Google rather than other Web search engines,

study participants’ responses included “comprehensive”, “accurate”, “neat

interface”, and “easy for use”. For example, typically,

The returned results from other Web search engines were covered by

Google. (Study Participant 1)

Cleaner and simpler search interface. (Study Participant 4)

A typical search engine with strong searching functions; habit for years.

(Study Participants 5, 7, 8 and 10)

5) The professional databases used (via QUT Library) included ScienceDirect,

Engineering Village, EBSCO, ProQuest, Willey InterScience, IEEE Xplore, ACM

Digital Library, and Web of Science.

95

6) In terms of research-type information, some study participants preferred

professional databases to Google Scholar for their “high quality”. The following

examples supported this choice:

The databases provide high quality research papers which were more

professional than those from Google Scholar. (Study Participant 2)

Professional databases provided journal articles and conference papers

which could be used as references. Google Scholar was only viewed as a

second option when searching for academic/research papers. (Study

Participants 3 and 7)

7) In contrast, other study participants believed that Google Scholar provided

reliable research resources. For example:

Google Scholar provided research papers’ “full text” link to the professional

databases. (Study Participants 4 and 6)

I could find authentic and scientific information from Google Scholar. (Study

Participant 10)

8) With regards to the study participants’ specific information problems, certain

Websites with a searching feature were preferred. These Websites included

bank Websites, airways Websites, government portals, job Websites, news

Websites, and on-line shops. For example,

Study Participant 4 employed IMDb, the internet movie website, to search

for his favourite movie.

Study Participant 40 employed www. realestate.com.au, the biggest real

estate Website in Australia, to look for a three-bedroom house.

9) Under this situation, Google was sometimes also employed but only acted as

navigator for the purpose of getting a Website URL. For example,

96

Study Participant 3 input the search query “eBay” in Google in order to find

the URL of eBay Website, followed by searching on eBay for a second-

hand bicycle.

Study Participant 11 went to Google first “because I don’t know how many

agencies in Brisbane and which of them are better”. She got a list of travel

agencies from Google, and then went straight to travel agency Websites to

search for the flight information.

97

4.4 Multitasking Behaviour during Web Searching

A multitasking episode was identified as descriptions of the variables, including the

ordering and switching between multiple information problems searching tasks, the

generation of evolving information problems including serendipity browsing

activities, and multiple search sessions. Multitasking data were identified through

an analysis of study participants’ post-search questionnaires and Web search logs.

Each study participant was required to answer the following three questions on the

post-search questionnaire:

1) how and why did you order your multiple information problems?

2) how and why did you switch your information problem searching tasks from one

to another?

3) was there any evolving information problem generated during your Web

searching? What were they?

Web search logs were examined for the evidence of the study participants’ multiple

search sessions as well as their answers to the above multitasking related

questions.

The following sections present multitasking results, including multiple information

problems search ordering, evolving information problems developing, information

problem searching task switching, and multiple Web search sessions within a

particular information problem searching.

98

99

4.4.1 Multiple Information Problems (IP)

Each study participant was asked to write down three personal information

problems on the pre-Web search questionnaire. There were no restrictions on the

problem topic. These information problems were called original information

problems (OIP) as subsequently the evolving information problems (EIP) were

probably generated during the successive Web searching process.

Table 4-10 indicates the original information problems that underpinned study

participants’ Web searching, and the topic area and status of each information

problem prior to the Web searching.

Table 4-10. Study participants’ information problems

Note: Topic Area—Research (RS), Assignment (AS), Jobs & Careers (JC), Travel & Airfares (TA), News (NE), Food & Entertaining (FE), Technology (TE), Finance (FI), Lifestyle (LI), Music & Movie (MM), Shopping (SH), Sports (SP), Answers (AN), Games (GA), Real estate/Rentals (RE)

Problem Status—D (A new problem area, precisely defined), F (A new problem area, still fuzzy), N (Not a new problem area) Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for the real cases of countermeasures to counter economic espionage internationally. NE D OIP2. Looking for the real cases of crimes committed by people who have been previously convicted and later released or paroled from prison.

NE D 1

OIP3. Looking for the real cases of any ferry boats sinkings throughout the world. NE D OIP1. Information on analogy relation between evaporative cooling and cooling tower. RS F OIP2. Looking for different applications of evaporative cooling: cool water or air. RS N

2

OIP3. Looking for the issues on evaporative cooling of air. RS F OIP1. Looking for a cheap bicycle, brand new or second-hand. SH D OIP2. Looking for the literature on the topic of traveller’s route-choice behaviour on urban public transit network.

RS N 3

OIP3. Looking for the latest news about China’s spaceship “Shenzhou-7”, in particular, about the astronauts’ food.

NE F

OIP1. Information relating to Marxian appropriation: contemporary use and current theories applying this perspective (keywords: natural, species & social being).

RS N

OIP2. Looking for a foreign film I saw on SBS. MM F

4

OIP3. Information on coming releases in industrial and EBM (Electronic Body Music) music. MM F OIP1. Looking for the companies with twenty to thirty people only in the Brisbane CBD. JC N OIP2. Looking for the journalists of media outlets on health & fitness. JC N

5

OIP3. Looking for right formula 1game Websites (play online). GA N OIP1. Information on multi-channel video content delivery. RS D OIP2. Information on how to transmit video data in the network environment. RS D

6

OIP3. Looking for the present methodology in video transcoding. RS D

100

Table 4-10. (Continued)

Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for the calculation methods for trip generation rates, person trips or vehicle trips, for Transit Oriented Developments (TODs).

RS D 7

OIP2. Information on travel behaviours of people living using a TOD: mode choice, trip length, walking trips, intra-zonal trips, etc.

RS N

OIP3. Information on statistical analysis of travel data including demographic, socioeconomic and travel/ trip details. E.g. the relation between mode choice and age of person.

RS F

OIP1. Information on the effect of passengers crowding at bus stop/station on the dwelling time(s) of buses. RS N OIP2. Information on the current practice of road pricing (Toll Road) around the world. RS N

8

OIP3. Looking for different day tours available in Malaysia: tour operators (guides) and cost involve. TA D OIP1. Looking for the real potential of finding a good job in statistics in Australia and the rest of the world. JC D OIP2. Information on the effects of USA financial crisis on Australian job market and economy in many ways in future and what I should do to protect myself.

JC D 9

OIP3. Looking for the scholarships for postgraduate research students in Australian Universities. JC N OIP1. Information on how the great chain of being has manifested over the millennia. RS D OIP2. Information on the link between food and education: socio-cultural explanations of the role of food in education.

RS F 10

OIP3. Information on septennial taxonomy of everything: seven planes describe/repeat all things known and unknown in the universe, and the relationship between what we know of these seven planes.

RS F

OIP1. Looking for the cheapest ticket to New Zealand during summer vacation. TA D OIP2. Looking for the activities I can do in New Zealand. TA F

11

OIP3. Information on advertising issue of fashion industry, e.g. using female body to sell products, such as clothes, watch, and perfume.

AS F

OIP1. Looking for the cheapest yet most wide-travelling route of car hire and travel from Atlanta to Las Vegas. TA F OIP2. Looking for the names of the schools or agencies that will offer a summer school camp in South Korea. I would like to teach English near Seoul and be there for more than 1 week.

JC D 12

OIP3. Looking for brief accounts of overseas peoples’ reclamation of sacred sites and Australian cases, Indigenous land and sea rights and protection of traditions in Australia.

AS F

101

Table 4-10. (Continued)

Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for general information on asset management and specific area such as public asset management: how the government in developed and developing countries manage their public asset, and whether there is any best practice or guidelines, etc.

RS N

OIP2. Information on how local government being structured in Australia, Queensland and in Brisbane, and the comparison with Indonesian local government’s structure.

RS F

13

OIP3. Information on the definition of public policy and the factors affecting the policy maker when making a particular policy and the process of public policy making.

RS D

OIP1. Information on Chinese consumers’ mobile phone adoption behaviours: How they respond to new mobile technologies, and What factors affect their adoption behaviours?

RS N

OIP2. Looking for the status quo of 3G technology in China market. RS N

14

OIP3. Looking for the differences of consumer behaviours toward new innovation between two China regional markets: North market and East market.

RS F

OIP1. Looking for the 10 schools, private or state, in radius 30 minutes drive from my house (Eight Mile Plains).

JC D

OIP2. Looking for caravan parks around Melbourne and those are in the way to Melbourne: name and price. TA D

15

OIP3. Looking for the cheapest accommodation and flight to New Zealand on the mid of January, for 2 weeks.

TA D

OIP1. Looking for some program examples written in JAVA and read in XML files from an RSS feed & display in a swing GUI (RSS News Aggregator—JAVA Based).

AS N

OIP2. Looking for racing statistics of Easy Rocking Filly (female baby horse, 2 years old): what distances this race horse will be best suited to run over.

SP F

16

OIP3. Looking for the predictions for surf conditions at the end of November at Byron Bay, northern NSW. SP D

102

Table 4-10. (Continued) Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for some directions for implementing e-portfolio systems (“e-portfolio roadmap”) in the VET (Vocational Education & Training) sector over the next 3-5 years.

JC N

OIP2. Looking for relevant research in the area of “Question–Answers (QA) research” in information systems. RS F

17

OIP3. Information on new tools that might be available for making natural language searching queries on the Web. RS D OIP1. Information on machine vision grade cameras: Australian suppliers, quality, the best resolution available, price, and fully integrated solutions (onboard storages).

TE F

OIP2. Information on how Rotary Engines work? What cars use rotary aside from Mazda’s? Are they more efficient? What is the history of the engine? Is there a reason if not used more frequently?

TE F

18

OIP3. Looking for the most likely power plants (Hydrogen, Ethonal, etc.) which will replace the combustion engine in modern cars.

TE F

OIP1. Overview information on the creative clusters development situation, especially in Australia, the UK and China.

RS D

OIP2. Information on the differences between the concepts of “clusters”, “precincts”, and ”parks”. RS F

19

OIP3. Looking for basic ideas about “media/culture convergence” in the creative clusters and the researchers who are doing research in this area.

RS F

OIP1. Looking for finite element modelling of cold-formed steel sections subject to local buckling. RS N OIP2. Information on experimental investigation of cold-formed steel beams. RS D

20

OIP3. Information on inelastic reserve capacities of cold-formed steel beams. RS D OIP1. Information on the differences among liability, responsibility and obligation, especially for the use of GNSS (Global Navigation Satellite System).

RS F

OIP2. Information on the rank of Arsenal in Britain Soccer League: Who kicked the goal in last match in the team? SP N

21

OIP3. Information on the technology Real-time kinematic (RTK) Integrity and its application in the surveying: How many companies are doing research on RTK integrity? Check the company, such as Leica, Trimble, Topcon, and Ashtech.

RS D

103

Table 4-10. (Continued)

Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Information on correlation analysis for the reference in my Literature Review chapter. RS D OIP2. Looking for presentation/ solution/ answer/ suggestion/ lecture notes on the basic skill of dealing with difficult people.

JC D 22

OIP3. Information on the way to lose weight especially for people who often sit in offices. LI D OIP1. Looking for interest rates for my home loan: the fixed & variable rates of Westpac, and compare them to the interest rates of other Australian banks.

FI N

OIP2. Information on USA presidential election: the likely winner based on polls, and historical comparison from turn around this far from an election.

NE F

23

OIP3. Information on Nuclear power in Australia: how long would a plant be up and running, the CO2 decrease, and the recent history of safety (1990s —now).

JC F

OIP1. Information on washing machine company Fisher & Paykel: company history, experience implementing Kaizen for one of their division, and related organisational behaviour theories to the case study of Kaizen adoption by Fisher & Paykel.

AS F

OIP2. Looking for SAP Business solution: definition of solution and what modules they offer. AS D

24

OIP3. Looking for sample of IT project framework. AS F OIP1. Looking for an example of text/discourse analysis from a socio-linguistic perspective and an example of discourse/text analysis of an Oscar Wilde text.

AS N

OIP2. Looking for general information about “Discourse Analysis”. AS N

25

OIP3. Looking for a detailed synopsis of the play “The Importance of Being Earnest” by Oscar Wilde. AS N OIP1. Looking for the update information of the instrument “distress thermometer” (DT). RS N OIP2. Information on the function of Shark Cartilage, help relieve the pain of arthritis or reduce the occurrence of pain?

LI D 26

OIP3. Information on the function of Omega-3, useful for decreasing cholesterol? LI F OIP1. For a holiday to Victor Harbor in South Australia, looking for what to do there and where we can stay. TA D OIP2. Looking for information about mobile phones in neighbouring Indonesia. RS N

27

OIP3. Information on the word “Beothuks” which came up in a novel set in Newfoundland in Canada. AN F

104

Table 4-10. (Continued) Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for the best/most highly-reviewed French restaurants that serve authentic food with a good ambience in Brisbane.

FE D

OIP2. Holiday packages to Vanuatu, Fiji & Cook Islands. Looking for a good deal covering airfare, accommodation, and local tours if available.

TA D

28

OIP3. Information on how gender, class, and perceived identity affects language use in the TV series ”Summer Heights High”.

AS D

OIP1. Looking for quality, economical and value-for-money tour packages to tourist attractions in Queensland. TA N OIP2. Looking for a casual job in Brisbane that provides me full flexibility and the Website is very user-friendly. JC N

29

OIP3. Looking for the online cheapest flights to Sydney. TA N OIP1. Looking for experts’ statements on IT procurement and strategy, and the reason why procurement is essential for IT department.

AS D

OIP2. Information on the latest movies or hit music. MM N

30

OIP3. Looking for the Chinese meaning of some IT acronyms, e.g. TCP, UDP, and the differences between the two.

RS F

OIP1. Looking for the recipe for Italian noodles cooking. FE D OIP2. Information on sunglasses shopping in Brisbane: shops, prices, and types, especially for those who are myopia/short-sighted.

SH D 31

OIP3. Looking for the Journals related to Information Systems and Management included in SCI, EI, or ISTP. RS F OIP1. Looking for papers and some comments on companies’ earnings forecasts. RS D OIP2. Looking for papers with comments on companies’ earnings forecasting reputation. RS D

32

OIP3. Looking for papers of analysts’ earning forecasts about some companies. RS D OIP1. Looking for significant information searching behaviour models related to Web searches. RS D OIP2. Looking for the literatures in query formulation/ modification analysis and their implications. RS N

33

OIP3. Looking for important conferences in image retrieval for paper writing plan. RS D

105

Table 4-10. (Continued)

Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for news and articles relating to the sub-prime credit crisis in the USA and its relation to the possible current recession in global financial markets.

AS F

OIP2. Looking for financial and strategic data regarding an Indian Biotechnology firm—Glenmark and its financial performance relative to the Biotechnology industry in India.

FI D

34

OIP3. Looking for career opportunities in the Venture Capital Industry globally. JC D OIP1. Information on the newest Mary Jane shoes from Marc Jacobs. SH N OIP2. Information on all infrastructure provider companies in Brisbane that has a large vehicle fleets division. RS F

35

OIP3. Information on the newest national code in good governance for Indonesia. RS D OIP1. Scholar papers about sea change: demographics, statistics, and background. RS F OIP2. Information on mediation organisations as a potential future employment in the Gold Coast area. JC F

36

OIP3. Scholar papers about diasporic media. RS F OIP1. Information on “Nutrition Transition” (NT): Who studied about it? What did they find? What is the significance? What sorts of projects/programs are doing now? Where? How about NT in Thailand?

RS F

OIP2. Information on “Migrant”: How many Thais have moved to other countries? Where? How many are Thais in Australia? More male or female? Where do Thais locate in Australia? How many are Thais in Queensland? How about their health status? What is the theory of migrants?

RS F

37

OIP3. Information on “Symbolic Interactionism” (SI): What is SI? Who created/developed/discovered SI? How? RS F OIP1. My ex-boss’ contact information in order to get a recommendation letter. JC D OIP2. Information on the concept of “Flow”. RS F

38

OIP3. Information on getting wireless access to QUT wireless service from my own MAC laptop and iPhone at my QUT office.

TE D

OIP1. Information on common strategies used in human resource management. AS D OIP2. Information on common traits across organisations within the biotechnology industry: size, skills, and competitive advantages.

AS F 39

OIP3. Information on common attributes regarding human resource management in the biotech industry: issues, solutions, and examples.

AS D

106

107

Table 4-10. (Continued)

Study Participant Original Information Problem (OIP) Topic

Area Status

OIP1. Looking for entertainment places in Brisbane. FE N OIP2. Information on “impact evaluation of university libraries” by some Web search engines. RS N

40

OIP3. Renting a house near the city. RE F OIP1. Information on ethical problems in new drugs development for HIV treatment. RS D OIP2. Looking for the recipe for curry cooking. FE F

41

OIP3. Information on tax refund. FI D OIP1. Information on grounded theory researchers in China. RS N OIP2. Information on grounded theory seminars worldwide. RS N

42

OIP3. Conference information in nursing. RS N

Information Problem Topic Area

Study participants’ 126 self-chosen information problems (Table 4-10, three

information problems per study participant * 42 study participants) were categorised

into fifteen topic areas (Table 4-11).

Table 4-11. Information problem topic area Types of Topic Area Number % Research (RS) 56 45 Assignment (AS) 15 12 Jobs & Careers (JC) 14 11 Travel & Airfares (TA) 10 8 News (NE) 5 4 Food & Entertaining (FE) 4 3 Technology (TE) 4 3 Finance (FI) 3 2 Lifestyle (LI) 3 2 Music & Movie (MM) 3 2 Shopping (SH) 3 2 Sports (SP) 3 2 Answers (AN) 1 1 Games (GA) 1 1 Real estate/Rentals (RE) 1 1 Total 126 100 • Generally, study participants’ Web searching information problems covered a

wide range of topic areas, relating to research, assignment, future jobs &

careers, travel, news, food & entertaining, technology, finance information,

lifestyle, favourite music & movies, online shopping, sports information, answers

to a specific inquiry, online gaming, and housing rentals.

• The searching topic areas listed in the Table 4-11 were similar to the Yahoo

category system provided on its Portal Website (http://au.yahoo.com/). There

was little difference between university postgraduate students and the mass in

terms of general Web searching topic areas (Research, Jobs & Careers,

Entertainment, etc.).

108

• Most topics were research (45%) related, followed by assignment (12%), jobs

(11%) and travel (8%). The rest of the topics were scattered across news (4%),

food (3%), technology (3%), finance (2%), lifestyle (2%), music & movie (2%),

online shopping (2%), sports (2%), answers (1%), online gaming (1%), and

rentals (1%).

Information Problem Status

The status of an information problem refers to either a new problem area that is

precisely defined, or a new problem area that is still fuzzy, or not a new problem

area. Table 4-12 summarises the status of original information problems given by

forty-two study participants prior to their searching.

Table 4-12. Status of original information problems prior to the searching

Status Number % A new problem area, precisely defined (D) 50 40 A new problem area, still fuzzy (F) 42 33 Not a new problem area (N) 34 27 Total 126 100

Most information problems were a new problem area, precisely defined (40%) or

still fuzzy (33%). The remaining 27% of the information problems were not a new

problem area for the study participants. Web searching episodes may vary from the

different status of information problems.

109

Related or Unrelated Information Problems

Each study participant’s three original information problems were either related or

unrelated. Table 4-13 provides a summary of the number of study participants who

conducted related or unrelated multiple information problems searching.

Table 4-13. Number of study participants with related or unrelated information problems searching

Study Participant Number %

Unrelated information problems 29 69 Related information problems 13 31 Total 42 100 Searching on multiple information problems, related or unrelated, is a common

Web searching behaviour (Spink, Ozmutlu & Ozmutlu, 2002). In this study, over

two-thirds (69%) of study participants pooled three unrelated information problems

together and conducted searching concurrently during one Web search episode.

The other 31% of the study participants conducted Web searching on three related

information problems.

Example: Unrelated information problems

• The three information problems were bicycle selling, research topic on

urban public transit, and recent Chinese spaceship news. (Study Participant

3)

• The first two information problems referred to the cheapest airfare to New

Zealand and activities available there, the third problem was course

assignment related. (Study Participant 11)

110

• The three information problems were an ethics problem in new drug

development, curry cooking and the tax refunds policy, respectively. (Study

Participant 41)

Example: Related information problems

• The three information problems were all related to the PhD research topic of

evaporative cooling and cooling tower. (Study Participant 2)

• All three problems were research topic related: the development status quo

of creative clusters, the differences between the concepts “clusters”,

“precincts” and “parks”, and media convergence research within creative

clusters. (Study Participant 19)

4.4.2 Factors Affecting Information Problem Search Ordering

4.4.2.1 Information Problem Search Ordering The results show that the study participants gave a priority to searching order

among their multiple information problems. Figure 4-1 demonstrates how the study

participants ordered the three Web search problems.

111

Figure 4-1. Ordering of information problems

0

2

4

6

8

10

12

14

16

18

20

Ordering of Information Problems

1st Search 6 0 1 3 0 5 0 0 1 20 0 1 0 1 4

2nd Search 3 0 1 1 0 6 1 2 2 19 0 1 2 1 3

3rd Search 6 1 1 0 1 3 2 1 2 18 1 0 1 2 3

AS AN FI FE GA JC LI MM NE RS RE SH SP TE TA

The figure data shows that:

• Nearly half of the study participants’ (20 in 42) first Web search was for

information on their Research (RS) topic, followed by Assignment (AS, 6 in

42) and Job information (JC, 5 in 42).

• Nearly half of the study participants (19 in 42) conducted their second Web

search on the research (RS) topic, followed by Job information (JC, 6 in 42),

and Assignment (AS, 3 in 42) and Travel information (TA, 3 in 42).

• Nearly half of the study participants (18 in 42) conducted their third Web

search on the research (RS) topic, followed by Assignment (AS, 6 in 42),

and Job (JC, 3 in 42) and Travel information (TA, 3 in 42).

Overall, research topic was a main information problem which concerned the study

participants while searching on the Web, either for the first search, the second

search, or the third search. This finding was not unexpected for postgraduate

112

113

student participants. Apart from the research topic, assignment (AS), jobs & careers

(JC), and travel & airfare (TA) information were frequent Web searching topics.

4.4.2.2 Reasons for Information Problem Search Ordering The study participants were asked to provide the reasons for their information

problem search ordering on the post-Web search questionnaire. Further

explanations were solicited during the post Web-search interview. Results show

that the perceived factors mainly decided the searching order.

Table 4-14 lists the perceived factors that affected information problem searching

order by each study participant.

Table 4-14. Factors affecting information problem search ordering

Factors

Study Participants

Ease of Information Finding—

High to Low

Ease of Information Finding—

Low to High

Future Usefulness

Personal Interest

—High to Low

Problem Familiarity —High to

Low

Problem Importance

Level —High to

Low

Problem Importance

Level —Low to

High

Problem Urgency

Level —High to

Low

Randomness Task Logic

1 √ 2 √ 3 √ √ √ 4 √ 5 √ √ 6 √ √ 7 √ 8 √ 9 √ √ 10 √ 11 √ √ 12 √ 13 √ 14 √ √ 15 √ 16 √ √ √ 17 √ 18 √ √ 19 √ √ 20 √

114

Table 4-14. (Continued)

Factors

Study Participants

Ease of Information Finding—

High to Low

Ease of Information Finding—

Low to High

Future Usefulness

Personal Interest

—High to Low

Problem Familiarity —High to

Low

Problem Importance

Level —High to

Low

Problem Importance

Level —Low to

High

Problem Urgency

Level —High to Low

Randomness Task Logic

21 √ 22 √ 23 √ 24 √ 25 √ 26 √ √ √ 27 √ √ 28 √ 29 √ 30 √ √ 31 √ √ 32 √ √ 33 √ 34 √ √ 35 √ √ 36 √ 37 √ 38 √ 39 √

115

116

Table 4-14. (Continued)

Factors

Study Participants

Ease of Information Finding—

High to Low

Ease of Information Finding—

Low to High

Future Usefulness

Personal Interest

—High to Low

Problem Familiarity —High to

Low

Problem Importance

Level —High to

Low

Problem Importance

Level —Low to

High

Problem Urgency

Level —High to Low

Randomness Task Logic

40 √ 41 √ √ √ 42 √ Total 10 4 2 6 3 12 2 6 11 7 % 24 10 5 14 7 29 5 14 26 17

Table 4-15 summarises the factors that affected study participants' information

problem searching order.

Table 4-15. Summary of the factors affecting multiple information problems search ordering

Factors Number % Level of problem importance—high to low 12 29 Randomness/ no specific reasons 11 26 Ease of finding information—high to low 10 24 Task Logic 7 17 Level of problem urgency—high to low 6 14 Task interest—high to low 6 14 Ease of finding information—low to high 4 10 Level of problem familiarity—high to low 3 7 Level of problem importance—low to high 2 5 Future usefulness 2 5 It shows that multiple information problem search ordering was affected by the

following factors: problem importance level, random order, ease of finding

information on the Web, task logic, problem urgency level, task interest, problem

familiarity level, and future usefulness.

• Problem importance level—high to low (N=12, 29%), randomness (N=11,

26%), and ease of finding information—high to low (N=10, 24%) were the

major factors in determining information problem search ordering.

• Followed by task logic (N=7, 17%), problem urgency level—high to low

(N=6, 14%), task interest (N=6, 14%), ease of finding information—low to

high (N=4, 10%), and problem familiarity level—high to low (N=3, 7%).

• The last two factors, problem importance level—low to high (N=2, 5%) and

future usefulness (N=2, 5%), were little considered when ordering the

searching tasks.

The findings are different from Spink, Park and Koshman’s (2006) conclusions, in

which task interest and problem familiarity level were the two major ordering factors.

117

For each factor identified in this study, detailed interpretations were presented in

the following paragraphs.

1). Level of Problem Importance—High to Low Almost 29% of the study participants ordered their information problems from the

most important to less important ones. It was the largest percentage compared to

other factors.

Examples:

• They were in order of the importance. Problem 1 was the most important. I

was interested in problem 2, but it was not really important. I’ve already

found some information on problem 3, it was less important. (Study

Participant 9)

• The first searched problem was the most important one. It was a necessary

topic in my PhD thesis. The second problem mattered to job hunting, but

was less important. And the third one was not important at all. (Study

Participant 22)

2). Randomness/ No Specific Reasons Interestingly, the second most considered factor was randomness. 26% of the

study participants stated that they ordered the multiple information problems

without any specific reasons.

Examples:

• Randomness was listed as the only factor. (Study Participant 1)

118

• They were just popped up in my mind, no specific reasons. (Study

Participant 8)

• Three problems were not related, there was no need to order them

intentionally. (Study Participant 18)

• I did not mean to order them, I just wrote them down in that order. They

were ordered pretty much randomly. (Study Participant 25)

3). Ease of Finding Information—High to Low Some 24% of the study participants listed high ease of finding information on the

Web as a major factor. They tended to leave the most involved problem to the last.

Examples:

• Problem 1 was easier. I thought I would take less time. Problem 2 was more

difficult which I may take longer time. Problem 3 was a very broad problem.

I knew it would take the longest time. (Study Participant 12)

• I was very confident to find the information easily on the first problem. For

Problem 2, a bit more difficult, also it was my main interest really today. And

problem 3 was a complex question in many ways. (Study Participant 17)

4). Task Logic Another 17% of the study participants ordered their multiple information problems in

a logical way.

Examples:

• The three information problems were ordered depending on the

development of the theory. (Study Participant 2)

119

• The searching sequence was the order that I would do my analysis. (Study

Participant 7)

• All three problems were related in a sequence manner. Problem 3 was

dependent on the first two problems. So I started with problem 1, followed

by problem 2 and ended up with problem 3. (Study Participant 20)

5). Level of Problem Urgency —High to Low Some 14% of the study participants listed high urgency as a major factor.

Examples:

• I was trying to find problem 1 out as I would have a party on this Friday. It’s

the most urgent one. I had to do problem 2. It was important but not urgent.

Problem 3 was neither urgent nor important. We would have a meeting next

week. (Study Participant 35)

• Problem 1 was about my assignment which was due to 23rd May, this

Friday, just two days after. It really mattered. I really needed some results.

(Study Participant 41)

6). Task Interest—High to Low Another 14% of the study participants put the most interesting problem as the first

search.

Examples:

• I liked playing games online which was my interest. I preferred to search for

information on it firstly. I did not like Problem 2, but I had to do it at

workplace. I always tried to avoid it and did it in the last minute. (Study

Participant 5)

120

• Problem 1 and problem 2 were the tasks I really wanted to do, they were

more interesting. Problem 3 was the task I had to do. (Study Participant 11)

7). Ease of Finding Information—Low to High Interestingly, four study participants preferred to work firstly on the problems which

they felt harder to find information on the Web.

Example:

• I did problem 1 first as I knew it must take most of time to find the

information that I wanted. For the other two problems, I did not think they

would take more than five minutes each. (Study Participant 4)

8). Level of Problem Familiarity—High to Low Three study participants ordered their Web searches according to the high

familiarity level.

Example:

• I started from the area which I knew better so that I knew where to start.

(Study Participant 13)

9). Level of Problem Importance—Low to High Two study participants firstly searched for information on the problems which were

viewed less important.

Example:

• I wanted to put the most important one at last and spent more time on it.

(Study Participant 31)

121

10). Usefulness in Future Two study participants considered future usefulness as a factor affecting searching

order.

Examples:

• The resolution of problem 1 would help me in my future workplace. (Study

Participant 5)

• Tax refund problem was listed as the third one because it would be helpful

in future. (Study Participant 41)

11). Multiple Factors Applied Multiple factors were applicable for the explanations of search ordering. In real-life

Web searching, information problems search ordering was normally determined by

more than one factor, which was more complex than our expectation. Listed here

are some examples:

• The first information problem of getting a cheap bicycle was a real problem

in my life and I needed it now. I knew it would be easier for me to find

information on the other two problems. The third problem was regarded to

my task interest. I searched it just for fun. (Study Participant 3)

Three factors—high problem urgency, task interest, and lease of finding

information on the Web (low-to-high)—were all considered when ordering

his three information problems.

122

• Problem 1 was the most important one. I was interested in Problem 2, but it

was not really important. Problem 3 was not interesting. (Study Participant 9)

Both problem importance level (high-to-low) and task interest were

considered in this case.

• I had different levels of confidence in finding the required information on the

three problems. I searched for the topics which I thought I would find first

and the less likely searches followed. Also I had to finish the first problem of

my assignment by this weekend. It was the most urgent one. After this

weekend, I could have a holiday like surfing. (Study Participant 16)

In this case, ease of finding information (high-to-low) and problem urgency

level (high-to-low) were viewed as two factors.

• Problem 1 was a background of my PhD research. I spent the longest time

on it as it was the most important information problem. Another reason I

ordered it as the first one was that I thought it could be solved, it was a

solvable question. I could get satisfied results. Problem 2 was also related

to my PhD research. But I was not that sure whether it could be solved

through the searching. (Study Participant 19)

Here, problem importance level (high-to-low) and ease of finding relevant

information (high-to-low) were two determined factors.

4.4.3 Evolving Information Problem

An evolving information problem (EIP) was a type of information problem that was

produced during the course of successive Web searching. It reflected a dynamic

and interactive property of the Web searching behaviour. Compared to an original

123

information problem which was consciously set by study participants, the

generation of an evolving information problem had an attribute of improvisation. It

could be conscious or unconscious.

Table 4-16 listed all the evolving information problems generated during forty-two

study participants’ Web searches.

124

Table 4-16. Evolving information problems generated per study participant Study Participant Evolving Information Problem (EIP)

1 None 2 None 3 EIP1. Looking for the Website URL of “eBay”. 4 None

EIP1. Looking for the Website URL of “EA Sports”. EIP2. Looking for the Website URL of “Microsoft Games”. EIP3. Information on names of gaming companies. EIP4. Looking for the Website URL of “MSN answers”. EIP5. Looking for the Website URL of “a CBD building”. EIP6. Looking for the Website URL of “another building”. EIP7. Looking for the Website URL of “another building”.

5

EIP8. Looking for the Website URL of “another CBD building”. 6 None 7 None

EIP1. Information on the transit visa in Malaysia. 8 EIP2. Looking for the differences between revenue generation and travel demand management. EIP1. Information on the job salary as a statistician. EIP2. Looking for scholarships for international students in Australia. EIP3. Looking for some news headlines about USA financial crisis. EIP4. Information on how to become an actuary.

9

EIP5. Information on how to become a statistician. EIP1. Looking for the book reviews of a book in Google Books. 10 EIP2. Looking for the book twentieth-century world in Google Books. EIP1. Looking for the cheapest ticket from Brisbane to Auckland. EIP2. Looking for the Website URL of “Jetstar”.

11

EIP3. Looking for the cheapest ticket from Brisbane to Christchurch. 12 EIP1. Information on how the hired car can be returned in USA.

EIP1. Information on the definition of concepts: local and municipal. EIP2. Looking for the local government structure in Canada. EIP3. Looking for the local government structure in Malaysia.

13

EIP4. Looking for the Website URL of “Yahoo Scholar”. 14 None

EIP1. Looking for the surrounding suburbs around my home in Google Maps. EIP2. Information on the distance between a parking place and Melbourne. EIP3. Looking for the cheapest flight from Brisbane to Auckland. EIP4. Looking for the cheapest accommodation in Auckland CBD.

15

EIP5. Information on the area in Auckland. EIP1. Looking for the Website URL of “Cuil”. EIP2. Looking for the Website URL of “Google”.

16

EIP3. Looking for easy rocking stallion statistics. EIP1. Information on the structure of questions & answers in automatic FAQ system. EIP2. Looking for the reason why there is a financial crisis.

17

EIP3. Looking for the reason why we age.

125

Table 4-16. (Continued) Study Participant Evolving Information Problem (EIP)

EIP1. Information on the concept of "gige" in Wikipedia. EIP2. Looking for the Website URL of “Wikipedia”. EIP3. Looking for the comparison between wankel engines and combustion engines. EIP4. Information on the future of the cars, like hybrid cars.

18

EIP5. Looking for the comparison of hybrid car technologies. 19 None 20 EIP1. Information on the experimental finite element modelling

EIP1. Information on the legal meaning of responsibility. 21 EIP2. Information on the legal meaning of liability. EIP1. Information on correlation analysis in data mining. EIP2. Information on correlation analysis association rule mining. EIP3. Information on how to deal with difficult people in software engineering.

22

EIP4. Looking for a diet plan for losing weight. EIP1. Information on the big four banks in Australia. EIP2. Looking for Website URL of “New York Times”. EIP3. Information on the USA presidential election date. EIP4. Information on Uranium Information Centre.

23

EIP5. Information on nuclear waste disposal. 24 EIP1. Looking for the motivational theories adopted by Fisher & Paykel. 25 None

EIP1. Looking for the URL of “PubMed” database. EIP2. Looking for the papers written by the author Yamagishi, A. EIP3. Information on the Chinese meaning of the word “angiogenesis”. EIP4. Looking for the Website URL of “online dictionary”.

26

EIP5. Information on the dosage of shark cartilage in order to cure arthritis. EIP1. Information on the relationship between Victor Harbor and Adelaide. EIP2. Information on the relationship between Victor Harbor and Granite Island.

27

EIP3. Information on the distance between Monarto Zoo and Victor Harbor. EIP1. Information on “Edgewater Resort”. EIP2. Looking for the reviews on “Edgewater Resort”. EIP3. Looking for the reviews on “Cook Islands” packages for couples.

28

EIP4. Information on the research done by the author Larissa Mclean. EIP1. Looking for tour packages from Brisbane to Bryon Bay. EIP2. Looking for cruise tour packages from Australia to anywhere.

29

EIP3. Looking for tour packages in Cairns. EIP1. Looking for the Website URL of “Wikipedia”. 30 EIP2. Information on the concept of “procurement”. EIP1. Looking for an English word with the meaning of “short-sighted”. 31 EIP2. Looking for the real location of the virtual store”eyeglasses4you”.

32 EIP1. Looking for the URL of “Google Scholar”. 33 None

126

Table 4-16. (Continued) Study Participant Evolving Information Problem (EIP)

EIP1. Looking for the resolution of subprime crisis. EIP2. Looking for the qualifications for jobs in Venture Capital industry. EIP3. Looking for career opportunities in the Venture Capital Industry in India.

34

EIP4. Looking for salary information in Indian venture capital industry. 35 None

EIP1. Looking for the contact details of the Australia Institute. EIP2. Looking for the Website URL of “Australia ABS”. EIP3. Looking for Australian government portals.

36

EIP4. Scholar papers about diasporic telecommunication media. 37 EIP1. Information on migrants in Australia.

EIP1. Looking for the Website URL of “McGraw-Hill publisher”. EIP2. Looking for the email from Stephanie Beames in the Yahoo email Inbox. EIP3. Looking for the books on flow in QUT Library. EIP4. Looking for a correct spelling of the book author’s name. EIP5. Information on the author of the book “Emotional Intelligence”.

38

EIP6. Looking for the books on emotional intelligence in QUT Library. 39 None 40 None 41 EIP1. Looking for the URL of “PubMed” database.

EIP1. Information on grounded theory seminars in China or in Australia. EIP2. Information on grounded theory seminars in Australia. EIP3. Looking for the Website URL of “all conferences”.

42

EIP4. Conference information in nursing focusing on qualitative methods.

127

Table 4-17 shows the number of study participants with evolving or non-evolving

information problems during Web searching.

Table 4-17. Number of study participants with evolving or non-evolving information problems developed

Study Participant Number of Evolving Information Problems Number % Zero 12 29 One 7 17 Two 5 12 Three 5 12 Four 6 14 Five 5 12 Six 1 2 Eight 1 2 Total 42 100 The number of evolving information problems per study participant ranged from

zero to eight.

• Twelve study participants (29%) did not generate any evolving information

problems, in the cases of Study Participants 1, 2, 4, 6, 7, 14, 19, 25, 33, 35,

39 and 40.

• Over 70% of the study participants developed evolving information

problems during the successive Web searching.

• Seven study participants (17%) developed one evolving information

problem, in the cases of Study Participants 3, 12, 20, 24, 32, 37 and 41.

• Five study participants (12%) developed two evolving information problems,

in the cases of Study Participants 8, 10, 21, 30 and 31.

• Another five study participants (12%) developed three evolving information

problems, in the cases of Study Participants 11, 16, 17, 27 and 29.

• Six study participants (14%) developed four evolving information problems,

in the cases of Study Participants 13, 22, 28, 34, 36 and 42.

128

• Five study participants (12%) developed five evolving information problems,

in the cases of Study Participants 9, 15, 18, 23 and 26.

• Only one study participant (2%) generated six evolving information

problems (Study Participant 38).

• One study participant (2%) generated eight evolving information problems

(Study Participant 5).

Compared with study participants’ original information problems, evolving

information problems were represented as changed problems or totally new ones.

An Evolving Information Problem as a Changed One Examples:

• Problem 1 was about statistics job market in Australia and the rest of the

world. After getting some information, he would like to check out specific

information on job salary for a statistician. (Study Participant 9)

• Problem 3 was about the scholarships for postgraduate students in

Australian Universities. The evolving information problem was about the

scholarships for international postgraduate students in Australian

Universities. It was identified as a changed problem by refining the scope to

international postgraduate students. (Study Participant 9)

An Evolving Information Problem as a New One Examples:

• The evolving information problem was searching for the URL of eBay

Website which was totally different from the original three problems. (Study

Participant 3)

129

• Problem 3 was for information on a day tour in Malaysia, including tour

guides and the cost involved. The evolving information problem was for

transit visa at Malaysia airport. (Study Participant 8)

4.4.4 Information Problem Searching Task Switching

Task switching necessarily existed during the multiple information problems Web

searching process. Information task switching behaviour was characterised as an

iterative process of ceasing to search or browse information on one task and

proceeding with another information task.

4.4.4.1 Types of Information Problem Searching Task Based on the analysis of Web search logs, three types of information problem

searching task were identified: searching on an original information problem (SOIP),

searching on an evolving information problem (SEIP), and serendipity browsing (SB)

on other topics.

1) Searching on an Original Information Problem (SOIP)

Study participants worked on the task of searching for information on their original

three information problems. SOIP1 means searching on the first original information

problem, SOIP2 means searching on the second original information problem, and

SOIP3 means searching on the third original information problem.

2) Searching on an Evolving Information Problem (SEIP)

Study participants worked on the task of searching for information on evolving

information problems. SEIP1 stands for searching on the first evolving information

problem, SEIP2 stands for searching on the second evolving information

problem, …, SEIPn stands for searching on the Nth evolving information problem.

130

3) Serendipity Browsing (SB) on Other Topics

Study participants browsed the opened Webpage/Website on other topics. It was

easily incurred by visual cues which prompted study participants to transfer their

attention to other interesting topics. SB1 means browsing on the first different topic,

SB2 means browsing on the second different topic, …, SBn means browsing on the

Nth different topic.

Study participants switched back and forth between three types of information

problem searching tasks, including searching on an original information problem

(SOIP), searching on an evolving information problem (SEIP), and serendipity

browsing (SB). Figure 4-2 illustrates the task switching behaviour of each study

participant.

131

Figure 4-2. Forty-two study participants’ information problem searching task switching

132

Figure 4-2. (Continued)

133

Figure 4-2. (Continued)

134

Figure 4-2. (Continued)

135

136

4.4.4.2 Task Switching Pattern During the forty-two study participants’ Web searches, the task of serendipity

browsing took place in only five cases: the Web searches of Study Participants 5, 8,

17, 38 and 41. Serendipity browsing was not found as a distinct behaviour in this

study. Therefore, this study focuses on original and evolving information problems

searching, reserving the analysis of serendipity browsing to further research.

All cases of the forty-two study participants’ searching task switches were clustered

into four distinct patterns, as shown in Table 4-18.

137

Table 4-18. Information problem searching task switching pattern

Pattern A Pattern B Pattern C Pattern D

Four sorts of task switching patterns, from the simplest to the most complicated,

were discussed as follows:

Pattern A

Task switching between original information problems searching: in order and

without iteration

Some study participants’ task switching processes were not complex. They switched

between three original information problems by order and without any iteration. This

situation was evidenced in the cases of Study Participants 1, 2, 7, 14, 25, 33, 35, 39

and 40.

Example: Study Participant 1

Pattern B

Task switching between original information problems searching: with iteration

In this pattern, study participants switched back and forth between their three

original information problems. It was evidenced in the cases of Study Participants 4,

6 and 19.

Example: Study Participant 6

138

Pattern C

Task switching between original information problems searching and one or two

evolving information problems searching

Under this circumstance, the course of task switching was more complicated than

the above two situations. During Web searching, one or two evolving information

problems were generated in addition to the original information problems. This

pattern was evidenced in the cases of Study Participants 3, 8, 10, 12, 20, 21, 24, 30,

31, 32, 37 and 41.

Example: Study Participant 10

Pattern D

Task switching between original information problems searching and three or more

evolving information problems searching

This pattern is much more complex. More evolving information problems were

generated. The maximum number of evolving information problems was eight in

one case. Under this circumstance, the task switching behaviour appeared very

active. Study participants frequently switched from one information problem

searching to another and jumped back and forth. This pattern was evidenced in

the cases of Study Participants 5, 9, 11, 13, 15, 16, 17, 18, 22, 23, 26, 27, 28, 29,

34, 36, 38 and 42.

Example: Study Participant 13

139

140

Table 4-19 provides a summary of task switching patterns across forty-two Web searches. Table 4-19. The percentage of each information problem searching task pattern Task Switching Pattern Number % Pattern A 9 21 Pattern B 3 7 Pattern C 12 29 Pattern D 18 43 Total 42 100 Pattern D is the main pattern of the searching tasks switching, in which 43% of

the study participants were situated. It means that most study participants

switched between original information problems searching and three or more

evolving information problems searching. The second most frequently occurring

task switching pattern was Pattern C (29%), followed by Pattern A (21%). There

were only three cases of Pattern B occurrences, accounting for 7% of the total

task switches.

4.4.4.3 Reasons for Information Problem Searching Task Switching This section elaborates the reasons for the searching tasks switching, that is,

what factors incurred users’ action of ceasing to search for or browse information

on one task and proceeding with another Web search task. Eight reasons were

revealed, based on the analysis of post-Web search questionnaires and post-

Web search interview notes. These are detailed in Table 4-20.

Table 4-20. Reasons for information problem searching task switching

Study Participant

Bored with the Current Information

Problem

Enough Information Was Found

Limited Time

Mental Fatigue

No Further Useful Information Could

be Found

Sufficient Time

Useful Links/ Resources Were Found

for Further Search Visual Cues

1 √ 2 √ √ 3 √ √ √ 4 √ √ √ 5 √ √ √ √ 6 √ √ 7 √ √ √ √ 8 √ √ √ 9 √ √ √ 10 √ √ 11 √ √ 12 √ √ 13 √ √ 14 √ 15 √ √ 16 √ √ 17 √ √ 18 √ √ √ 19 √ √ √ 20 √ 21 √ √ 22 √ √

141

142

Table 4-20. (Continued)

Study Participant

Bored with the Current Information

Problem

Enough Information Was Found

Limited Time

Mental Fatigue

No Further Useful Information Could

be Found

Sufficient Time

Useful Links/ Resources Were Found for Further

Search

Visual Cues

23 √ 24 √ √ 25 √ √ √ 26 √ √ 27 √ √ 28 √ √ √ 29 √ 30 √ 31 √ √ 32 √ √ 33 √ √ √ 34 √ √ 35 √ 36 √ √ 37 √ √ 38 √ √ √ 39 √ √ 40 √ 41 √ 42 √ √ Total 1 34 23 1 16 3 9 2 % 2 81 55 2 38 7 21 5

Table 4-21 provides a list of the reasons for study participants’ task switching.

Table 4-21. Summary of reasons for searching tasks switching Reasons Number % Enough information was found 34 81 Limited time 23 55 No further useful information could be found 16 38 Useful links/resources were found for further search 9 21 Sufficient time 3 7 Visual cues 2 5 Bored with the current information problem 1 2 Mental fatigue 1 2 Task switching was due to the following reasons: enough information was found,

limited time, no further useful information could be found, useful links/resources

were found for further search, sufficient time, visual cues, bored with the current

information problem, and mental fatigue. Enough information was found and

Limited time were the two major reasons.

1). Enough Information was Found The overwhelming study participants (81%) switched the searching tasks when

they were satisfied with the information they had found and believed that the

gathered information could solve the current information problem well.

Examples:

• I was happy with the results I got on problem 1, so I moved on to problem

2. (Study Participant 1)

• I’ve found what I wanted. After finding the solution for problem 1, I

switched to problem 2. (Study Participant 5)

• I got enough information for problem 1. I was happy with the results I got.

(Study Participant 11)

143

2). Limited Time Over half of the study participants viewed limited time as a factor pushing them to

switch between the tasks. The results showed that 74% of the study participants

chose number 3, 4, or 5 on the 5-point Likert scale (The number 1 stands for low

time pressure, number 5 stands for high time pressure), which means those

study participants felt high time pressure during multiple information problems

Web searching. Accordingly, they consciously allocated a certain time for each

information problem searching.

Examples:

• I needed more time to judge the quality of the returned results, but time

was limited, so I moved on to the third problem. (Study Participant 2)

• I was aware that I had only one hour, so consciously I wanted to allocate

equal amount of time to all the three problems. (Study Participant 5)

• As time was limited, if I got four or five good articles, I stopped my search.

(Study Participant 7)

• All right, 40 minutes passed, the second problem was done. Now, I had to

go on with the third information problem. (Study Participant 10)

• I was not quite satisfied with what I had got from this search, but I decided

to try next problem because the time was limited. (Study Participant 42)

3). No Further Useful Information Could be Found Another 38% of the study participants did not expect that they could find any

further useful information on the current information problem. Thus, it was

unnecessary to continue searching.

144

Examples:

• I did not think I could find any further useful information. No point for the

continuing searching. I did not want to waste my time. (Study Participant 3)

• I did not think there were a lot of Websites which had information on the

current problem. (Study Participant 16)

4). Useful Links/Resources Were Found for Further Search About 21% of the study participants believed that some useful links/resources

were found which could be traced later, even though the punchy information had

not been retrieved yet.

Example:

• I had gathered several related references and added them into my

Favourite. Then I checked them out one by one to see whether there was

any information I needed. (Study Participant 9)

5). Sufficient Time

Under this circumstance, still some time was left after the last problem searching

had been finished. Nearly 7% of the study participants were situated in this case,

who went back to the previous problems searching in order to check out further

useful information.

Examples:

• I still had a lot of time after I finished searching on the last problem, so I

went back to the previous ones to do further searching. (Study Participant

4)

145

• Still 40 minutes left, I’d like to find out more detailed information on the

first information problem as I had only got a few results about it. (Study

Participant 6)

6). Visual Cues For 5% of the study participants, their task switching was prompted by visual

cues. This happened more frequently among several related searching tasks.

Example:

• I found a connection from the current information problem to another one.

Hence, I temporarily stopped the searching on the current problem to this

problem. (Study Participant 13)

7). Bored with the Current Information Problem This reason refers to an emotional factor. Only one study participant’s task

switching occurred due to an interest shift when he became bored with the

current information problem and attempted to try another one.

Example:

• I got to leave that Marx stuff there because I was bored about it. (Study

Participant 4)

• 8). Mental Fatigue

Only one study participant complained that the search process wound down by

exhaustion and ended with a whimper. Information overload and mental fatigue

were in evidence.

146

Example:

• I partially solved problem 2, after that I got tired. So I stopped searching.

(Study Participant 5)

9). Multiple Reasons Applied Task switching is a humans’ advanced and complicated cognitive activity. In most

cases, study participants switched the tasks based on the consideration of

multiple factors.

Examples:

• I was happy with the searching results on problem 1 and I switched to

problem 2 searching. After searching for a while, I did not think I could find

any further useful information. I may need to consult with the

professionals instead. So I wished to have some changes and tried

problem 3. (Study Participant 26)

Here, two reasons for the tasks switching, “enough information was

found” and “no further useful information could be found” were applied.

• I was aware that I had only one hour searching time, so consciously I

wanted to allocate equal amount of time to all the three problems. After I

found what I wanted for problem 1, I switched to problem 2. I partially

solved problem 2, after that I got tired. So I stopped searching. (Study

participant 5)

Here, three reasons for the tasks switching, namely, “limited time”,

“enough information was found” and “mental fatigue” were applied.

147

4.4.5 Multiple Web Search Sessions

The first definition of a session given in the Oxford English Dictionary is “a period

devoted to a particular activity”. In this study, accordingly, a Web search session

is regarded as a period devoted to a particular Web information problem search.

A Web search session refers to the users’ practice of submitting the entire

sequence of queries through the interactions with the Web search systems over

time in the windows/tabs when searching on a particular information problem.

Four basic elements are involved in a Web search session:

1) the queries sequences

2) the selected Web search systems

3) the opened windows/tabs

4) a particular information problem

At the least, a user submits one query through conducting one Web search

interaction with one Web search system in one window/tab when searching on a

particular information problem. A Web search session definition more closely

discovers users’ actual/dynamic Web search interactions than previous definitions

(Jansen & Pooch, 2001; Spink, 1996; Spink, Park, Jansen & Pedersen, 2006), in

that it reflects users’ multitasking searching reality—the interactions with multiple

Web search systems in multiple opened windows/tabs.

Spink (1996) suggested that people conducted multiple search sessions by

developing and modifying their search queries and search strategies over many

subsequent search sessions. They found that 56.5% of the end-users conducted

multiple search sessions with an approximate mean number of search sessions of

148

3.33. However, the detailed analysis of the components of multiple search sessions

was not provided in her study.

The results of this study not only verified that users conducted multiple Web

search sessions, but that a Web search session had multiple characteristics, with

multiple queries, multiple Web search systems, and multiple windows/tabs.

Table 4-22 shows the number of Web search sessions conducted by the forty-two

study participants during the current Web searching.

Table 4-22. Number of search sessions conducted during the current Web searching

Study Participants Number of Search Sessions Number % 3 9 21 4 4 10 5 4 10 6 4 10 7 3 7 8 5 12 9 2 5 10+ 11 26 Total 42 100 All of the study participants conducted more than one search session during the

current Web searching. Study participants reported wide variation in the number

of search sessions from 3 to 16, with 26% reporting having conducted 10 or more

search sessions. The forty-two study participants conducted 315 search sessions

in total, with a mean number of search sessions of 7.5. Study participants

conducted multiple Web search sessions probably due to the current information

problem evolving or changing.

Table 4-23 indicates the mean number of the submitted queries, the selected

Web search systems, and the opened windows/tabs within each study

participant’s Web search session.

149

Table 4-23. Mean queries, Web search systems, and windows/tabs for an information problem searching per study participant Study Participant Mean No. of Queries Mean No. of Web

Search Systems Mean No. of

Windows/Tabs 1 6 1 5 2 4 1 2 3 8 3 5 4 7 1 13 5 6 3 2 6 5 2 1 7 9 2 11 8 4 1 6 9 3 1 1 10 5 2 2 11 1 1 1 12 6 1 1 13 5 3 5 14 4 1 9 15 3 2 10 16 6 2 2 17 5 2 2 18 10 1 4 19 1 1 13 20 4 2 3 21 12 3 10 22 3 3 2 23 7 2 2 24 4 1 1 25 2 1 1 26 4 2 2 27 2 1 7 28 3 2 10 29 2 3 1 30 2 2 3 31 10 3 13 32 3 2 2 33 7 2 9 34 2 1 1 35 3 1 1 36 3 2 3 37 6 2 2 38 4 3 2 39 6 2 2 40 2 1 5 41 5 2 2 42 2 1 1 Mean 5 2 4

150

A Web search session entailed mental processes and coordination work. The

degree of complexity of a Web search session depends on how many times the

queries, the Web search systems and the windows/tabs were utilised.

The Web search session data, over forty-two study participants’ Web searches,

shows that:

• For a particular information problem search, the number of the submitted

queries varied from 1 to 12, with the mean of 5; the number of the

employed Web search systems varied from 1 to 3, with the mean of 2;

and the mean number of the opened windows/tabs was 4, with a range of

1 to 13.

• On average, a study participant conducted a Web search session through

submitting five queries with two Web search systems in four opened

windows/tabs.

Study participants were not satisfied with the results from the first interaction or

they wanted to get more results through the subsequent interactions. A Web search

session within an information problem searching was characterised as changing

the search strategies, such as modifying the queries and adopting another Web

search system, followed by the first search interaction.

Multiple queries and multiple Web search systems were needed in order to

constitute multiple Web search interactions during a Web search session.

Interestingly, study participants handled multiple windows/tabs concurrently

without being confused. Instead, they believed that browsing multiple

windows/tabs at the same time was an efficient way of time management. In

addition, study participants did not browse the opened Websites in order. They

151

jumped from one window/tab to another on a “first in, first served” basis because

they did not want to waste time to wait for a certain Website to fully open. Namely,

they browsed the window/tab which turned up first.

Similar results were found in Park’s (2008) study. Multiple search tools, such as

academic databases and commercial Web search engines, and multiple tabbed

browsing techniques were utilised when people coordinating multiple information

tasks during Web information seeking and retrieval context.

4.4.6 Summary

Section 4.4 Multitasking during Web Search reported the results of multitasking

featured Web searching behaviour, based on the analysis of study participants'

multiple information problems including the evolving information problems

searching process.

Multitasking Web searching behaviour was characterised as a two-dimensional

activity. The first dimension referred to the multiple information problems

searching by task switching. The second dimension referred to the multiple

search sessions during Web searching.

The analysis revealed the factors affecting study participants’ information

problem search ordering, including problem importance level, randomness, ease

of finding information, task logics, problem urgency level, task interest, problem

knowledge and future usefulness. The results confirm and extend the findings of

Spink, Park & Koshman (2006), who suggested that task interest and problem

knowledge were the major factors determining search order on assigned

information problems searching.

152

Evolving information problems were explored in detail in study participants’

successive Web search processes. Compared to the original information

problems, the evolving information problems were represented as changed

problems or totally new ones.

Study participants switched their information problem searching tasks between

original information problem searching, evolving information problem searching

and serendipity browsing on other topics. Four task switching patterns were

identified and all the forty-two study participants’ task switching sequences were

classified into the patterns.

Further, a Web search session was investigated within a particular information

problem searching process. Results showed that multiple queries instead of a

single query were submitted, multiple Web search systems instead of a single

Web search system were employed, and such Web search interactions were

conducted in the multiple opened windows/tabs.

Next Section 4.5 reports the results of the cognitive shifts which occurred during

the Web searching processes.

153

4.5 Cognitive Shifts during Web Searching

A Web search interaction was also the situation during which study participants

experienced various cognitive shifts. Study participants engaged in two

categories of cognitive shifts: holistic cognitive shifts and cognitive state shifts.

Holistic shifts were defined as the changes on study participants’ perception over

an information problem before and after a Web search. Forms of holistic shifts

were identified via the analysis of the pre- and post-Web search questionnaires.

Each study participant was asked to indicate the shifts on the 5-point Likert scale

before and at the end of Web searching.

State shifts were the changes between study participants’ cognitive states during

the searching interactions. Search logs analysis provided the details of study

participants’ cognitive states and the changes between the states.

154

155

4.5.1 Holistic Cognitive Shifts

Holistic shifts were the users’ cognitive changes on the information problems

which were measured before and after Web searching, including shifts in

information problem understanding, information problem stage, information

seeking stage, shifts in personal knowledge, and the perception of the

contribution to the information problem resolution due to the Web search

interactions.

A 5-point Likert scale was utilised to quantify a cognitive shift on the information

problem understanding prior to and after Web searching. The number “1” stands

for no changes occurred, and the number “5” stands for the significant changes

occurred. Users’ judgment of contribution to the information problem resolution

was also quantified using a 5-point Likert scale: the number “1” stands for no

contribution, and the number “5” represents the significant contributions were

made. All the holistic shifts related data were collected via the pre- and post-Web

search questionnaires.

Table 4-24 shows the holistic cognitive shifts that forty-two study participants

experienced due to the Web search interactions.

Table 4-24. Holistic cognitive shifts data

Study Participant

Judgment of Information Problem Understanding (1-5)

Shift in Information Problem Stage

Shift in Information Seeking Stage

Shift in Personal Knowledge

Judgment of Contribution to the Information Problem Resolution (1-5)

OIP1 4 2 stages+ Forward/specific 2 stages+ 5 OIP2 4 2 stages+ Forward/specific 4 stages+ 5

1

OIP3 4 2 stages+ Forward/specific 4 stages+ 5 OIP1 1 Same Same 3 stages+ 4 OIP2 3 2 stages- Backward/broad 1 stage- 3

2

OIP3 1 3 stages+ Forward/specific 4 stages+ 5 OIP1 2 1 stage+ Backward/broad 1 stage+ 2 OIP2 3 1 stage+ Backward/broad Same 4

3

OIP3 5 3 stages+ Forward/specific 4 stages+ 5 OIP1 3 1 stage+ Same 1 stage+ 3 OIP2 1 1 stage+ Backward/broad Same 1

4

OIP3 5 3 stages+ Multi-stage/forward 4 stages+ 5 OIP1 5 1 stage+ Forward/specific 1 stage+ 5 OIP2 3 1 stage+ Forward/specific Same 3

5

OIP3 4 1 stage+ Forward/specific 3 stages+ 4 OIP1 1 Same Same Same 2 OIP2 1 Same Same Same 3

6

OIP3 2 Same Same Same 3 OIP1 4 Same Backward/broad 2 stages+ 3 OIP2 3 Same Forward/specific 1 stage+ 3

7

OIP3 2 Same Forward/specific 1 stage+ 3 OIP1 4 Same Same Same 4 OIP2 4 Same Backward/broad 1 stage+ 3

8

OIP3 2 2 stages+ Forward/specific 2 stages+ 5 OIP1 4 1 stage+ Forward/specific 2 stages+ 4 OIP2 3 1 stage+ Backward/broad 2 stages+ 2

9

OIP3 4 Same Forward/specific 3 stages+ 4

156

Table 4-24. (Continued)

Study Participant

Judgment of Information Problem Understanding (1-5)

Shift in Information Problem Stage

Shift in Information Seeking Stage

Shift in Personal Knowledge

Judgment of Contribution to the Information Problem Resolution (1-5)

OIP1 1 1 stage+ Multi-stage/forward 3 stages+ 1 OIP2 1 2 stages+ Multi-stage/backward 1 stage+ 1

10

OIP3 1 1 stage+ Multi-stage/backward Same 1 OIP1 4 1 stage+ Forward/specific 2 stages+ 4 OIP2 3 2 stages+ Backward/broad 2 stages+ 3

11

OIP3 3 1 stage+ Backward/broad 2 stages+ 3 OIP1 4 1 stage+ Same 1 stage+ 4 OIP2 5 Same Backward/broad 1 stage+ 4

12

OIP3 3 1 stage+ Forward/specific 2 stages+ 3 OIP1 3 1 stage+ Backward/broad 1 stage- 4 OIP2 4 Same Forward/specific 1 stage+ 3

13

OIP3 4 Same Backward/broad 2 stages+ 4 OIP1 1 1 stage- Backward/broad Same 4 OIP2 1 1 stage- Same Same 4

14

OIP3 1 1 stage+ Forward/specific 1 stage+ 5 OIP1 1 1 stage+ Multi-stage/forward Same 5 OIP2 1 1 stage+ Multi-stage/forward 1 stage- 5

15

OIP3 2 1 stage+ Multi-stage/forward 1 stage- 5 OIP1 4 Same Same Same 4 OIP2 4 3 stages+ Backward/broad 1 stage+ 5

16

OIP3 5 Same Forward/specific 1 stage+ 4 OIP1 1 Same Backward/broad 1 stage+ 4 OIP2 4 Same Forward/specific 1 stage+ 4

17

OIP3 3 Same Forward/specific 1 stage+ 3 OIP1 3 Same Backward/broad Same 3 OIP2 4 1 stage+ Backward/broad 1 stage+ 4

18

OIP3 3 Same Backward/broad 1 stage+ 4

157

Table 4-24. (Continued)

Study Participant

Judgment of Information Problem Understanding (1-5)

Shift in Information Problem Stage

Shift in Information Seeking Stage

Shift in Personal Knowledge

Judgment of Contribution to the Information Problem Resolution (1-5)

OIP1 2 Same Backward/broad 1 stage- 5 OIP2 3 2 stages+ Forward/specific Same 5

19

OIP3 4 1 stage+ Backward/broad 1 stage+ 5 OIP1 3 2 stages+ Backward/broad Same 3 OIP2 4 1 stage+ Forward/specific Same 4

20

OIP3 2 1 stage- Same 1 stage- 2 OIP1 4 Same Backward/broad 2 stages+ 4 OIP2 5 Same Same 1 stage+ 5

21

OIP3 4 1 stage+ Backward/broad Same 4 OIP1 2 Same Same 1 stage- 2 OIP2 4 Same Same Same 4

22

OIP3 3 Same Same 2 stages+ 3 OIP1 3 Same Backward/broad 1 stage+ 3 OIP2 3 1 stage+ Forward/specific Same 4

23

OIP3 2 1 stage- Same 1 stage- 2 OIP1 2 1 stage+ Multi-stage 1 stage+ 5 OIP2 3 1 stage+ Multi-stage Same 5

24

OIP3 3 1 stage+ Multi-stage Same 5 OIP1 1 Same Same Same 4 OIP2 1 Same Same Same 4

25

OIP3 1 Same Forward/specific Same 5 OIP1 4 Same Forward/specific Same 3 OIP2 3 Same Backward/broad 1 stage+ 4

26

OIP3 3 1 stage+ Backward/broad 2 stages+ 4 OIP1 4 Same Backward/broad 1 stage+ 4 OIP2 4 1 stage+ Backward/broad 2 stages+ 4

27

OIP3 3 2 stages+ Forward/specific 3 stages+ 5

158

Table 4-24. (Continued)

Study Participant

Judgment of Information Problem Understanding (1-5)

Shift in Information Problem Stage

Shift in Information Seeking Stage

Shift in Personal Knowledge

Judgment of Contribution to the Information Problem Resolution (1-5)

OIP1 4 2 stages+ Backward/broad Same 5 OIP2 5 1 stage+ Backward/broad 2 stages- 4

28

OIP3 3 Same Backward/broad 4 stages- 2 OIP1 4 1 stage+ Backward/broad 1 stage+ 4 OIP2 3 1 stage+ Backward/broad 1 stage+ 3

29

OIP3 4 1 stage+ Forward/specific 2 stages+ 5 OIP1 4 1 stage+ Backward/broad 1 stage- 4 OIP2 2 Same Same 3 stages+ 3

30

OIP3 2 1 stage+ Same 2 stages+ 5 OIP1 5 1 stage+ Same 4 stages+ 5 OIP2 4 1 stage- Backward/broad 1 stage+ 4

31

OIP3 4 1 stage+ Forward/specific 3 stages+ 3 OIP1 1 Same Backward/broad Same 4 OIP2 1 Same Same Same 4

32

OIP3 3 1 stage+ Backward/broad Same 3 OIP1 4 Same Forward/specific 1 stage+ 4 OIP2 4 Same Forward/specific Same 4

33

OIP3 1 1 stage+ Forward/specific Same 5 OIP1 4 1 stage+ Backward/broad Same 5 OIP2 3 2 stages+ Forward/specific 3 stages+ 5

34

OIP3 5 1 stage+ Forward/specific Same 5 OIP1 2 Same Forward/specific 1 stage+ 5 OIP2 4 2 stages+ Forward/specific 2 stages+ 4

35

OIP3 2 1 stage+ Forward/specific 2 stages+ 5 OIP1 2 1 stage- Forward/specific Same 4 OIP2 3 Same Forward/specific 2 stages+ 4

36

OIP3 2 1 stage- Forward/specific 1 stage+ 4

159

160

Table 4-24. (Continued)

Study Participant

Judgment of Information Problem Understanding (1-5)

Shift in Information

Problem Stage

Shift in Information Seeking Stage

Shift in Personal

Knowledge

Judgment of Contribution to the Information Problem Resolution

(1-5) OIP1 1 1 stage- Forward/specific 1 stage+ 1 OIP2 1 2 stages+ Forward/specific 1 stage+ 1

37

OIP3 2 Same Forward/specific 2 stages+ 3 OIP1 4 1 stage+ Backward/broad Same 4 OIP2 4 2 stages- Backward/broad 1 stage+ 4

38

OIP3 5 1 stage+ Forward/specific 1 stage+ 5 OIP1 2 1 stage+ Forward/specific 4 stages+ 4 OIP2 2 1 stage+ Forward/specific 1 stage+ 4

39

OIP3 2 Same Forward/specific 3 stages+ 4 OIP1 3 3 stages+ Same Same 4 OIP2 1 Same Backward/broad Same 4

40

OIP3 2 Same Forward/specific 1 stage+ 3 OIP1 3 Same Backward/broad 1 stage+ 3 OIP2 4 3 stages+ Forward/specific 1 stage+ 5

41

OIP3 4 Same Backward/broad Same 5 OIP1 2 1 stage- Multi-stage/backward Same 2 OIP2 4 Same Multi-stage/forward 1 stage+ 4

42

OIP3 5 Same Multi-stage/forward 1 stage+ 3

1). Shifts in the Information Problem Understanding Table 4-24 shows that different study participants experienced different levels of

change in terms of information problem understanding due to their Web search

interactions. Even for the same study participant, the understandings of three

information problems were different as well.

Examples:

• Study Participant 1 chose number 4 to indicate significant changes

occurred in each information problem understanding after searching on

the Web.

• Study Participant 3 considered number 2 to indicate a few changes

happened in the problem 1 understanding, number 3 to indicate that the

changes did take place in the problem 2 understanding, and number 5

was chosen to show the great changes that occurred in problem 3.

2). Shifts in the Information Problem Stage Table 4-24 indicates that different study participants experienced different levels

of shifts in their information problem stage due to the Web search interactions.

• Around 39% of the study participants stayed in the same information

problem stage, measured before and after Web searching.

• Over 50% of the study participants experienced positive shifts in the

information problem stage, measured in the beginning and at the end of the

Web search.

o Almost 37% of the study participants shifted one information problem

stage.

161

o Around 10% of the study participants shifted two information

problem stages.

o Near 5% of the study participants shifted three information problem

stages.

• The remaining 10% of the study participants experienced negative shifts in

the information problem stage.

o Over 7% of the study participants shifted to one previous information

problem stage.

o Less than 2% of the study participants shifted to two previous

information problem stages.

3). Shifts in the Information Seeking Stage Table 4-24 shows that different study participants experienced different levels of

shifts in their information seeking stage due to the Web search interactions. For

the same study participant, the levels of shifts in the information seeking stage

over three information problems were different as well.

• Nearly 38% of the study participants shifted to a forward/specific seeking

stage. (For example, Study Participants 1, 35 and 36)

• Around 34% of the study participants shifted to a backward/broad seeking

stage. (For example, Study Participants 18 and 26)

• Nearly 17% of the study participants remained in the same information

seeking stage. (For example, Study Participants 6 and 22)

• Over 10% of the study participants situated at multiple seeking stages. (For

example, Study Participants 10, 15 and 24)

162

4). Shifts in Personal Knowledge

Table 4-24 indicates that different study participants experienced different levels of

shifts in their personal knowledge on each information problem due to the Web

search interactions.

• More than 60% of the study participants reported that the searching

interactions were good for their personal knowledge.

o Over 32% of the study participants shifted one personal knowledge

stage.

o Around 16% of the study participants shifted two personal

knowledge stages.

o Over 7% of the study participants shifted three personal knowledge

stages.

o Over 6% of the study participants shifted four personal knowledge

stages.

• Near by 31% of the study participants stayed in the same personal

knowledge stage.

• Over 7% of the study participants shifted to one previous personal

knowledge stage.

5). Judgments on Contribution to the Information Problem Resolution Table 4-24 also indicates the study participants’ judgments on the contribution of

Web search interactions to the information problem resolution. Most of them (67%)

selected number 4 or 5 on the 5-point Likert scale, reporting that the searching

interactions significantly contributed to the resolution of their information problems.

163

In summary, study participants reported levels of holistic shifts: forward, backward,

and no shift, with respect to the information problems understanding and

knowledge contribution. In most situations, they experienced forward shifting.

Namely, Web search interactions had a positive impact on their information problem

solving. After Web searching, study participants got a better understanding of the

information problems, positive changes occurred at the information problem stage,

seeking stage and personal knowledge, and thus the contributions to the

information problems resolution were achieved.

4.5.2 Cognitive State Shifts

State shifts were the cognitive changes in focus of the interactions between a study

participant and a Web search system with respect to the study participant’s

cognitive states. It reflects how study participants moved between different

cognitive states, for instance, from the state of evaluation on the search results to

the state of query reformulation.

Cognitive state shifts were recognised through the “cue words” which indicated a

change. These cue words varied from the subtle to the obvious, and therefore,

utterance-search segments were studied carefully in order to perceive such cues.

164

4.5.2.1 Types of Cognitive State Five types of cognitive states were identified in the Web searching context via

analyzing the transcribed utterance-search segments.

1). Topic (TOP): the state focusing on the specific subject area/problem guiding the

search.

• Cue words: “problem”, “topic”, “look for”, “find”, “next”, etc.

2). Strategy (STR): the state concerning with the search strategies adoption and

adjustment, for example, term selection, query (re)formulation, Web search system

selection.

• Cue words: “query”, “change”, “Google”, “next page”, “save”, “make notes”,

etc.

3). Evaluation (EVA): the state focusing on the judgments regarding the relevance,

magnitude, etc. of the system output (the list of results), positively or negatively.

• Cue words: “(not) good”, “(ir)relevant”, “(ir)related”, “yes/no”, “(not) suitable”,

“(not) satisfied”, “too much/little”, etc.

4). View (VIE): the state focusing on viewing and examining a specific opened URL

link against the aim of the information problem, making sense, such as availability,

format, useful or not.

• Cue words: “full text”, “access”, “available”, “pdf document”, “easy”, “idea”,

“useful”, “good”, “what I want/ am looking for”, etc.

5). Overview (OVE): the state focusing on the overall search outcome, and time

allocation among the information problems.

• Cue words: “enough”, “finish”, “solve”, “done”, “time”, “stop”, “move on”, etc.

165

Table 4-25 shows the study participants’ cognitive state data during Web searching.

Table 4-25. Cognitive state data Note: TOP (Topic) STR (Strategy) EVA (Evaluation) VIE (View) OVE (Overview)

Type and Number of Cognitive State Occurrences Study Participant TOP EVA STR VIE OVE Total 1 3 25 38 17 2 85 2 3 16 34 4 2 59 3 3 36 59 23 3 124 4 4 61 65 48 6 184 5 14 22 66 39 3 144 6 5 18 27 16 4 70 7 3 44 72 33 4 156 8 6 36 53 27 4 126 9 11 23 58 38 6 136 10 6 16 42 27 3 94 11 6 6 31 26 2 71 12 5 30 68 43 3 149 13 9 22 58 28 2 119 14 3 68 50 29 5 155 15 7 14 64 40 3 128 16 7 20 54 44 3 133 17 10 60 64 47 6 187 18 13 60 72 36 4 185 19 4 35 45 40 7 131 20 8 51 43 28 2 132 21 6 30 81 50 2 169 22 16 44 44 48 6 158 23 13 24 52 48 3 140 24 5 19 40 30 3 97 25 3 26 25 28 2 84 26 9 25 39 33 3 109 27 7 26 30 51 3 117 28 8 48 59 56 3 174 29 6 16 27 24 3 76 30 6 7 23 11 3 50 31 6 48 64 49 3 170 32 5 65 59 56 3 188 33 3 52 63 32 3 153 34 8 28 24 31 3 94 35 3 11 19 4 3 40 36 9 27 38 23 3 100 37 5 58 76 17 3 159 38 12 31 66 37 4 150 39 3 51 58 23 3 138 40 3 26 25 37 3 94 41 8 19 39 16 3 85 42 7 22 39 11 3 82 Total 281 1366 2053 1348 142 5159 Mean 7 33 49 32 3 124 % 5 26 40 26 3 100

166

• The total number of cognitive states that study participants experienced was

5159, with a mean of 124.

• The most experienced cognitive states were strategy (40%), evaluation

(26%), and view (26%), respectively. The three states formed more than

90% of the total cognitive states during the searching interactions.

• In other words, study participants paid more attention to the search

strategies adoption and adjustment, and the system output and specific

search results evaluation.

4.5.2.2 Shifts of Cognitive State Study participants not only experienced diverse cognitive states, but also moved

between the states. This section reports how study participants shifted from one

cognitive state to another. Table 4-26 shows the shifts between the five types of

cognitive states.

167

Table 4-26. Shifts between cognitive states Note: TOP (Topic) STR (Strategy) EVA (Evaluation) VIE (View) OVE (Overview) Types of Cognitive State Shift (From —> To)

Frequency of Occurrence

TOP to TOP 1 TOP to EVA 0 TOP to STR 275 TOP to VIE 4 TOP to OVE 0 EVA to TOP 30 EVA to EVA 330 EVA to STR 447 EVA to VIE 539 EVA to OVE 24 STR to TOP 132 STR to EVA 835 STR to STR 695 STR to VIE 317 STR to OVE 55 VIE to TOP 25 VIE to EVA 201 VIE to STR 573 VIE to VIE 483 VIE to OVE 65 OVE to TOP 65 OVE to EVA 1 OVE to STR 31 OVE to VIE 4 OVE to OVE 1 The cognitive state shifts data show that the main shifts took place from evaluation

(EVA) to another type of cognitive state, between strategy (STR) and another type

of cognitive state, and from the state of view (VIE) followed by another type of

cognitive state. Table 4-27 sorts out the frequency of cognitive state shifting

occurrence, from the highest to the lowest.

168

Table 4-27.Summary of cognitive state shifts occurrence Note: TOP (Topic) STR (Strategy) EVA (Evaluation) VIE (View) OVE (Overview) Types of Cognitive State Shift (From —> To) Frequency of Occurrence STR to EVA 835 STR to STR 695 VIE to STR 573 EVA to VIE 539 VIE to VIE 483 EVA to STR 447 EVA to EVA 330 STR to VIE 317 TOP to STR 275 VIE to EVA 201 STR to TOP 132 OVE to TOP 65 VIE to OVE 65 STR to OVE 55 OVE to STR 31 EVA to TOP 30 VIE to TOP 25 EVA to OVE 24 OVE to VIE 4 TOP to VIE 4 OVE to EVA 1 OVE to OVE 1 TOP to TOP 1 TOP to EVA 0 TOP to OVE 0

• Cognitive state was verified to be constantly transformable and the shifts

occurred frequently.

• Most shifts occurred between the states of strategy (STR), evaluation (EVA),

and view (VIE), including STR to EVA, STR to STR, VIE to STR, EVA to VIE,

VIE to VIE, EVA to STR, EVA to EVA, STR to VIE, and VIE to EVA.

• The findings demonstrate that study participants recurrently divert their

attention/cognitive state from the adoption of the search strategy to the

following evaluation of the search results, then to the examination of the

opened URL link.

169

• Fewer shifts took place between the state of topic (TOP) /overview (OVE)

and the other types of states. There were no shifts directly from topic (TOP)

to evaluation (EVA), or from topic (TOP) to overview (OVE).

• However, TOP to STR was verified as a frequent type of shift. That means

study participants regularly moved from the state of the searching topic to

the following state of search strategy consideration.

4.5.3 Summary

This section presented the results of cognitive shifts during Web searching,

including holistic cognitive shifts and cognitive state shifts. It provided a

comprehensive understanding of the cognitive shifts due to the searching

interactions.

The results confirmed and extend Spink’s (2002) findings, which were limited to the

shifts before and after the Web search, namely, the holistic shifts level. The results

also extend Robins’ (2000) findings which were focused only on the shifts occurring

within information problems searching.

The analysis of cognitive states and the shifts between the states may deepen our

understanding of the cognitive and interactive Web searching behaviour.

170

4.6 Cognitive Coordination during Web Search

The analysis unit was a cognitive coordination occurrence. The behaviour of

cognitive coordination occurrence was embodied within the iterative searching

process, including:

1) an information problem identification,

2) search terms (re)selection and queries (re)formulation,

3) system outputs in response to the search strategies,

4) the study participant’s utterances including judgment feedback and sense-

making process relating to the systems output, and

5) the study participant’s further actions relating to the relevant items retrieved.

Thinking aloud audio data (utterances), along with the search logs, were

concurrently recorded by the software Camtasia Studio. Each study participant was

required to verbalise their thoughts, actions and reasons at the same time as

conducting the Web search. Each study participant’s utterances related to the Web

searching activities were transcribed onto the section of their search logs which

recorded those activities.

Cognitive coordination occurrences were identified through the analysis of think-

aloud transcripts and Web search logs, that is, the verbal protocol analysis of the

utterance-search segments. The identified cognitive coordination occurrences were

then classified into different levels. The following sections present the results of

cognitive coordination behaviour, including three cognitive coordination levels.

171

4.6.1 Cognitive Coordination Results Overview

Table 4-28 provides the overview results of cognitive coordination across forty-two

study participants’ Web searches.

172

Table 4-28. Overall results

Information Task Number

Study Participant SOIP SEIP SB TotalWeb Search

System Query

1 3 0 0 3 1 17 2 3 0 0 3 2 12 3 3 1 0 4 6 26 4 3 0 0 3 3 20 5 3 8 1 12 6 27 6 3 0 0 3 2 14 7 3 0 0 3 2 26 8 3 2 2 7 1 18 9 3 5 0 8 4 21 10 3 2 0 5 4 18 11 3 3 0 6 4 11 12 3 1 0 4 2 18 13 3 4 0 7 6 22 14 3 0 0 3 1 12 15 3 5 0 8 4 19 16 3 3 0 6 5 22 17 3 3 1 7 5 23 18 3 5 0 8 1 39 19 3 0 0 3 2 4 20 3 1 0 4 2 15 21 3 2 0 5 6 38 22 3 4 0 7 6 12 23 3 5 0 8 3 26 24 3 1 0 4 2 15 25 3 0 0 3 1 7 26 3 5 0 8 4 18 27 3 3 0 6 4 9 28 3 4 0 7 4 14 29 3 3 0 6 7 8 30 3 2 0 5 6 8 31 3 2 0 5 7 31 32 3 1 0 4 2 11 33 3 0 0 3 4 20 34 3 4 0 7 1 13 35 3 0 0 3 1 9 36 3 4 0 7 4 14 37 3 1 0 4 3 22 38 3 6 1 10 10 21 39 3 0 0 3 3 19 40 3 0 0 3 4 7 41 3 1 3 7 4 21 42 3 4 0 7 3 15

173

• When searching for information on the Web, study participants did basic

cognitive coordination activities involving original information problems

searching (SOIP), evolving information problems searching (SEIP) and

serendipity browsing (SB), and Web search systems selection and queries

(re) formulation as well.

• During the period of the one-hour Web searching episode, the number of

coordinated information searching tasks ranged from three to twelve, the

number of employed Web search systems ranged from one to ten, and the

number of submitted queries varied from four to thirty-nine.

The following sections 4.6.2—4.6.4 discuss the three cognitive coordination levels

identified in this study.

4.6.2 Level One: Information Task Coordination (TC)

Information task level coordination was represented as the coordination process

between the information tasks (IT), including original information problem

identification, evolving information problem generation, problem searching task

switching, and windows/tabs browsing.

The explanation was endowed within the following example. Example: Study Participant 10

Information Task Coordination Number OIP1 Identification 1 OIP2 Identification 1 OIP3 Identification + Searching Task Switching (SEIP2 —> SOIP3) 2 EIP1 Generation 1 EIP2 Generation 1 Browsed windows/tabs 5 Total 11

174

Study Participant 10’s information task coordination process consisted of:

• the identification of the first original information problem (OIP1) once,

• the identification of the second original information problem (OIP2) once,

• the identification of the third original information problem (OIP3) once,

• the generation of the first evolving information problem (EIP1) once,

• the generation of the second evolving information problem (EIP2) once,

• the searching task switching from SEIP2 to SOIP3 once, and

• browsing information in five opened windows/tabs.

In total, eleven times of the information task coordination occurred during the

course of Web searching.

4.6.3 Level Two: Cognitive Coordination Mechanism (CM)

The cognitive coordination mechanism was the underlying system supporting

information task coordination; and it also explained how the cognitive coordination

among multiple information tasks was achieved.

During Web searching, the coordination mechanism involved a series of cognitive

processing activities, stated as follows:

– content relevance feedback (CRF) in which relevance judgments were

made on the returned results,

– magnitude feedback (MF) in which judgments were made in terms of the

size of output,

– self-learning and regulating (SLR) process within which the gathered

information was examined,

175

– tactical review feedback (TCF) in which strategy-related adjustments were

made based on the retrieved results,

– term relevance feedback (TRF) consisting of users identifying a term (terms)

within the retrieved results.

Details of each cognitive processing activity will be discussed in Section 4.6.5

Cognitive Coordination Behaviour on Three Levels. The following example

describes the occurrence of the cognitive coordination mechanism.

Example: Study Participant 10

Cognitive Coordination Mechanism No. Content

Relevance Feedback

Magnitude Feedback

Self-learning and

Regulating

Tactical Review

Feedback

Term Relevance Feedback

SOIP1 1 0 10 1 1 SOIP2 5 1 8 1 1 SOIP3 4 1 7 2 0 SEIP1 0 1 1 0 0 SEIP2 0 0 1 0 1 Total 10 3 27 4 3

Study Participant 10’s coordination mechanism consisted of:

ten content relevance feedback — relevance judgments were made once,

five times and four times during the process of SOIP1, SOIP2, and SOIP3

searching respectively. No content relevance judgment was made during

the searching of SEIP1 or SEIP2;

three magnitude feedback — magnitude judgments over the size of system

output were made once during the searching course of SOIP2, SOIP3, and

SEIP1 respectively. No magnitude feedback occurred within SOIP1 or

SEIP2 searching;

176

twenty-seven self-learning and regulating processes — Study Participant

10 made sense of the gathered information during each information

problem searching;

four tactical review feedback — strategy-related adjustments were made

once during the searching process of SOIP1, once during SOIP2 and twice

during SOIP3 searching process. No strategy adjustment was made during

SEIP1 and SEIP2 searching;

three term relevance feedback — a search term was identified within the

retrieved results during the problem searching task of SOIP1, SOIP2, and

SEIP2 respectively.

4.6.4 Level Three: Cognitive Strategy Coordination (SC)

Strategy coordination was considered as a strategic plan for solving information

problems within the resources available. The available resources in this study

encompassed the usable Web searching tools as well as the limited timeframe of

the one-hour searching duration.

Cognitive strategy comprised problem specific strategy (PSS) and global strategy

(GS).

– Problem specific strategy (PSS) was viewed as the collection of tactics on

Web searching tools that study participants adopted for particular

information problem solving.

– The tactics referred to Web search systems selection, search queries

(re)formulation, results set (pages) review, and the relevant results saving

as well.

177

– Problem specific strategies were in the form of selecting a Web search

system, submitting a query, reformulating search query or just browsing

more results by clicking on the “next” link to next pages, and saving the

results either to the hard drive, flash disk or to the Web browser

favourite/bookmark, or by making notes of the Website URL, or by copying

and pasting the Website content to Word application document.

– Global strategy (GS) was an overall plan for the whole searching process,

which was presented as study participants’ decisions on the allocation of the

one-hour searching time between multiple information problems searching,

in order to solve all of them within the limited timeframe.

Table 4-29 shows study participants’ Web searching duration and time allocation

for each information problem searching.

178

Table 4-29. Time allocation between multiple information problems searching (Global Strategy)

Time Allocation Study Participant SOIP1 SOIP2 SOIP3 SEIP Total Duration1 9’53’’ 12’27’’ 2’59’’ 0 25’19’’ 2 11’40’’ 17’25’’ 6’17’’ 0 35’22’’ 3 27’19’’ 23’00’’ 8’45’’ 0’56’’ 59’60’’ 4 28’45’’ 26’49’’ 4’26’’ 0 59’60’’ 5 7’37’’ 6’27’’ 24’55” 12’16’’ 51’15’’ 6 14’27’’ 8’37’’ 6’26’’ 0 29’30’’ 7 15’16’’ 30’01’’ 12’17’’ 0 57’34’’ 8 15’30’’ 12’40’’ 7’32’’ 16’09’’ 51’51’’ 9 5’12’’ 21’10’’ 0’26’’ 27’25’’ 54’13’’ 10 23’55’’ 14’45’’ 18’14’’ 2’09’’ 58’53’’ 11 2’52’’ 10’38’’ 8’47’’ 21’03’’ 43’20’’ 12 7’40’’ 11’36’’ 39’15’’ 1’18’’ 59’49’’ 13 26’56’’ 5’25’’ 16’35’’ 9’34’’ 58’30’’ 14 13’44’’ 4’27’’ 22’30’’ 0 40’41’’ 15 26’40’’ 9’32’’ 0’56’’ 12’39’’ 49’47’’ 16 10’22’’ 4’51’’ 21’31’’ 4’47’’ 41’31’’ 17 18’14’’ 10’29’’ 16’43’’ 16’49’’ 1h02’15’’ 18 17’21’’ 19’53’’ 7’54’’ 17’53’’ 1h01’51’’ 19 30’22’’ 18’49’’ 7’49’’ 0 57’00’’ 20 21’07’’ 8’46’’ 6’49’’ 5’30’’ 43’00’’ 21 19’49’’ 6’59’’ 30’54’’ 2’21’’ 1h0’03’’ 22 18’04’’ 18’14’’ 12’58’’ 10’44’’ 59’60’’ 23 13’23’’ 14’11’’ 11’23’’ 17’01’’ 50’58’’ 24 12’30’’ 5’05’’ 8’50’’ 1’51’’ 28’16’’ 25 32’49’’ 14’03’’ 4’54’’ 0 51’46’’ 26 17’01’’ 16’06’’ 7’41’’ 6’73’’ 48’01’’ 27 19’13’’ 21’52’’ 2’36’’ 1’59’’ 45’40’’ 28 25’13’’ 6’36’’ 13’25’’ 14’48’’ 1h02’52’’ 29 5’32’’ 3’41’’ 12’49’’ 5’54’’ 27’56’’ 30 7’56’’ 4’22’’ 7’25’’ 3’58’’ 23’31’’ 31 8’53’’ 18’22’’ 31’27’’ 2’43’’ 1h01’25’’ 32 28’31’’ 12’09’’ 16’58’’ 0’29’’ 58’27’’ 33 26’32’’ 22’57’’ 10’26’’ 0 59’55’’ 34 7’12’’ 10’50’’ 6’17’’ 18’48’’ 43’07’’ 35 3’30’’ 7’50’’ 1’38’’ 0 12’58’’ 36 17’46’’ 14’56’’ 4’51’’ 5’16’’ 42’49’’ 37 30’46’’ 8’29’’ 8’30’’ 5’30’’ 53’15’’ 38 13’18’’ 2’52’’ 8’08’’ 27’32’’ 51’50’’ 39 23’37’’ 5’30’’ 32’10’’ 0 1h01’17’’ 40 21’48’’ 19’38’’ 10’43’’ 0 52’09’’ 41 21’32’’ 2’51’’ 15’32’’ 2’31’’ 42’26’’ 42 11’36’’ 6’53’’ 2’30’’ 13’38’’ 34’37’’

179

The column of SOIP1 stands for the time allocation on the first original

information problem searching, the column of SOIP2 stands for the time

allocation on the second original information problem searching, the column

of SOIP3 stands for the time allocation on the third original information

problem searching, and the column of SEIP stands for the time allocation

on the evolving information problems searching. The number of evolving

information problem ranged from zero to eight.

In terms of searching duration, most of the study participants (nearly 40%)

used up the one-hour searching time, 26% of the study participants utilised

over three-fourths of the one-hour searching time, another 21% of the study

participants utilised around two-thirds of the one-hour searching time, and

the remaining 14% only used less than half an hour.

Table 4-29 also shows that, under most circumstances, the searching

duration was unevenly allocated between multiple information problems. In

the cases of Study Participants 3, 4, 5, 9, 12, 13, 15, 19, 21, 25, 31, 37 and

39, almost half an hour was given to one information problem searching

and only a few minutes to another. Only in four cases of Study Participants

22, 23, 30 and 35, searching time was allocated nearly evenly.

Park (2008) found the subjects were well aware of spending time in an

optimal way among multiple information tasks. Coordinating activities in

Web seeking and retrieval context entailed task switching, tabbed browsing,

strategic search planning, and information evaluation, which were all

closely related to time management. The findings of this study also show

that time allocation was not a random behaviour but a conscious strategy.

Here are some examples gathered from post-Web search questionnaires:

180

I knew I would need to spend a lot of time searching on problem 3,

so I left it till last. Problem 1 was the most straight-forward out of the

three and I believed it would need the shortest time to find the

information. (Study Participant 28)

The first problem was quite related to my study, I thought it would

take long time and I needed to be focused. (Study Participant 26)

I left the most involved problem until last as I knew it would take the

longest time and if I didn’t fully complete it in the session, I would

have at least completed the other two. Problem 1 was easier, I

thought it would take less time; Problem 2 was more difficult which

would take longer time; and Problem 3 was a very broad problem

which must have taken the longest time. (Study Participant 12)

The following example describes the occurrence of cognitive strategy coordination,

including problem specific strategy and global strategy.

Example: Study Participant 10 Strategy Coordination Problem Specific Strategy Global Strategy SOIP1 7 0 SOIP2 13 1 SOIP3 14 1 SEIP1 2 1 SEIP2 2 0 Total 38 3

The occurrence of Study Participant 10’s strategy coordination process

consisted of 38 adopted problem specific strategies during the searching of

SOIP1 (7), SOIP2 (13), SOIP3 (14), SEIP1 (2) and SEIP2 (2), and 3

adopted global strategies on the SOIP2 (1), SOIP3 (1) and SEIP1 (1)

searching process.

181

The next section examines the interplay between the three cognitive coordination

levels from the perspectives of occurrences frequency on each level and the

transition between the three cognitive coordination levels.

4.6.5 Cognitive Coordination Behaviour on Three Levels

Study participants experienced the complex cognitive coordination processes

embedded within the problem-solving oriented Web search interactions. The

cognitive coordination behaviour was embodied as above three cognitive

coordination levels. Strategy coordination affected users’ coordination mechanism

that further supported their task coordination process. For example,

Study Participant 42 generated a new information problem (information task

coordination level) due to content relevance feedback (coordination

mechanism level) and strategy coordination level of selecting another Web

search engine.

This section presents the interplay between three cognitive coordination levels. At

first, the descriptions and examples of each cognitive coordination level are

elaborated as follows. The examples are excerpted from study participants’

utterance-search segments.

182

Level One: Task Coordination (TC)

Information Task (IT): original information problem identification, evolving

information problem generation, problem searching task switching, and

windows/tabs browsing.

Examples:

• The second problem I'm trying to find out the (medical) function of shark

cartilage. (original information problem identification, Study Participant 26)

• (I met a word "angiogenesis" in this article). What is angiogenesis? I need to

look it up in the dictionary before continuing the current problem searching.

(evolving information problem generation, Study Participant 26)

• Now move on to the third problem. (information problem searching task

switching, Study Participant 26)

• Click on the result link and open it in a new tab, then browse information on

the opened website. (tabs browsing, Study Participant 36)

Level Two: Coordination Mechanism (CM)

Term Relevance Feedback (TRF): a term (terms) was (were) identified within

the retrieved results, subsequently used to modify the search query.

Example:

• Oh, Google provides related searches to 'shark cartilage arthritis'. I will try

this (as search keywords). (Study Participant 26)

183

Content Relevance Feedback (CRF): a relevance judgment was made on a

system returned entry/result before the corresponding URL link was clicked on.

Example:

• Yes, here we go, this one is good. (followed by clicking on the 3rd entry link).

(Study Participant 26)

Magnitude Feedback (MF): a judgment was made based on the size of system

output, not necessarily followed by a modified/reformulated query.

Examples:

• No results, what happened? (followed by reformulating the search query).

(Study Participant 26)

• There are 34 related articles and 2 reviews turned up. That’s good.

(followed by viewing the returned results rather than reformulating the search

query). (Study Participant 26)

Tactical Review Feedback (TCF): strategy-related adjustments were made

based on the retrieved results, such as the decision on changing a search

keyword or changing information source.

Examples:

• I may add another keyword to make sure that I'll not miss any information.

(decision on change a search keyword, Study Participant 26)

• (It seems that I need to pay for it) I may (try) go to QUT library database to

find the full text. (decision on change information source, Study Participant 26)

184

Self-learning and Regulating (SLR): refers to users’ examination on the opened

Website/Webpage content followed by a sense-making process. SLR suggests

that study participants were in continuous states of knowledge constructing

based on new information gained through focusing on various aspects of

information problems.

Examples:

• This is not what I want. (after browsing the opened Website content, Study

Participant 26)

• It makes sense. The reason of using shark cartilage in the treatment of

arthritis is that it can help to build.... This is a high quality research, quite good.

So shark cartilage is useful to the treatment of arthritis. (after reading through

the pdf document, Study Participant 26)

Level Three: Strategy Coordination (SC)

Problem Specific Strategy (PSS): consists of the tactics on Web search

systems selection, search query (re)formulation, results set (pages) review, and

relevant results saving as well.

Examples:

• Hopefully I can find something in the academic database. But I cannot be

sure. I know it's a really narrow topic. (Web search systems selection, Study

Participant 28)

• The first problem I'm going to solve today is to find updated information

about instructions on 'distress thermometer' using in my PhD study. So I enter

185

'distress thermometer' as my search query. (search query formulation, Study

Participant 26)

• Click on the 'next' link at the end of Google results list to check information

shown on the next page. (next result page review, Study Participant 26)

• I'd like to save this pdf document to the desktop. (relevant results saving,

Study Participant 26)

Global Strategy (GS): refers to the tactics on searching duration and time

allocation between multiple information problem searching tasks.

Examples:

• Oh, time is up, ok, stop here. (searching duration, Study Participant 28)

• What time is now? 25 minutes left? So I still can keep looking for this

problem. (time allocation, Study Participant 22)

4.6.5.1 Frequency of Occurrences Table 4-30 shows a detailed picture of occurrence frequency at each cognitive

coordination level within each Web searching episode. As an example, Study

Participant 10‘s cognitive coordination occurrences at each level have been

described in the previous Sections 4.6.2—4.6.4.

186

Table 4-30. Cognitive coordination behaviour data

TC (Level 1)

CM (Level 2)

SC (Level 3)

No.

Study Participant IT TRF CRF MF TCF SLR PSS GS SOIP1 1 2 6 3 3 6 6 1 SOIP2 1 0 12 0 9 8 18 0 SOIP3 1 0 2 0 0 3 2 1

1

Sub-Total 3 2 20 3 12 17 26 2 SOIP1 1 0 3 2 1 0 4 0 SOIP2 1 0 1 9 8 4 18 1 SOIP3 1 0 0 1 0 0 3 1

2

Sub-Total 3 0 4 12 9 4 25 2 SOIP1 1 0 15 3 9 13 19 1 SOIP2 1 0 5 10 11 5 16 1 SOIP3 1 0 2 1 0 5 4 1 SEIP1 1 0 0 0 0 0 2 0

3

Sub-Total 4 0 22 14 20 23 41 3 SOIP1 1 0 20 7 8 21 23 2 SOIP2 2 0 28 4 6 21 25 3 SOIP3 1 0 3 0 2 6 5 1

4

Sub-Total 4 0 51 11 16 48 53 6 SOIP1 1 0 5 0 2 6 12 0 SOIP2 1 0 0 0 0 4 2 2 SOIP3 4 1 5 5 8 10 21 0 SEIP1 1 0 0 0 0 0 1 0 SEIP2 1 0 0 0 0 1 1 0 SEIP3 1 0 0 0 1 1 2 0 SEIP4 1 0 1 0 0 0 2 1 SEIP5 1 0 0 0 1 1 1 0 SEIP6 1 0 0 0 1 1 1 0 SEIP7 1 2 2 0 1 8 8 0 SEIP8 1 1 0 0 0 6 1 0

5

Sub-Total 14 4 13 5 14 38 52 3 SOIP1 2 0 4 1 2 8 8 1 SOIP2 2 0 8 1 3 3 8 2 SOIP3 1 0 2 2 1 5 5 1

6

Sub-Total 5 0 14 4 6 16 21 4 SOIP1 1 0 16 1 7 11 13 1 SOIP2 1 2 17 2 9 15 24 2 SOIP3 1 0 5 1 9 7 10 1

7

Sub-Total 3 2 38 4 25 33 47 4 SOIP1 1 0 12 2 4 6 12 0 SOIP2 2 2 8 1 5 6 10 0 SOIP3 1 0 2 2 1 4 5 2 SEIP1 1 0 1 0 0 2 2 0 SEIP2 1 0 6 0 5 5 7 1 SB1 0 0 0 0 0 1 1 0 SB2 0 0 0 0 0 1 1 1

8

Sub-Total 6 2 29 5 15 25 38 4

187

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 1 0 2 0 2 5 3 0 SOIP2 2 0 13 0 6 12 19 2 SOIP3 1 0 0 0 0 0 1 0 SEIP1 2 0 1 0 0 3 6 1 SEIP2 2 0 4 0 2 13 10 2 SEIP3 1 0 1 1 1 1 3 0 SEIP4 1 0 1 0 0 3 3 0 SEIP5 1 0 0 0 0 1 2 1

9

Sub-Total 11 0 22 1 11 38 47 6 SOIP1 1 1 1 0 1 10 7 0 SOIP2 1 1 5 1 1 8 13 1 SOIP3 2 0 4 1 2 7 14 1 SEIP1 1 0 0 1 0 1 2 1 SEIP2 1 1 0 0 0 1 2 0

10

Sub-Total 6 3 10 3 4 27 38 3 SOIP1 1 0 2 0 1 1 2 0 SOIP2 1 0 0 0 0 8 4 1 SOIP3 1 0 4 0 0 6 7 1 SEIP1 1 0 0 0 0 5 7 0 SEIP2 1 0 0 0 1 0 2 0 SEIP3 1 0 0 0 2 6 5 0

11

Sub-Total 6 0 6 0 4 26 27 2 SOIP1 2 0 0 0 1 4 7 1 SOIP2 1 0 7 0 2 7 12 1 SOIP3 1 0 22 0 7 32 36 1 SEIP1 1 0 1 0 1 0 2 0

12

Sub-Total 5 0 30 0 11 43 57 3 SOIP1 1 4 5 1 0 14 11 1 SOIP2 2 1 1 0 1 5 5 0 SOIP3 2 0 5 0 7 5 17 0 SEIP1 1 0 0 0 1 0 2 0 SEIP2 1 0 4 0 1 3 7 0 SEIP3 1 0 0 0 1 1 2 1 SEIP4 1 0 1 0 1 0 2 0

13

Sub-Total 9 5 16 1 12 28 46 2 SOIP1 1 0 15 1 3 16 9 3 SOIP2 1 0 6 0 1 3 5 1 SOIP3 1 0 46 0 4 10 28 1

14

Sub-Total 3 0 67 1 8 29 42 5

188

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 1 0 7 1 6 21 25 1 SOIP2 1 1 2 0 2 5 9 1 SOIP3 1 0 0 0 0 1 2 0 SEIP1 1 0 0 0 1 1 4 0 SEIP2 1 2 1 0 0 3 5 0 SEIP3 1 0 0 0 1 4 2 0 SEIP4 0 0 1 0 1 4 3 1 SEIP5 1 0 0 0 1 1 2 0

15

Sub-Total 7 3 11 1 12 40 52 3 SOIP1 2 0 8 3 8 4 13 1 SOIP2 1 0 2 0 2 8 5 0 SOIP3 1 1 5 0 5 28 16 1 SEIP1 1 0 0 0 0 0 2 0 SEIP2 1 0 0 0 0 0 2 0 SEIP3 1 0 1 0 0 4 1 1

16

Sub-Total 7 1 16 3 15 44 39 3 SOIP1 1 0 16 2 5 12 12 2 SOIP2 2 1 7 1 0 8 6 1 SOIP3 1 0 16 0 7 17 10 1 SEIP1 1 1 10 2 3 4 13 1 SEIP2 3 0 2 1 0 5 5 1 SEIP3 2 0 1 1 0 0 3 0 SB1 0 0 0 0 0 1 0 0

17

Sub-Total 10 2 52 7 15 47 49 6 SOIP1 2 2 9 0 7 12 12 1 SOIP2 2 4 16 0 11 8 15 1 SOIP3 3 1 13 0 5 4 6 1 SEIP1 1 1 0 0 1 1 2 0 SEIP2 1 0 0 0 1 1 1 0 SEIP3 1 1 8 0 3 3 3 0 SEIP4 2 0 2 0 1 4 3 0 SEIP5 1 0 3 0 0 3 1 1

18

Sub-Total 12 9 51 0 29 36 43 4 SOIP1 2 0 13 0 3 21 18 4 SOIP2 1 0 14 0 0 13 15 1 SOIP3 1 0 8 0 1 6 8 2

19

Sub-Total 4 0 35 0 4 40 41 7 SOIP1 3 0 24 6 4 13 16 1 SOIP2 2 0 10 0 3 7 7 0 SOIP3 2 0 4 2 1 5 9 0 SEIP1 1 0 5 0 0 3 3 1

20

Sub-Total 8 0 43 8 8 28 35 2

189

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 2 1 11 0 4 17 25 0 SOIP2 1 0 2 0 0 8 2 0 SOIP3 1 2 10 2 6 24 42 1 SEIP1 1 0 1 0 0 0 1 0 SEIP2 1 0

21

1 0 0 1 1 1 Sub-Total 6 3 25 2 10 50 71 2 SOIP1 7 0 15 1 3 13 16 3 SOIP2 2 0 6 1 3 14 6 2 SOIP3 1 0 8 0 0 16 5 0 SEIP1 3 0 7 3 2 3 5 0 SEIP2 1 0 1 0 0 0 1 0 SEIP3 1 0 1 0 0 1 1 0 SEIP4 1 0 1 0 0 1 2 1

22

Sub-Total 16 0 39 5 8 48 36 6 SOIP1 2 0 7 0 4 12 13 1 SOIP2 3 2 9 0 4 11 14 1 SOIP3 2 0 2 0 2 9 3 1 SEIP1 1 0 1 0 0 7 4 0 SEIP2 2 0 0 0 0 4 4 0 SEIP3 1 0 1 0 0 2 1 0 SEIP4 1 1 0 0 0 1 1 0 SEIP5 1 0 1 0 0 2 1 0

23

Sub-Total 13 3 21 0 10 48 41 3 SOIP1 2 0 9 0 3 11 14 2 SOIP2 1 0 2 0 0 6 7 0 SOIP3 1 0 5 1 2 11 11 1 SEIP1 1 0 2 0 1 2 2 0

24

Sub-Total 4 0 18 1 6 30 34 3 SOIP1 1 0 19 1 2 14 17 1 SOIP2 1 0 2 0 0 9 3 0 SOIP3 1 0 4 0 0 5 3 1

25

Sub-Total 3 0 25 1 2 28 23 2 SOIP1 1 0 5 6 4 11 11 1 SOIP2 1 1 5 0 1 8 5 0 SOIP3 1 0 2 0 2 5 3 1 SEIP1 1 0 1 0 0 0 2 0 SEIP2 1 0 0 0 1 1 1 0 SEIP3 2 0 0 0 0 2 4 0 SEIP4 1 0 1 0 0 0 1 0 SEIP5 1 0 2 0 2 5 3 1

26

Sub-Total 9 1 16 6 10 32 30 3

190

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 2 1 3 0 1 30 9 1 SOIP2 1 1 15 0 2 15 10 1 SOIP3 1 0 4 0 0 3 2 1 SEIP1 1 0 0 0 1 1 2 0 SEIP2 1 1 0 0 0 1 1 0 SEIP3 1 1 0 0 0 1 2 0

27

Sub-Total 7 4 22 0 4 51 26 3 SOIP1 1 3 15 0 3 29 18 1 SOIP2 1 0 2 0 1 7 4 0 SOIP3 2 1 14 1 6 8 15 1 SEIP1 1 1 2 0 1 2 1 0 SEIP2 1 0 3 0 1 5 1 0 SEIP3 1 1 4 0 0 3 5 1 SEIP4 1 0 0 1 1 2 2 0

28

Sub-Total 8 6 40 2 13 56 46 3 SOIP1 1 0 3 1 0 8 2 0 SOIP2 1 0 4 0 0 3 5 1 SOIP3 1 0 3 0 2 8 11 1 SEIP1 1 1 0 0 0 1 2 0 SEIP2 1 1 0 0 1 2 1 0 SEIP3 1 1 2 0 0 2 3 1

29

Sub-Total 6 3 12 1 3 24 24 3 SOIP1 2 0 3 0 1 4 6 1 SOIP2 1 0 0 0 0 2 2 1 SOIP3 1 0 3 1 2 4 7 1 SEIP1 1 0 0 0 0 0 2 0 SEIP2 1 0 0 0 1 1 2 0

30

Sub-Total 6 0 6 1 4 11 19 3 SOIP1 1 0 4 0 0 5 4 1 SOIP2 2 2 8 2 4 19 12 1 SOIP3 1 0 26 4 8 22 31 1 SEIP1 1 0 0 0 1 1 2 0 SEIP2 1 1 1 0 1 2 1 0

31

Sub-Total 6 3 39 6 14 49 50 3 SOIP1 2 4 24 0 4 25 21 1 SOIP2 1 1 14 0 2 12 13 1 SOIP3 1 2 20 0 2 19 15 1 SEIP1 1 0 0 0 0 0 2 0

32

Sub-Total 5 7 58 0 8 56 51 3 SOIP1 1 1 15 1 3 12 16 1 SOIP2 1 1 14 0 4 11 13 1 SOIP3 1 6 14 0 6 9 21 1

33

Sub-Total 3 8 43 1 13 32 50 3

191

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 1 0 4 0 1 2 2 0 SOIP2 1 0 7 0 1 7 4 1 SOIP3 2 0 3 0 2 5 3 0 SEIP1 1 0 3 0 0 2 1 1 SEIP2 1 1 3 0 1 7 1 0 SEIP3 1 0 1 0 1 2 2 0 SEIP4 1 0 6 0 2 6 3 1

34

Sub-Total 8 1 27 0 8 31 16 3 SOIP1 1 0 0 1 0 2 2 1 SOIP2 1 0 5 4 6 2 9 1 SOIP3 1 0 1 0 0 0 2 1

35

Sub-Total 3 0 6 5 6 4 13 3 SOIP1 2 1 5 1 3 9 8 1 SOIP2 2 0 11 1 3 9 14 1 SOIP3 1 0 3 0 0 2 3 0 SEIP1 1 0 1 0 0 1 2 0 SEIP2 1 0 1 0 0 0 1 0 SEIP3 1 0 1 0 0 0 2 0 SEIP4 1 0 2 1 0 2 2 1

36

Sub-Total 9 1 24 3 6 23 32 3 SOIP1 1 0 14 8 8 4 24 0 SOIP2 2 0 20 2 3 4 21 1 SOIP3 1 0 5 0 1 5 9 1 SEIP1 1 1 8 0 1 4 9 1

37

Sub-Total 5 1 47 10 13 17 63 3 SOIP1 2 0 13 1 8 8 15 1 SOIP2 1 0 1 0 1 1 2 1 SOIP3 1 0 0 0 3 11 5 1 SEIP1 1 1 1 0 0 1 1 0 SEIP2 1 0 1 0 1 1 1 0 SEIP3 2 0 6 1 1 9 11 0 SEIP4 1 0 1 0 0 0 3 0 SEIP5 2 0 3 0 2 2 5 1 SEIP6 1 0 2 0 2 2 3 0 SB1 0 0 0 0 0 2 2 0

38

Sub-Total 12 1 28 2 18 37 48 4 SOIP1 1 0 11 6 7 8 17 1 SOIP2 1 0 2 1 2 2 5 1 SOIP3 1 0 27 4 7 13 20 1

39

Sub-Total 3 0 40 11 16 23 42 3 SOIP1 1 0 15 0 5 21 13 1 SOIP2 1 0 6 1 0 9 4 1 SOIP3 1 0 4 0 0 7 3 1

40

Sub-Total 3 0 25 1 5 37 20 3

192

Table 4-30. (Continued)

TC (Level 1)

CM (Level 2)

SC (Level 3)

No. Study Participant IT TRF CRF MF TCF SLR PSS GS

SOIP1 2 1 9 2 8 7 20 1 SOIP2 1 0 2 0 0 1 3 1 SOIP3 4 0 1 0 0 6 3 1 SEIP1 1 0 5 0 0 1 6 0 SB1 0 0 0 0 0 0 0 0 SB2 0 0 0 0 0 2 0 0 SB3 0 0 0 0 0 1 0 0

41

Sub-Total 8 1 17 2 8 18 32 3 SOIP1 1 0 5 2 2 4 9 1 SOIP2 1 0 1 1 1 4 3 0 SOIP3 0 0 1 1 1 1 1 0 SEIP1 1 1 2 0 2 0 2 0 SEIP2 1 0 2 0 1 1 2 1 SEIP3 2 0 2 0 0 1 4 0 SEIP4 1 1 5 1 4 0 6 1

42

Sub-Total 7 2 18 5 11 11 28 3

193

Table 4-31 provides succinct statistics of cognitive coordination data during the

forty-two study participants’ Web searches detailed in Table 4-30.

194

Table 4-31. Number and type of cognitive coordination occurrences per study participant Study Participant TC CM SC Total

IT TRF CRF MF TCF SLR PSS GS 1 3 2 20 3 12 17 26 2 85 2 3 0 4 12 9 4 25 2 59 3 4 0 22 14 20 23 41 3 127 4 4 0 51 11 16 48 53 6 189 5 14 4 13 5 14 38 52 3 143 6 5 0 14 4 6 16 21 4 70 7 3 2 38 4 25 33 47 4 156 8 6 2 29 5 15 25 38 4 124 9 11 0 22 1 11 38 47 6 136 10 6 3 10 3 4 27 38 3 94 11 6 0 6 0 4 26 27 2 71 12 5 0 30 0 11 43 57 3 149 13 9 5 16 1 12 28 46 2 119 14 3 0 67 1 8 29 42 5 155 15 7 3 11 1 12 40 52 3 129 16 7 1 16 3 15 44 39 3 128 17 10 2 52 7 15 47 49 6 188 18 12 9 51 0 29 36 43 4 184 19 4 0 35 0 4 40 41 7 131 20 8 0 43 8 8 28 35 2 132 21 6 3 25 2 10 50 71 2 169 22 16 0 39 5 8 48 36 6 158 23 13 3 21 0 10 48 41 3 139 24 4 0 18 1 6 30 34 3 96 25 3 0 25 1 2 28 23 2 84 26 9 1 16 6 10 32 30 3 107 27 7 4 22 0 4 51 26 3 117 28 8 6 40 2 13 56 46 3 174 29 6 3 12 1 3 24 24 3 76 30 6 0 6 1 4 11 19 3 50

195

Table 4-31. (Continued) Study Participant TC CM SC Total

IT TRF CRF MF TCF SLR PSS GS 31 6 3 39 6 14 49 50 3 170 32 5 7 58 0 8 56 51 3 188 33 3 8 43 1 13 32 50 3 153 34 8 1 27 0 8 31 16 3 94 35 3 0 6 5 6 4 13 3 40 36 9 1 24 3 6 23 32 3 101 37 5 1 47 10 13 17 63 3 159 38 12 1 28 2 18 37 48 4 150 39 3 0 40 11 16 23 42 3 138 40 3 0 25 1 5 37 20 3 94 41 8 1 17 2 8 18 32 3 89 42 7 2 18 5 11 11 28 3 85 Total 280 78 1,146 148 446 1,346 1,614 142 5,200

Table 4-31 indicates a total of 5,200 cognitive coordination behaviour occurrences,

either task coordination (information task), coordination mechanism (term relevance

feedback, content relevance feedback, magnitude feedback, tactical review

feedback, and self-learning and regulating), or strategy coordination (problem

specific strategy and global strategy) during forty-two Web searches. The mean

number of cognitive coordination occurrence per search episode was 124 with a

range from 40 to 189.

196

Table 4-32 provides a summary of the cognitive coordination types and

occurrences.

Table 4-32. Summary of cognitive coordination types and occurrences Cognitive Coordination

Number of Searches

(out of 42)

Number of Occurrences

% Mean Range

IT 42 280 5 6.7 3-16 Task Coordination (TC)

TRF 25 78 1.5 3.1 1-9 CRF 42 1,146 22 27.3 4-67 MF 34 148 3 4.4 1-14 TCF 42 446 8.5 10.6 2-29 SLR 42 1,346 26 32.0 4-56 3,164 61

Coordination Mechanism (CM)

PSS 42 1,614 31 38.4 13-71 GS 42 142 3 3.4 2-7 1,756 34

Strategy Coordination (SC)

Total 5,200 100

Except for TRF and MF, other cognitive coordination behaviours all

occurred at each study participant’s Web search interactions. They were

indispensable activities during Web searching. Study participants did

identify an information problem to be searched and evaluate the retrieved

results relevant or not, and, therefore, adjust their search strategy.

Term relevance feedback and magnitude feedback did not take place at

each Web search episode. That means identifying a term was not essential

for a smooth Web search interaction, neither was the behaviour of judging

the size of system output.

Cognitive coordination mechanism played an important role in multitasking

Web search sessions. Study participants experienced most coordination on

197

the mechanism level (61%). 34% coordination occurred at strategy level

and only 5% coordination was for information tasks.

On average, for per Web search episode involving multiple information

problem searching, 6.7 task coordination occurred, 3.1 term relevance

feedback took place, 27.3 content relevance feedback occurred, 4.4

magnitude judgments took place, 10.6 tactical review feedback occurred,

32 self-learning and regulating processes occurred, with the adoption of

38.4 problem specific strategies and 3.4 global strategies.

4.6.5.2 Transition Analysis of Cognitive Coordination Levels Cognitive coordination occurrences can also be analysed in sequences. Study

participants transited from one cognitive coordination level to another cognitive

coordination level during Web searching. For example,

Study Participant 22 began with problem specific strategies of selecting

Google as Web search system and entering "correlation analysis" as search

query at strategy coordination level, and then shifted to content relevance

feedback on the second returned result that "it may be not very academic",

followed by a tactical review feedback of "I need to change the search

query" at coordination mechanism level.

Content analysis helped to identify many transition steps between different

cognitive coordination levels that study participants experienced.

Table 4-33 summarises the transition steps that per study participant experienced

during their Web searching process.

198

Table 4-33. Summary of cognitive coordination transition steps per study participant Study Participant

Steps of Transition

1 85 2 54 3 128 Steps of Transition Number4 183 Minimum 37 5 139 Maximum 187 6 63

Mean 122 7 145 8 122 9 135 10 93 11 73 12 148 13 122 14 151 15 127 16 127 17 183 18 183 19 131 20 131 21 169 22 152 23 138 24 95 25 83 26 105 27 116 28 172 29 74 30 49 31 168 32 187 33 162 34 94 35 37 36 100 37 150 38 146 39 137 40 93 41 85 42 84

199

200

The mean transition steps between cognitive coordination levels per search

were 122 with a range from 37 to 187. It shows that the transition between

cognitive coordination levels was a frequent behaviour. The transitions

were needed in order to achieve Web searching process.

Content analysis on the utterance-search segments also identified the first and the

last transition steps in most of the forty-two study participants’ Web searches. For

example, most study participants’ first transition step was from IT to PSS, that is,

they normally began with identification of an information problem, and then shifted

to select a proper Web search system and search query to start the course of Web

searching.

The data of transition steps between cognitive coordination levels formed the basis

for the following analysis of cognitive coordination sequences.

Tables 4-34 (Study Participants 1 to 21) and 4-35 (Study Participants 22 to 42)

show a full picture of cognitive coordination level sequences per study participant,

in which there was a change from one cognitive coordination level to another

cognitive coordination level.

Table 4-34. Sequence on cognitive coordination levels per study participant (Study Participants 1 to 21) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

IT to IT 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 IT to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to CRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to TCF 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to SLR 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 IT to PSS 3 3 4 4 13 5 3 5 10 5 6 5 9 3 7 7 10 13 4 8 6 IT to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to IT 0 0 0 0 2 0 0 0 0 0 0 0 0 0 1 0 0 2 0 0 0 TRF to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to CRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to TCF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to SLR 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to PSS 2 0 0 0 0 0 2 2 0 3 0 0 5 0 2 1 2 7 0 0 3 TRF to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CRF to IT 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 CRF to TRF 0 0 0 0 2 0 0 0 0 0 0 0 0 0 1 0 1 2 0 0 0 CRF to CRF 3 1 1 11 1 1 8 6 1 0 0 1 0 33 1 0 11 19 20 18 4 CRF to MF 0 1 0 1 0 0 0 0 0 0 0 0 2 0 1 0 1 0 0 0 1 CRF to TCF 7 0 8 3 0 1 5 6 6 0 1 3 3 4 0 7 3 14 2 6 2 CRF to SLR 8 0 5 20 3 7 17 12 14 5 4 15 6 10 2 8 27 15 5 16 7 CRF to PSS 2 2 7 15 6 3 2 5 1 4 0 10 5 19 6 1 6 1 5 3 10CRF to GS 0 0 1 3 1 1 2 0 0 1 1 1 0 1 0 0 0 0 2 0 0 Study Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

201

Table 4-34. (Continued) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

MF to IT 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 MF to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 MF to CRF 1 2 7 4 1 1 3 1 0 0 0 0 0 0 0 0 2 0 0 4 0 MF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 MF to TCF 1 7 5 1 1 1 1 0 0 1 0 0 0 0 0 2 0 0 0 1 0 MF to SLR 1 0 1 2 1 2 0 1 0 1 0 0 1 1 1 1 3 0 0 0 2 MF to PSS 0 2 1 3 2 0 0 3 0 0 0 0 0 0 0 0 1 0 0 2 0 MF to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 TCF to IT 0 0 0 0 4 0 0 2 2 0 0 0 2 0 0 1 0 6 1 0 0 TCF to TRF 1 0 0 0 2 0 0 1 0 0 0 0 0 0 0 0 0 4 0 0 2 TCF to CRF 1 0 1 1 0 1 1 1 0 0 0 1 0 2 2 0 1 0 2 0 0 TCF to MF 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 TCF to TCF 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TCF to SLR 1 0 2 4 0 0 2 2 0 0 1 0 3 0 0 0 3 0 0 0 1 TCF to PSS 10 8 15 10 6 3 19 9 9 4 3 9 7 6 9 14 11 17 1 8 7 TCF to GS 0 0 1 0 1 0 1 0 0 0 0 1 0 0 0 0 0 1 0 0 0 SLR to IT 0 0 0 0 1 1 0 1 2 1 1 0 0 0 1 0 0 1 0 0 2 SLR to TRF 1 0 0 0 0 0 0 1 0 2 0 0 2 0 0 0 1 0 0 0 1 SLR to CRF 4 0 2 12 1 3 3 1 6 1 2 4 0 6 0 6 10 1 4 6 1 SLR to MF 0 0 0 1 0 0 0 1 0 1 0 0 0 0 0 0 2 0 0 0 0 SLR to TCF 3 0 5 7 8 2 9 5 4 3 1 2 2 1 5 4 10 15 0 0 2 SLR to SLR 4 1 8 17 16 4 7 6 12 7 12 14 9 10 16 25 9 14 10 12 29SLR to PSS 4 3 6 9 10 2 12 9 11 10 12 22 14 8 17 7 10 2 24 10 17SLR to GS 2 0 1 3 2 3 0 1 3 2 0 1 1 2 1 2 4 3 3 0 1 Study Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

202

Table 4-34. (Continued) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

PSS to IT 1 1 1 2 5 2 0 1 2 5 3 3 5 0 3 4 5 1 0 8 2 PSS to TRF 1 0 0 0 0 0 1 0 0 1 0 0 3 0 2 1 0 3 0 0 0 PSS to CRF 11 1 11 23 10 7 21 20 15 9 4 24 15 26 8 10 26 31 8 15 20PSS to MF 3 10 14 8 4 4 4 4 1 1 0 0 1 1 0 3 3 0 0 8 1 PSS to TCF 2 1 1 3 5 0 7 3 1 0 2 6 7 3 6 2 1 0 1 1 6 PSS to SLR 3 3 6 6 17 2 5 4 12 13 11 14 9 5 21 10 4 6 25 0 15PSS to PSS 6 6 11 5 10 3 7 4 12 10 7 10 8 6 9 8 6 2 6 2 26PSS to GS 0 2 0 0 0 0 1 2 4 0 1 0 1 1 2 1 2 0 2 1 1 GS to IT 1 1 2 1 1 2 2 1 2 0 1 2 1 2 2 1 3 2 2 0 1 GS to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to CRF 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to TCF 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 0 0 GS to SLR 0 0 0 0 1 0 0 0 0 0 0 0 0 2 0 0 0 0 1 0 0 GS to PSS 0 0 0 2 0 1 0 1 3 2 0 0 1 0 0 1 1 1 2 1 0 GS to GS 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Study Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

203

Table 4-35. Sequence analysis on cognitive coordination levels per study participant (Study Participants 22 to 42) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

IT to IT 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to CRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to TCF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IT to SLR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 IT to PSS 15 13 5 3 9 7 8 6 6 6 5 3 8 3 9 5 12 3 3 5 7 IT to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to IT 0 2 0 0 2 0 1 3 0 2 1 0 1 0 0 0 1 0 0 0 1 TRF to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to CRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to TCF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to SLR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TRF to PSS 0 1 1 0 1 2 5 0 0 1 6 8 0 0 1 1 0 0 0 0 0 TRF to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 CRF to IT 3 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 CRF to TRF 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CRF to CRF 7 3 1 7 0 7 16 1 1 7 13 14 4 0 7 9 3 11 4 1 0 CRF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 0 0 2 CRF to TCF 1 2 3 9 0 0 3 0 1 2 2 6 2 5 0 4 8 4 2 2 9 CRF to SLR 27 14 12 13 13 13 19 9 3 24 38 16 21 0 11 15 14 15 18 4 2 CRF to PSS 1 2 2 4 3 1 2 2 1 5 4 6 0 0 3 16 3 8 1 8 3 CRF to GS 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 0 2 0 0 1 Study Participant 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

204

Table 4-35. (Continued) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

MF to IT 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 MF to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 MF to CRF 2 0 1 4 1 0 0 4 0 1 0 1 0 2 1 3 0 5 0 1 2 MF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 MF to TCF 1 0 0 0 1 0 1 0 1 1 0 0 0 0 0 3 0 4 0 1 1 MF to SLR 1 0 0 0 2 0 1 0 0 3 0 0 0 1 2 0 1 0 1 0 0 MF to PSS 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 2 1 1 0 0 1 MF to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 TCF to IT 2 2 1 1 1 0 3 0 0 0 0 0 3 0 0 0 0 2 0 1 1 TCF to TRF 0 2 0 0 0 1 3 0 0 3 3 2 1 0 0 1 0 2 0 0 2 TCF to CRF 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 3 0 0 0 TCF to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TCF to TCF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 TCF to SLR 0 0 0 0 0 1 1 0 0 0 0 0 0 0 1 0 3 0 0 0 0 TCF to PSS 5 7 4 2 5 2 6 3 4 11 4 11 4 5 5 10 10 14 5 6 8 TCF to GS 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 SLR to IT 2 5 0 0 2 0 0 0 0 0 0 0 1 0 2 0 0 0 0 2 0 SLR to TRF 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 SLR to CRF 15 3 1 4 6 4 6 5 2 7 16 3 13 0 1 4 4 5 9 0 2 SLR to MF 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 SLR to TCF 3 6 2 0 7 4 4 2 2 8 6 4 5 0 6 0 6 2 2 4 0 SLR to SLR 18 20 12 15 8 30 29 8 3 18 8 10 8 2 5 3 12 5 18 6 3 SLR to PSS 4 12 15 9 6 11 16 6 2 14 26 15 1 0 7 11 12 10 7 4 5 SLR to GS 6 3 0 0 2 2 1 1 2 0 0 0 3 2 2 0 2 1 1 2 0 Study Participant 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

205

206

Table 4-35. (Continued) Level of Cognitive Coordination Sequences (From —> To) Number of Sequences

PSS to IT 7 3 1 2 4 2 1 0 3 1 1 0 0 0 3 1 6 0 0 4 4 PSS to TRF 0 0 0 0 0 2 2 2 0 0 4 5 0 0 1 0 0 0 0 1 0 PSS to CRF 15 15 15 15 10 11 18 5 3 24 28 24 10 4 15 29 21 17 12 14 14PSS to MF 4 0 1 1 5 0 2 1 1 5 0 1 0 4 1 8 2 10 1 2 2 PSS to TCF 1 3 1 1 0 0 5 0 0 3 0 3 1 0 0 4 4 6 0 0 0 PSS to SLR 2 15 6 0 7 7 6 7 5 3 10 6 2 1 4 0 6 3 0 5 6 PSS to PSS 6 5 6 4 4 3 9 7 6 12 6 7 3 3 7 14 6 6 4 3 1 PSS to GS 0 0 3 2 0 1 2 2 1 2 2 3 0 0 1 3 1 0 2 1 1 GS to IT 1 1 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 0 0 GS to TRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to CRF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to MF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to TCF 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 GS to SLR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GS to PSS 3 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2 1 GS to GS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Study Participant 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Table 4-36 provides a summary of the sequences at each type of cognitive

coordination level detailed in Tables 4-34 and 4-35.

207

Table 4-36. Summary of each type of cognitive coordination level sequences

Level of Cognitive Coordination Sequences (From —> To)

Number of Sequences

Level of Cognitive Coordination Sequences

(From —> To)

Number of Sequences

IT to IT 1 SLR to IT 25 IT to TRF 0 SLR to TRF 11 IT to CRF 0 SLR to CRF 183 IT to MF 0 SLR to MF 7 IT to TCF 1 SLR to TCF 161 IT to SLR 4 SLR to SLR 483 IT to PSS 274 SLR to PSS 412 IT to GS 0 SLR to GS 65 TRF to IT 19 PSS to IT 97 TRF to TRF 0 PSS to TRF 29 TRF to CRF 0 PSS to CRF 634 TRF to MF 0 PSS to MF 121 TRF to TCF 0 PSS to TCF 90 TRF to SLR 2 PSS to SLR 292 TRF to PSS 56 PSS to PSS 286 TRF to GS 2 PSS to GS 48 CRF to IT 8 GS to IT 65 CRF to TRF 7 GS to TRF 0 CRF to CRF 256 GS to CRF 1 CRF to MF 12 GS to MF 0 CRF to TCF 146 GS to TCF 7 CRF to SLR 507 GS to SLR 4 CRF to PSS 188 GS to PSS 24 CRF to GS 20 GS to GS 1 MF to IT 3 Total 5133 MF to TRF 0 MF to CRF 54 MF to MF 1 MF to TCF 35 MF to SLR 30 MF to PSS 22 MF to GS 2 TCF to IT 35 TCF to TRF 30 TCF to CRF 19 TCF to MF 2 TCF to TCF 3 TCF to SLR 25 TCF to PSS 316 TCF to GS 7

208

Table 4-37 further summarises nine different types of sequences on three cognitive

coordination levels of the observed 5,133 sequences.

Table 4-37. Summary of cognitive coordination transition on three levels

Level of Cognitive Coordination Sequences (From —> To) Number of Sequences %

TC to TC 1 TC to CM 5 TC to SC 274 Sub-Total 280 5% CM to TC 90 CM to CM 1,974 CM to SC 1,090 Sub-Total 3,154 62% SC to TC 162 SC to CM 1,178 SC to SC 359 Sub-Total 1,699 33% Total 5,133

There were 5,133 cognitive coordination sequences during the forty-two

Web searches. Most frequent sequence was coordination mechanism (62%)

followed by another level of cognitive coordination, then strategy

coordination (33%) and task coordination (5%).

Specifically, the analysis of the sequences shows that the most frequently

occurring sequence was one type of coordination mechanism followed by

another type of coordination mechanism. The second most frequently

occurring coordination sequence was one type of strategy coordination

followed by one type of coordination mechanism.

209

4.6.6 Summary

This section reported the results of cognitive coordination attributed Web searching

behaviour, based on the analysis of utterance-search segments. Three levels of

cognitive coordination were identified, namely, task coordination, coordination

mechanism, and strategy coordination. The occurrences on each level and the

transition between the three cognitive coordination levels were also investigated.

The analysis revealed 5,200 coordination occurrences in total, related to users'

concern with the task coordination, coordination mechanism, or strategy

coordination. The analysis also discovered 5,133 cognitive coordination sequences,

with more sequences of coordination mechanism followed by another cognitive

coordination level.

210

4.7 Chapter Summary

This chapter presented the results of this study, including the cognitive and

behavioural characteristics of multitasking, cognitive coordination, and cognitive

shifts during the course of Web searching. It thereby addressed three minor

research questions: (1) How do users conduct their Web searches on multiple

information problems? (2) What types of cognitive shifts occur during Web search?

And (3) What levels of cognitive coordination occur during Web search?

The next chapter provides the interpretation of the findings. Most importantly, the

relationship between multitasking, cognitive coordination and cognitive shifts during

Web searching is identified and modelled.

211

212

Chapter 5 Discussion This chapter discusses the findings reported in the Results chapter relating to the

issues addressed by the research. Based on the research findings, a revised model

illustrating the relationship between multitasking, cognitive coordination and

cognitive shifts is built up and presented. The chapter then discusses the

implications of the study’s findings. The theoretical implications to be discussed are

the relationship of the findings to existing theories and models in Web search

studies and in cognitive and interactive IR research.

5.1 Key Findings of the Study

This section discusses the key findings of the study. It highlights the new findings

from the research and elaborates the differences between this study and other

related studies.

5.1.1 Multitasking during Web Search

The results have provided valuable insights into human multitasking information

behaviour in the Web searching context. Multitasking was found to be a major

element within Web search interactions. Users batched their multiple information

problems at hand when conducting Web search activity and ordered problems

consciously or unconsciously. Evolving information problems were likely to be

generated within the successive searches.

During the course of multitasking-featured Web searches, users switched the task

between multiple information problems including evolving information problems

searching. Furthermore, within each information problem searching, users usually

employed more than one Web search system, and modified/reformulated search

213

queries several times in parallel browsing windows/tabs concurrently. A session

with such characteristics was called a Web search session. Multiple Web search

sessions were applied into a Web search episode during which multiple information

problems searching occurred.

Figure 5-1 portrays a two-dimensional multitasking Web search behaviour.

Figure 5-1. Two-dimensional Multitasking Web search behaviour

Multitasking during Web searching is demonstrated as a two-dimensional behaviour.

The first dimension (IP axis) is multiple information problems searching through

task switching. The second dimension (A Web search episode axis) is multiple Web

search sessions. A Web search episode from logging on to logging off consists of

several Web search sessions linked to several information problem searches. One

information problem search is undertaken within a Web search session.

214

Dimension One: Multiple Information Problems Searching by Task Switching During Web searching, multiple information problems incorporate original

information problems, which motivate users’ Web searching activity, and evolving

information problems, which are generated during the continual Web searching

process. Multiple information problem searching tasks are switched within a Web

search episode.

Original Information Problems The findings show that the study participants’ multiple information problems which

initiated Web searching referred to a wide range of topic areas: Doctoral or Masters

research, course assignment, jobs & careers, travel, news, food & entertaining,

technology, finance information, lifestyle, favourite music & movies, online shopping,

sports information, answers to a specific inquiry, online gaming, as well as housing

rentals.

The results show that multiple information problems in a Web search episode were

mostly concentrated on research topic (45%) and course assignment (12%),

together with jobs (11%) and travel (8%) related information. Also, most of the

information problems undergoing a search involved a new problem area and were

unrelated to each other.

Factors Affecting Multiple Information Problems Search Ordering Study participants made a decision on the order of their multiple information

problems searching, based on several perceived factors, prior to the Web search.

Spink, Park and Koshman (2006) suggested that personal interest and problem

familiarity were two major factors affecting assigned information problems ordering.

215

The results of this study indicate that problem importance level—high to low,

randomness, and ease of finding information—high to low were the three major

factors in determining non-assigned information problem search ordering.

Other factors were also considered, such as task logic, problem urgency level, task

interest, ease of finding information—low to high, problem familiarity level, and

problem importance level—low to high and future usefulness. Task interest and

problem familiarity were not verified as major factors affecting non-assigned

information problems ordering. This finding supports Spink, Park and Koshman’s

(2006) call for further research into information problem prioritisation and ordering

in non-assigned problem environments.

Additionally, multiple factors were applicable for information problems search

ordering. In real-life Web searching, users determined multiple information

problems searching order based on the consideration of a combination of several of

these factors, which was more complex than our expectation.

Evolving Information Problem Developed during the Successive Web Searches Spink (1996, 2004) found evolving information problems existed during successive

searches and that a continuum of search sessions possibly was driven by an

evolving and changing information problem. Their results were confirmed in the

current study, in that over 70% of the study participants developed evolving

information problems during the successive Web searching. The number of

evolving information problems per study participant ranged from one to eight.

With respect to the reasons for evolving information problems generation, Spink

(2004) claimed that serendipity browsing was regarded as one factor. She found a

serendipity browsing episode involved the information seeker physically browsing

the library shelves for other information. However, serendipity browsing was not

216

discovered, either as important behaviour during Web searching or as the factor

leading to the generation of evolving information problems. Instead of diverting their

attention to browse other topics, study participants stuck to the searching activity on

certain topics.

The results of this study show that the development of evolving information

problems was linked to the reasons for users’ information task switching, which

were explored from cognitive coordination perspective.

Information Problem Searching Task Switching Much multitasking during Web searching studies suggested an evolving information

problem existed, but few have explicitly addressed the searching on evolving

information problems as part of the information tasks switching. For example, in

Spink’s (2004) study, information task switching was explored between several

information tasks set by the information seeker in advanced and serendipity

browsing. The study reported in this thesis originally explores the evolving

information problem searching task as a component of task switching during Web

searches.

Three types of information problem searching task were identified: searching on an

original information problem (SOIP), searching on an evolving information problem

(SEIP), and serendipity browsing (SB) on other topics. Four information problem

task switching patterns were abstracted from forty-two study participants’ Web

searches. The most frequent pattern (43%) was the switch between original

information problems searching and three or more evolving information problems

searches.

217

Dimension Two: A Web Search Session within One Information Problem Searching A Web search session is a ubiquitous but unexplored phenomenon during Web

searching. A Web search session refers to the practice by users of submitting the

entire sequence of queries by conducting Web search interactions with the Web

search systems over time in the windows/tabs when searching for information

related to a particular information problem.

The findings indicate that a study participant, on average, achieved a Web

search session by submitting five queries with two Web search systems in four

opened windows/tabs. Multiple Web search sessions were applied into multiple

information problems searching environment. Multiple Web search sessions

reflect an important multitasking trait of the Web searching process. The findings

may have implications for the more efficient design of Web searching tools.

Interestingly, the opened windows/tabs consisted of a primary searching

window/tab and several secondary searching windows/tabs. Study participants

went back and forth between several secondary windows/tabs and the primary

one. The move between multiple windows/tabs appeared chaotic. Nevertheless,

study participants believed that multiple windows/tabs browsing at the same time

was an efficient way of time management.

Similar results were found in Park’s (2008) study, that the behaviour of tabbed

browsing was closely related to time management, especially when people faced

demanding information tasks. She provided an example that one subject did not

wait till the database page was fully opened and decided to open a new tab to

search with Google. When failing to find satisfactory results, the subject switched

to another tab which was already opened to continue the search. Multiple tabs

browsing simultaneously resulted in high efficiency of task performance.

218

5.1.2 Cognitive Shifts during Web Search

This section examines the key findings concerning cognitive shifts. The different

findings from previous related studies were also indicated.

Cognitive shifting is the internal brain’s response to external stimuli. Cognitive

shifting was found as an essential element of Web searching interaction and

humans’ Web searching behaviour. This study revealed two types of cognitive shifts.

For one thing, the holistic cognitive shifts included users’ perception on the

information problem and overall information evaluation. For another, cognitive state

shifts reflected users’ focus changes between different cognitive states.

Holistic Cognitive Shifts

Measured prior to and after Web search interactions, holistic cognitive shifts were

represented as the changes of users’ perception over an information problem

embodied with respect to information problem understanding, information problem

stage, and information seeking stage, along with personal knowledge and

contribution to the information problem resolution.

The study results support Spink’s (2002), and Spink and Dee’s (2007) findings that

study participants reported tendencies of holistic shifts: forward, backward, and no

shift with respect to information problem understanding and knowledge contribution.

And different study participants reported different degrees of the shifts. In most

situations, they experienced forward shifting. The results show that 67% of the

study participants reported the Web search interaction significantly contributed to

the resolution of their information problems.

219

The Web search interactions had a positive impact on their information problem

resolving. Study participants got a better understanding of information problems;

positive changes occurred at information problem stage, seeking stage and

personal knowledge; and the contribution to the information problems resolution

was achieved after Web searching.

Cognitive State Shifts Unlike Spink’s (2002) and Spink and Dee’s (2007) studies, which were restricted to

the shifts on holistic cognition level, measured before and after Web search

interactions, more specific shifts on users’ cognitive state level during the Web

search interactions were also considered in this thesis study. Cognitive state shifts

were the cognitive changes in focus of the interaction between a user and a Web

search system with respect to the users’ cognitive states. It reflects how users

move from one cognitive state to another during the interaction with Web search

systems.

Five types of cognitive states were identified under Web searching circumstances,

including topic, strategy, evaluation, view and overview. Each of them stands for

five distinct phases in the Web searching process. The results show that the most

experienced cognitive states were strategy, evaluation, and view. The three states

formed more than 90% of the total cognitive states during the Web search

interactions.

As a result, most shifts occurred between the states of strategy (STR), evaluation

(EVA), and view (VIE). Study participants recurrently diverted their

attention/cognitive state from the search strategy adoption to evaluation of the

upcoming search results, then to the examination on the opened Webpage. Similar

results were reported in Robins’ (2002) study, that users’ higher concentration on

220

strategy and evaluation, together, strategic and evaluative foci constitute 60% (44%

and 16%, respectively) of all foci occurrences. It indicated that the majority of foci

dealt with input and output from the IR system.

The following table (Table 5-1) outlines the different findings in the current study

and Robins’ (2000) study as to the definition and types of focus shifts.

Table 5-1. Cognitive State Shift (this study) vs. Information Problem Shift (Robins, 2000) Cognitive State Shift Information Problem Shift Definition Cognitive changes in focus

of the interaction between a user and a Web search system in terms of users’ cognitive states.

Any change in focus of the interaction between a user and a search intermediary with respect to users’ information problem (p. 919).

TOP TOPIC STR STRAT EVA EVAL VIE DOC OVE N/A N/A I (indiscernible passage) N/A SNSR (social issues) N/A ST (the experiment) N/A SYS (IR system) N/A TECH (technical issues)

Types of Focus

N/A USER (users’ background)

Information problem shifts occurred under a mediated IR search environment

and the interaction was between a user and a search intermediary, whereas a

cognitive state shift took place in the Web searching context, and the

interaction was between an end-user and a Web search system.

Information problem shift referred to any change in terms of any aspect of

users’ information problem, while cognitive state shifts referred only to the

cognitive changes with respect to users’ cognitive states.

221

As such, Robins’ foci on problem shifts covered topic, strategy, evaluation,

document (view), indiscernible passage, social issues not related to the search,

discussion related to the experiment itself, discussion related to the IR system

itself, technical issues associated with the search, and users’ background

related to the search. The cognitive states included only the first four of Robins’

foci, plus the extra state of overview of users’ focusing on the overall search

outcome.

The Interplay between Holistic Cognitive Shifts and Cognitive State Shifts

In summary, this study explored the nature of the cognitive shifts, with both holistic

cognitive shifts over the perception of the information problem before and after Web

search, and cognitive state shifts in focus of the cognitive states and their shifts

during the searching interactions. Figure 5-2 illustrates these two types of cognitive

shifts.

222

Figure 5-2. Holistic cognitive shifts and cognitive state shifts

The cognitive state shifts experienced during the Web search interactions have

impact on the holistic cognitive shifts that occurred after Web searching. The states

of evaluation on the search strategies (STR), on the system returned results (EVA),

and on the gathered information (VIE) against the searching aim occurred a lot.

More occurrences of these states affected the shifts of holistic cognition on the

information problem understanding and the overall judgment on the contribution of

the search interactions to the information problem resolution. The forms of the

impact and how the impact is produced are important issues that deserve further

attention.

223

5.1.3 Cognitive Coordination during Web Search

Cognitive coordination was considered as an important behaviour during the course

of Web searching. The core of cognitive coordination was management

mechanisms, binding together with other cognitive processes including multitasking

and cognitive shifts, in order to move through users’ Web searching behaviour.

Study participants experienced complicated cognitive coordination processes

embedded within problem-solving oriented Web search interactions.

Three cognitive coordination levels have been identified: information task

coordination (level one), coordination mechanism (level two), and strategy

coordination (level three).

Level One: Information Task Coordination (TC) Information task level coordination was represented as the coordination process

between the information tasks (IT), including original information problem

identification, evolving information problem generation, problem searching task

switching, and windows/tabs browsing.

Level Two: Cognitive Coordination Mechanism (CM)

Cognitive coordination mechanism was the underlying system supporting

information task coordination. The results show that the coordination mechanism

involved a series of cognitive processing activities – most of these were content

relevance feedback (CRF) (36%) of making relevance judgments on the returned

results, and s self-learning and regulating (SLR) process (43%) of making sense of

the gathered information.

Other forms/types of coordination mechanism included tactical review feedback

224

(TCF), which made strategy-related adjustments based on the retrieved results,

term relevance feedback (TRF), consisting of users identifying a term (terms) within

the retrieved results, and magnitude feedback (MF) that made judgments on the

size of output.

This study’s results show that tactical review feedback essentially existed during

each Web search episode, even though the occurrence was not frequent. Term

relevance feedback and magnitude feedback seldom occurred within the

information problem searching. These two forms of feedback mechanism

constituted only 7% of all the mechanism occurrences.

The following table compares the cognitive coordination mechanism discussed in

this study with the feedback mechanism explored in Spink’s (1997) study.

Table 5-2. Cognitive coordination mechanism vs. feedback mechanism

Cognitive Coordination Mechanism Feedback Mechanism Term Relevance Feedback (2%) Term Relevance Feedback (8%) Content Relevance Feedback (36%) Content Relevance Feedback (40%)Magnitude Feedback (5%) Magnitude Feedback (45%) Tactical Review Feedback (14%) Tactical Review Feedback (6%)

Types (Occurrence %)

Self-learning and Regulating (43%) Term Review Feedback (1%)

Adapted from Spink’s (1997) feedback mechanism, the cognitive coordination

mechanism focuses on users’ interactive and dynamic searching process.

Two mechanisms were examined in different contexts. Feedback mechanisms

were investigated under a mediated IR search environment. Feedback was

initiated either by a user or a search intermediary, whereas the cognitive

coordination mechanism occurred in end-user Web search context and

therefore was initiated only by a user.

New meaning was given to the cognitive coordination mechanism due to the

Web searching context. For example, tactical review feedback in Spink’s (1997)

225

study was described as users’ input followed by a strategy related judgment to

display the search strategy history influencing the subsequent query. Spink

(1997) suggested that tactical review feedback would result in query

reformulation. In the current study, tactical review feedback as one type of

coordination mechanism was viewed as strategy-related adjustments which

were made based on the retrieved results review, represented as the decision

on changing a search keyword or changing an information source. As such,

tactical review feedback in the Web searching context would result in query

reformulation or strategic search re-planning, such as changing a Web search

engine to an academic database.

In addition to content relevance feedback, a self-learning and regulating

process was observed as an important interaction of coordination mechanisms

in which users examined information shown on the opened Website/Webpage,

followed by a sense-making process.

Magnitude feedback constituted only 5% of all the coordination mechanism

occurrences. It was observed as trivial interaction mechanism. In Spink’s (1997)

study, however, magnitude feedback was found to be a major interaction

among other types of feedback, occurring in 45% of the observed mediated

online search process.

Level Three: Cognitive Strategy Coordination (SC) Strategy coordination was considered as a strategic plan for solving information

problems within the resources available. It was represented as two kinds of

strategies: problem specific strategies (PSS) and global strategy (GS).

Problem specific strategy (PSS) was viewed as the collection of tactics on usable

Web searching tools for each information problem solving, which referred to the

226

selection of Web search systems, the adoption of search queries, the reviewing of

results set (pages), as well as the saving of relevant results.

Global strategy (GS) was an overall plan guiding the whole searching process,

presented as users’ decisions on the allocation of the one-hour searching time

between multiple information problems searching. Time allocation was not found as

a random behaviour but as a conscious strategy. Park (2008) also found that the

subjects were well aware of spending time in an optimal way among multiple

information tasks. Coordinating activities in Web seeking and retrieval context

entailed task switching, tabbed browsing, strategic search planning and information

evaluation, which were all closely related to time management.

The Interplay between Three Cognitive Coordination Levels The most basic level was the information task level, standing for the coordination of

interleaving between multiple information tasks. The second level was the cognitive

coordination mechanism level. Mechanisms included feedback, and a self-learning

and regulating process. The highest level was the cognitive strategy level, in which

the coordination occurred between the strategies: global strategy (plan for the

entire search) and problem specific strategy (sub-plan for each information problem

search).

The occurrences of coordination at strategy and mechanism levels supported the

coordination process among multiple information tasks.

The results also show that cognitive coordination occurred as sequences.

Transition analysis was presented to show the sequence shift between three

cognitive coordination levels. Most frequent transition (62%) was at the

coordination mechanism level, followed by another cognitive coordination level.

227

The figure below (Figure 5-3), which exemplifies the interplay between the three

coordination levels, was adapted from Study Participant 36’s Web searching

flowchart.

Figure 5-3. Flowchart example of interplay between the three coordination levels (Study Participant 36)

228

As illustrated, the generation of the evolving information problem, and the task

switching from searching on the original information problem “mediation

orgnisation” to searching on the evolving information problem “government portal

information”, were fundamentally based on the reasoning mechanisms, such as

content relevance feedback, a self-learning and regulating process, and tactical

review feedback. The occurrences of coordination at strategy and mechanism

levels strongly explained how the coordination among multiple information tasks

was achieved. The transition process between three coordination levels is shown in

Figure 5-4:

Figure 5-4. Shifts between the three coordination levels (Study Participant 36)

Three cognitive coordination levels occurred as a sequence. Information tasks

initiated the process of coordination and then were pushed by the strategies and

cognitive mechanisms. The task level coordination did not continue without the

existence and support of the coordination on strategies and cognitive mechanisms

levels.

229

Figure 5-5 demonstrates the interplay structure between three levels of cognitive

coordination behaviour.

Figure 5-5. Interplay between three cognitive coordination levels

Strategy coordination and coordination mechanisms support/explain the task

coordination process. Sequence transitions take place between three cognitive

coordination levels, that is, TC <—> CM, CM <—> SC, and SC <—> TC. The shifts

also occur between individual types of coordination mechanisms, and between

individual types of cognitive strategies.

230

231

5.2 Revised Relationship Model

The study aims to develop a sound Web search model which can illustrate users’

searching behaviours during Web search interactions, especially users’ cognitive

processes within such interactions. The major research problem to be addressed in

this study is:

What is the relationship between multitasking, cognitive coordination, and

cognitive shifts during Web search?

The research to this point has laid a foundation to explore any possible relationship

between the three behaviours that may emerge from the results. In this section, the

theoretical model presented in Figure 2-2 (Chapter 2), based on the review of

related literature, is revisited with empirical findings of this study.

Based on the findings of this study, a revised model has been built up to illustrate

the relationship between multitasking, cognitive coordination and cognitive shifts

during Web search, as shown in Figure 5-6.

232

Figure 5-6. Multitasking, Cognitive Coordination and Cognitive Shifts (MCC) Model

Web Search System User

Web Search Problems

System Returned Results

Web Search Interaction

Cognitive Shifts

Cognitive Coordination

Multitasking

Multiple IPs Ordering

Level 1: Task Coordination

Level 2: Coordination Mechanism

Level 3: Strategy Coordination

External Force: • Individual Differences • IP Attributes

Brain’s Response

Task Switching: • OIP Identification and Searching • EIP Generation and Searching • Serendipity Browsing

Multiple Web Search Session Windows/Tabs Browsing

Holistic Shifts State Shifts

• Compared to the initial literature review model, the MCC model has been

empirically validated through forty-two study participants’ Web searches.

• The MCC model illustrates how users’ multitasking, cognitive coordination

and cognitive shifts interplay in order to obtain a smooth Web search

interaction. For example, how the searching on multiple information

problems (IP) is accomplished, and how the users’ cognitive shifts occur.

• Web searching is a dynamic interaction between users and Web search

systems, during which information problems ordering, evolving information

problems generation, searching task switching, task and mental

coordinating occur, and at a deeper level, cognitive shifts take place.

• Users explicitly conduct Web searching on original information problems

(OIP) which initiate the searching activity. Task level coordination is closely

linked to multitasking behaviour, and is embodied as original information

problem identification, evolving information problem (EIP) generation,

problem searching task switching, and windows/tabs browsing.

• The implicit mechanism level coordination and strategy level coordination

support/explain the task coordination process, especially for the

development of evolving information problem and information problem

searching task switching.

• The occurrence of coordination mechanisms directly results in the users’

shifts on cognitive states during Web searching. Furthermore, the

occurrence of state shifts influences their perception on information problem

understanding and the judgment on the contribution to problem solving after

Web searches, namely, holistic shifts.

233

• Cognitive coordination is humans’ sub-conscious but vital ability to support

multitasking Web searches and to provoke the occurrence of cognitive shifts.

• Cognitive coordination, the hinge linking multitasking and cognitive shifts

together, moves users through dynamic and interactive Web searches.

Without cognitive coordination, neither multitasking Web search nor

complicated mental shifting can occur.

• The user brain’s response of cognitive shifts may also be affected by

external forces, such as individual differences and information problem

attributes. In this study, different study participants experienced different

cognitive shifts, and again the same study participant experienced different

cognitive shifts over different searched information problems. Further

research on how individual differences and information problem

characteristics are associated with cognitive shifts needs to be explored.

234

5.3 Theoretical Implications

As discussed above, the study identified the nature of and the relationship between

multitasking, cognitive coordination, and cognitive shifts in the Web searching

context. It provides researchers with a framework which can form the basis of a

better understanding of users’ Web searching behaviour. This section discusses the

study’s implications for research and the possible contributions to the theoretical

construction on the research area of Web search and interactive IR.

5.3.1 Implications for Multiple Search Sessions Model

This study identified the occurrences of multiple Web search sessions within

multiple information problems searching process. Three facets of a Web search

session (queries, search systems, and windows/tabs) were derived from the data

analysis. Study participants submitted the sequence of several queries into multiple

Web search systems over time in multiple opened windows/tabs. On average, a

study participant achieved a Web search session by submitting 5 queries via

interacting with 2 Web search systems in 4 opened windows/tabs.

This finding provides implications for the development of a multiple session model

in the Web IR context. The results confirm Spink’s (1996) statement which

suggested that users conducted a continuum of search sessions possibly driven by

an evolving and changing information problem, and support Spink’s (1996) call for

further exploration on the dimension of a multiple search session model as

representative of users’ behaviour in the interactive IR context. Existing studies

which discussed users’ multiple searches were limited to a single information

problem and a single IR system (Huang, 1992; Saracevic, Mokros & Su, 1990;

Spink, 1996). The present research expands the model from a single information

problem searching to multiple information problems searching when figuring out the

235

multiple search sessions.

5.3.2 Implications for Cognitive IR Model: The Role of Cognitive Coordination

Multitasking and Task Switching Research Task switching is an integral component of multitasking behaviour. Multitasking is a

critical human behaviour that allows people to cope with complex environments by

handling multiple tasks simultaneously through task switching (Spink, Park, Jansen

& Pedersen, 2006).

This study explored task switching behaviour on the cognitive level. It found that the

occurrence of cognitive coordination played an active role in task switching

activities. Previous Web search studies (Spink, Park, Jansen & Pedersen, 2006;

Spink, Park & Koshman, 2006) did trials on task switching investigation. However,

few of them were from cognitive perspectives. For example, Spink (2004)

suggested that the information seeker experienced a shift of interest, became

bored with one information task and wanted to proceed with another one. In her

study, interest shift was considered as the factor invoking information seeker’s task

switching. This thesis study revealed task switching behaviour was prompted by

several factors, especially by cognitive coordination factors.

Revisiting Task Switching from Cognitive Coordination Perspective As stated previously in the Results chapter, there were eight reasons for users’

Web searching tasks switching, which were obtained from the analysis of the study

participants’ written statements on post-Web search questionnaires and the oral

retrospective explanations during the post-Web search interviews. The eight

reasons could be clustered into four types of factors, prompting the study

236

participants’ task of switching behaviour.

The first two factors existed but were minor. One, an emotional factor, was

concerned with study participants’ emotion. Study participants switched from one

searching task to another due to being bored with the current information problem.

This is the same finding as the interest shift identified in Spink’s (2004) study. The

other factor invoking task switching was incurred by visual cues in which study

participants “followed his or her nose” to switch from one task to another

unconsciously.

The occurrence of coordination mechanisms was a major factor supporting task

switching. Study participants constructed a consciously reasoned coordination

process. Information tasks were switched, based on the cognitive feedback,

including relevance judgment, self-learning and regulating process, and tactical

review feedback. Evidence could be elicited from study participants’ statements,

such as “I have found enough related information” and “I did not think the

information that I have found was useful”.

The cognitive strategy of the time allocation between multiple information problems

also mainly affected the behaviour of task switching. Study participants switched

between multiple searching tasks due to limited time or sufficient time. Here are

some examples:

Ok, that's the first one. Oh, I only got 30 minutes left. I have to go to the

second information problem searching. (Limited time, Study Participant 10)

Still 40 minutes left, I’d like to find out more detailed information on the first

information problem. (Sufficient time, Study Participant 6)

Therefore, task switching not only occurs at physical multitasking level and

emotional level, it is a behaviour correlated closely with humans’ cognitive

237

processing. The exploration of task switching on cognitive coordination levels is an

indispensable dimension of a cognitive IR model.

238

Chapter 6 Conclusion and Further Research

This chapter begins with a summary of the dissertation research, followed by the

discussions of its significance and contributions, limitations, as well as the areas of

further research. The limitations section provides an overview of the constraints

realised in the study. The further research section provides an explanation of

additional areas to be considered for extension of the research program.

6.1 Summary of the Study

This study provides important results and conclusions from our observations on the

study of participants’ verbal and written statements associated with the behaviours

of multitasking, cognitive coordination and cognitive shifting during Web searching.

The roles of multitasking, cognitive coordination, and cognitive shifts within the

course of Web searching were discussed. Cognitive coordination mechanisms,

including mainly content relevance feedback, tactical review feedback and a self-

learning and regulating process, drive the task coordination behaviour of task

switching and efficient multiple-task performance. The occurrence of coordination

mechanisms directly results in the users’ shifts on cognitive states during Web

searching. Further, the occurrence of state shifts influences the users’ perception

on information problem understanding and judgment on the contribution to problem

solving after Web searches, that is, holistic shifts. Cognitive coordination is a

humans’ sub-conscious but vital ability to support multitasking Web searches and to

invoke the occurrence of cognitive shifts.

The results reported in this research confirm and extend the findings of previous

studies on multitasking Web search behaviour. The study results support Spink’s

239

(2004) and Spink, Park and Koshman’s (2006) suggestions that people usually

gather multiple information problems at hand when conducting Web searching

activity. The study participants were capable of searching multiple information

problems concurrently by ordering and switching between information problems.

During Web searching, evolving information problems and serendipity browsing

occurred. The generation of evolving information problems delayed or suspended

the users’ current information problem searching.

The results also confirm the findings of Spink’s (2002) and Spink and Dee’s (2007)

studies. Study participants experienced some levels of cognitive shifts (positive,

negative, or no shifts) in their information problem understanding, information

problem stage, information seeking stage, personal knowledge, and contribution to

the information problem resolution due to Web search interaction.

The research findings represent our efforts to understand the human cognitive

nature of interactive IR, including 1) study participants experienced the complex

cognitive coordination process embedded within problem-solving oriented Web

search interaction; 2) different study participants experienced a different degree of

complicated cognitive coordination process; 3) the relationship between

multitasking, cognitive coordination and cognitive shifts during Web searching was

investigated and modelled. The relationship model has been developed and

provisionally verified through systematic data collection and analysis of data

pertaining to the Web search interactions. Compared to the initial literature review

model, the revised model is more comprehensive and detailed, illustrating users’

cognitive process, mental tasks and mechanical tasks during Web searching; it also

models how these variables interplay to support the users’ dynamic Web searching

activity.

It is the belief that such an understanding is fundamental to basic research directed

240

toward theories and models of interactive information retrieval involving a variety of

Web search systems, and to applied research and development directed toward the

improvement of the human–information/computer interaction involved in Web

searching.

6.2 Significance of the Study

Research investigating the complex nature of users’ multitasking searching of Web

search systems is still in its infancy. Existing multitasking research during Web

search sessions is restricted to multiple information problems level, including

information and non-information multitasking interplay (Spink, 2004; Spink & Cole,

2005; Spink, Ozmutlu & Ozmutlu, 2002; Spink & Park, 2005; Spink, Park, Jansen &

Pedersen, 2006; Spink, Park & Koshman, 2006). Multitasking is humans’ ability

associated closely with their cognitive processing and coordination capabilities.

There is still much research to be done in order to reach a greater understanding of

users’ Web searching behaviour.

This study makes a trial to start holistic and synthetic research investigating users’

underlying cognitive mechanisms which prompt and support the occurrence of

multitasking behaviour and provoke users’ shifts in cognition afterward in order to

achieve effective Web searches. It is a new and significant research area with

major implications for cognitive and interactive IR theories and models, for

information system designers, and for our understanding of information searching

behaviours in general.

6.3 Contributions of the Study

The exploratory study provides a full picture of Web searching interaction, by taking

multitasking, cognitive coordination, and cognitive shifts as an interplay structure to

241

move through users’ Web searching process. The exploration of multitasking,

cognitive coordination and cognitive shifting extends the Web search model to

include cognitive mechanisms within multitasking searching environments.

Different from previous Web search and interactive IR models, the MCC model

proposed in this study provides a detailed micro-analysis and explanation of the

process of Web search interaction. It has contributed to a deep insight into how

users’ multitasking of information problems searching at an operational level and

cognitive changes at a higher mental level were achieved.

In contrast to prior work on Web search modelling, which concentrates more on

users’ cognitive efforts involved in single topic Web search interaction, the MCC

model examined how users’ cognitive efforts were made during multiple topics

involved in the Web searching process. The development of the MCC model not

only greatly enhances the value of the thesis but also contributes to the research

literature.

In addition to providing an understanding of multitasking, cognitive coordination and

cognitive shifts under an interactive Web searching context, the model can be used

to enlighten existing research of information seeking behaviour in other contexts,

such as e-commerce purchasing, e-government, and library search.

It also can be used as a theoretical base to design Web search systems that

support smooth multitasking Web searches involving lots of cognitive effort on the

part of users. The strength of an IR system is based on the strength of the models

that underpin its development. Users’ multitasking, task switching and task

coordination is little understood or supported by current search technologies. The

comprehensive MCC model may provide significant insights into the development

of adaptive Web technologies.

242

6.4 Limitations

Due to the exploratory nature of the study, there are some limitations to the findings.

First, the data collection was based on a small sample of postgraduate students at

the Queensland University of Technology, Brisbane, Australia. This may decrease

the generalisability of power of the study’s conclusions. The lack of generality can

be addressed by expanding the scale of the next study in different contexts.

Second, study participants were asked to verbalise about their thoughts and actions

at the same time as conducting their Web search. Thus, the acquisition of study

participants’ real-time thinking and cognitive reasoning process was possible. The

results show that roughly 76% of the study participants felt no discomfort in terms of

the think-aloud method and showed their searching activities were not affected

either. However, over 20% of the study participants did not frequently speak aloud

during their Web searching or could not provided sufficient verbal data, as they

were constrained by individual habits or differences. For example, study

participants stated “I don't know exactly what to say and how to say”, “speaking

cannot follow my thinking, quite stressful”, “when I am reading I have to speak, I

cannot concentrate on reading”.

Think-aloud is a timely method for research on the cognitive elements and

processes involved in interactive Web searches. Nevertheless, the method imposes

an extra cognitive burden on the study participants, diverting their concentration to

extra cognitive activity of translating the implicit thinking into explicit spoken

language. One study participant believed that the think-aloud method affected his

searching strategies a lot on the very open problem: he could not reflect his real

searching process as he normally does this without the pressure of speaking aloud.

The impact of the extra cognitive burden on the Web searching process was not

considered in the derivation of the study results.

243

Retrospective memory may be a supplementary choice with respect to obtaining

users’ thoughts. Study participants may keep diaries to record their thoughts and

actions or may be interviewed after Web searching.

While these limitations exist, the study allowed the collection of basic data and the

conclusion of results to form the basis for further research as discussed below.

6.5 Further Research

This study analysed the impact of cognitive coordination on multitasking activities

and cognitive shifts experienced as the users’ brain response. As recognised, Web

searching occurs as an on-going process. A user may re-conduct another Web

searching episode on the identical information problem which was searched before.

His or her memory imprinting on cognitive shifts may react on the new-run Web

searching process. For example, the adoption of search strategies and

corresponding cognitive coordination process may be experienced. The effect of

the “imprinting” of cognitive experience obtained before, such as the understanding

of the information problem, on the continuing Web searching behaviours needs

more exploration in future.

More efforts are needed on the research of exploring other cognitive variables and

developing theories and models of interactive IR and Web search by building upon

this thesis study.

Other areas of further research are listed below. Multiple search sessions were discussed as one of the multitasking components

under the Web searching context. Study participants normally submit more than one

query into several Web search systems in multiple opened windows/tabs during a

Web search session. Future research may be required depending on the nature

244

and occurrence of multiple Web search sessions and the development of a multiple

Web search sessions model.

Additionally, in order to identify details of the transition between cognitive

coordination levels, more research is needed to examine the transition steps, and

potentially any pattern from one cognitive coordination level to another. For

example, what are the first transition step and the last transition step? What are the

first 10 steps and the last 10 steps?

We also need to learn how individual differences, as well as the information

problem attributes, influence the cognitive Web searching process. The current

findings suggest that different study participants experienced different cognitive

shifts; and that the same study participant experienced different cognitive shifts on

three different information problems. User-based variables and the attributes of

information problem are factors that should not be ignored.

In terms of data analysis methods, the analysis of this study’s results was mainly

based on qualitative research methodology. The relationship between multitasking,

cognitive coordination and cognitive shifts was derived from grounded theory and

content/verbal protocol analysis techniques. It may be possible to provide a

statistical test of the relationship since numerical data were collected; for example,

the correlation analysis was employed by previous Web searching behaviour

studies.

245

246

Appendix A

Participation Information and Consent Form

247

PARTICIPANT INFORMATION for QUT RESEARCH PROJECT

Multitasking, Cognitive Coordination and Cognitive Shifts During Web Searching

Research Team Contacts Jia Tina Du – PhD student Professor Amanda Spink – Supervisor

Phone: 3138 9577 Phone: 3138 9583 Email: jia.du@student.qut.edu.au Email: ah.spink@qut.edu.au

Description This project is being undertaken as part of a PhD project for Jia Tina Du. The purpose of this project is to study how people interact with Web search engines. The research team requests your assistance because the research involves studying people’s Web information seeking and retrieving behaviours during the interaction with Web search engines. The behaviours are focused on how to solve information problems, how to cognitively coordinate between multiple information problems and available resources, and what types of cognitive shifts experienced while searching on the Web.

Participation Your participation in this project is voluntary. If you do agree to participate, you can withdraw from participation at any time during the project without comment or penalty. Your decision to participate will in no way impact upon your current or future relationship with QUT (for example your grades). Your participation will involve: (1) complete a pre-questionnaire; (2) conduct searching activity and think aloud your actions and reasons during the searching sessions, the researcher will sit aside to keep observation notes, and your search logs and thinking aloud data will be recorded simultaneously by Camtasia Studio software installed in the computer; (3) complete a post-questionnaire; (4) complete after-Web-search interview with the researcher. Your information problems will not be predefined. You will seek information on your own information problems related to research or wok or everyday life affairs in order to better simulate Web searching reality. In addition, there are no Web search engines restrictions. You will search the Web using Web search engines you prefer to employ in order to make the situation as real-life as possible. The whole process is estimated to take around one hour and the study will be conducted in a computer laboratory at Faculty of IT at QUT. Please note that the anonymous pre- and post-questionnaires you will fill out in the study will not be possible to withdraw, once you have submitted.

Expected benefits It is expected that this project may benefit you by helping you better understand your own Web searching behaviour interacting with Web search engines.

Risks There are no risks beyond normal day-to-day living associated with your participation in this project.

Confidentiality All comments and responses are anonymous and will be treated confidentially. The names of individual persons are not required in any of the responses. All of your responses that may have any identification will be turned into anonymous responses and your personal details will be erased from the records as soon as you submit your response and will not be revealed to anyone at any time including to QUT. Your comments during the interview are to be verified by you prior to final inclusion.

248

The thinking aloud audio recordings

• are to be verified by you prior to final inclusion; • will be destroyed after the contents have been transcribed; • will not be used for any other purpose; • the researcher only, with academic access upon request to the audio/video recording

(confidentiality/anonymity issues).

Consent to Participate We would like to ask you to sign a written consent form (enclosed) to confirm your agreement to participate.

Questions / further information about the project Please contact the researcher named above if you have any questions answered or if you require further information about the project.

Concerns / complaints regarding the conduct of the project QUT is committed to researcher integrity and the ethical conduct of research projects. However, if you do have any concerns or complaints about the ethical conduct of the project you may contact the QUT Research Ethics Officer on 3138 2340 or ethicscontact@qut.edu.au. The Research Ethics Officer is not connected with the research project and can facilitate a resolution to your concern in an impartial manner.

249

CONSENT FORM for QUT RESEARCH PROJECT

Multitasking, Cognitive Coordination and Cognitive Shifts During Web Searching

Statement of consent By signing below, you are indicating that you:

• have read and understood the information document regarding this project

• have had any questions answered to your satisfaction

• understand that if you have any additional questions you can contact the research team

• understand that you are free to withdraw at any time, without comment or penalty

• understand that you can contact the Research Ethics Officer on 3138 2340 or ethicscontact@qut.edu.au if you have concerns about the ethical conduct of the project

• agree to participate in the project

• understand that the project will include audio recording

Name

Signature

Date / /

250

Appendix B

Pre-Web Search Questionnaire

251

For project use: User Number___________________________ Date__________________________________

Pre-Web Search Questionnaire 1. Demographic variables: 1.1 What is your age? ______a. Under 20 ______b. 20–29 ______c. 30–39 ______d. 40–49 ______e. 50–59 ______f. 60 and over 1.2 Please indicate your gender: ______Male ______Female 1.3 Your faculty/institute/division: __________________________________________________________________ 1.4 Student status:

______a. Full time ______b. Part time Degree sought: _____________________________________________________

2. Web using experience: 2.1 How long have you been using the Web to look for information? ______a. One year–five years ______b. Six years–ten years ______c. Eleven years and over 2.2 Which Web browser is used most frequently for information viewing? (Ex.

Internet Explorer, Netscape, Mozilla, Maxthon, MyIE, and FireFox) __________________________________________________________________ 2.3 Which Web search engines do you use most frequently for information searching? __________________________________________________________________

252

3. Please describe your three information problems driving Web search in

detail: Information Problem 1: Information Problem 2: Information Problem 3:

253

4. For each information problem, is it a new problem area for you? Please tick

in the corresponding box.

Answer Problem 1 Problem 2 Problem 3 4.1 No.

Precisely defined 4.2 Yes. Is it precisely defined, or is it still fuzzy in any way? Still fuzzy

5. What is your current information problem stage for each information

problem? Please tick in the corresponding box.

Current information problem stage Problem 1 Problem 2 Problem 3 5.1 Problem Recognition – I am trying to determine whether or not the information problem I’m interested in is a real problem from the point of view of the discipline or area that interests me. I need a search so that I can discover whether others have identified the same issue as problematical.

5.2 Problem Definition – I have identified a real problem and now need to define it more closely or carefully so that I can determine how to approach the problem and how it relates to other information problems in the field. I need a search to help me define my research objectives.

5.3 Problem Resolution – I am in the process of resolving the problem and now need information to enable me to proceed with and complete that work. The question deals with a particular problem that I need to resolve.

5.4 Problem Completion – I have effectively finished the work I was doing and I am either tying up loose ends, or finding out from related work how best to report my research or where best to report it.

254

6. What is your current information seeking stage for each information

problem? Please tick in the corresponding box.

Current information seeking stage Problem 1 Problem 2 Problem 3 6.1 Collection – Having focused my problem I am now collecting specific relevant information problem.

6.2 Exploration – I am now identifying specific information sources that I think will be useful.

6.3 Formulation – The information I have found has enabled me to form a clearer focus on the problem.

6.4 Initiation – I have recognised that I need information at this stage of my work.

6.5 Presentation – I am in the process of finishing the collection of information for this stage of my work.

6.6 Selection – I have identified the general area in which I need information.

7. Please rank (circle) your current specific personal or internal knowledge in

relation to each information problem on the following 5-point Likert scale. Problem 1 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge Problem 2 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge Problem 3 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge

255

8. Please rank (circle) the degree of your uncertainty in relation to each

information problem on the following 5-point Likert scale. For Problem 1, How certain are you have: 8.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain For Problem 2, How certain are you have: 8.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain For Problem 3, How certain are you have: 8.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 8.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain

** END OF PRE-QUESTIONNAIRE ** **THANK YOU**

256

Appendix C

Post-Web Search Questionnaire

257

For project use: User Number___________________________ Date__________________________________

Post-Web Search Questionnaire 1. Which Web search systems were employed in your current searches? __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ 2.1 How did you order your information problems when searching on the Web?

Please make the corresponding choices.

The first information problem you searched is: ______a. Problem 1 ______b. Problem 2 ______c. Problem 3

The second information problem you searched is: ______a. Problem 1 ______b. Problem 2 ______c. Problem 3

The third information problem you searched is: ______a. Problem 1 ______b. Problem 2 ______c. Problem 3

2.2 Why did you order your three information problems in that way?

258

3.1 How did you switch information problems from one to another during the

search process? 3.2 Why did you switch your information problems in that way?

259

4.1 Was there any new information problem developed during your Web search? No/Yes: ________________________________

4.2 If yes, what were they?

4.3 What factors pushed you to generate new information problem(s) and thus deferred or suspended current information problem searching?

260

5. Did any changes occur in your understanding of each information problem as result of search? Please indicate (circle) each on the following 5-point Likert scale.

Problem 1 No changes in my Significant changes in my

understanding 1—--2—--3—--4—--5 understanding Problem 2 No changes in my Significant changes in my

understanding 1—--2—--3—--4—--5 understanding Problem 3 No changes in my Significant changes in my

understanding 1—--2—--3—--4—--5 understanding 6. What is current information problem stage for each information problem?

Please tick in the corresponding box.

Current information problem stage Problem 1 Problem 2 Problem 3 6.1 Problem Recognition – I am trying to determine whether or not the information problem I’m interested in is a real problem from the point of view of the discipline or area that interests me.

6.2 Problem Definition – I have identified a real problem and now need to define it more closely or carefully so that I can determine how to approach the problem and how it relates to other information problems in the field.

6.3 Problem Resolution – I am in the process of resolving the problem and now need information to enable me to proceed with and complete that work.

6.4 Problem Completion – I have effectively finished the work I was doing and I am either tying up loose ends.

261

7. What is current information seeking stage for each information problem? Please tick in the corresponding box.

Current information seeking stage Problem 1 Problem 2 Problem 3

7.1 Collection – Having focused my problem I am now collecting specific relevant information problem.

7.2 Exploration – I am now identifying specific information sources that I think will be useful.

7.3 Formulation – The information I have found has enabled me to form a clearer focus on the problem.

7.4 Initiation – I have recognised that I need information at this stage of my work.

7.5 Presentation – I am in the process of finishing the collection of information for this stage of my work.

7.6 Selection – I have identified the general area in which I need information.

8. Please rank (circle) your current specific personal or internal knowledge in

relation to each information problem after searching on the following 5-point Likert scale.

Problem 1 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge Problem 2 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge Problem 3 Little specific Considerable specific

knowledge 1—--2—--3—--4—--5 knowledge 9. Please estimate (circle) the contribution the Web searching has made to

the resolution of each information problem on the following 5-point Likert scale.

Problem 1 Nothing Substantial

contributed 1—--2—--3—--4—--5 contributed Problem 2 Nothing Substantial

contributed 1—--2—--3—--4—--5 contributed Problem 3 Nothing Substantial

contributed 1—--2—--3—--4—--5 contributed

262

10. Please rank (circle) the degree of your uncertainty in relation to each

information problem on the following 5-point Likert scale. For Problem 1, How certain are you have: 10.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain For Problem 2, How certain are you have: 10.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain For Problem 3, How certain are you have: 10.1 Recognised a real problem to investigate?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.2 Defined the problem appropriately?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.3 A problem that can be resolved?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.4 An effective way of presenting the results can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain 10.5 Relevant information is available and can be found?

Very uncertain 1—--2—--3—--4—--5 Very certain

263

11. Please rank (circle) the degree of your feeling on time pressure due to the

rate or pace at which information problems solved on the following 5-point Likert scale.

1—--2—--3—--4—--5 Low High

time pressure time pressure

** END OF POST-QUESTIONNAIRE ** **THANK YOU**

264

Appendix D

Semi-structure Interview Questions

265

For project use: User Number___________________________ Date__________________________________

Semi-structure Interview Questions 1. Why not try other popular Web search engines, such as Yahoo? Live Search? 2. Can you categorise your topics? It’s about your research, or personal interest, or anything else? 3. Why did you order your three information problems in that way?

266

4. Why did you switch your three information topics searching in that way? Under what kinds of circumstances, you would be happy with the current search results and went on with next topic searching? 5. Are you happy with your search results? 6. What do you feel when I am sitting beside you? Does it make you uneasy? What about speak aloud while searching? Do you think speaking aloud affected your searching performance?

267

Appendix E

Web Searching Process as Flowchart (Examples)

268

Appendix E-1. Flowchart-Study Participant 2

269

Appendix E-1. (Continued)

270

Appendix E-1. (Continued)

271

Appendix E-2. Flowchart-Study Participant 6

272

Appendix E-2. (Continued)

273

Appendix E-2. (Continued)

274

Appendix E-2. (Continued)

275

Appendix E-3. Flowchart-Study Participant 11

276

Appendix E-3. (Continued)

277

Appendix E-3. (Continued)

278

Appendix E-3. (Continued)

279

Appendix E-4. Flowchart-Study Participant 30

280

Appendix E-4. (Continued)

281

Appendix E-4. (Continued)

282

Appendix F

Steps of Transition between

Cognitive Coordination Behaviours

283

Appendix F. Cognitive coordination behaviours transition per study participant Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,30,78 1,11,50 1,5,68,117 IT to GS TRF to IT TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 8,24 TRF to GS CRF to IT CRF to TRF CRF to CRF 5,14,47 4 12 CRF to MF 5 CRF to TCF 6,17,33,36,55,58,67 8,13,17,21,48,55,82,

111 CRF to SLR 15,26,43,53,61,70,81,84 24,32,34,75,87 CRF to PSS 48,65 9,31 27,44,46,63,80,121,1

23 CRF to GS 66 MF to IT MF to TRF MF to CRF 4 8,30 7,11,74,79,86,110,12

0 MF to MF MF to TCF 10 5,14,17,20,23,26,40 38,71,98,101,114 MF to SLR 20 104 MF to PSS 43,53 96 MF to GS TCF to IT TCF to TRF 7 TCF to CRF 13 65 TCF to MF TCF to TCF 106 TCF to SLR 75 14,56

284

TCF to PSS 11,18,34,37,41,51,56,59,68,72

6,15,18,21,24,27,38,41

9,18,22,39,49,52,72,77,83,90,93,99,102,1

07,112 TCF to GS 115 SLR to IT SLR to TRF SLR to CRF 16,54,64,83 31,33 SLR to MF SLR to TCF 40,50,71 51,76,89,92,105 SLR to SLR 21,22,27,82 33 41,57,58,59,88,125,1

26,127 SLR to PSS 23,44,62,76 34,36,45 15,25,29,35,42,60 SLR to GS 28,85 128 PSS to IT 77 10 4 PSS to TRF 24 PSS to CRF 25,32,35,42,46,52,57,60

,66,69,80 3 16,20,23,26,43,45,47

,54,62,81,122 PSS to MF 3,9,19 7,13,16,19,22,25,29,

39,42,52 6,10,37,70,73,78,85,95,97,100,103,109,1

13,119 PSS to TCF 12,74 37 64 PSS to SLR 39,49,63 32,35,44 28,30,40,50,91,124 PSS to PSS 2,31,38,45,73,79 2,12,28,46,47,51 2,3,19,36,53,61,69,8

4,94,108,118 PSS to GS 48,54 GS to IT 29 49 67,116 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS GS to GS Study Participant

1 2 3

285

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,80,147,165 2,11,19,23,28,34,64,68,

94,98,103,124,132 2,15,28,45,53

IT to GS TRF to IT 102,123 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 48,109 TRF to GS CRF to IT CRF to TRF 47,122 CRF to CRF 9,28,75,95,101,118,12

1,124,167,178,179 77 61

CRF to MF 4 CRF to TCF 57,63,109 56 CRF to SLR 12,18,29,34,46,60,76,9

6,102,112,125,130,152,158,168,170,172,174,

176,180

5,50,88 8,12,23,39,41,47,62

CRF to PSS 10,15,21,44,68,73,83,90,92,94,105,119,122,1

35,137

43,62,75,78,80,106 6,21,59

CRF to GS 78,144,162 66 26 MF to IT MF to TRF MF to CRF 8,27,72,143 4 38 MF to MF MF to TCF 5 17 18 MF to SLR 42,85 13 4,30 MF to PSS 32,66,107 36,60 MF to GS TCF to IT 18,27,93,97 TCF to TRF 101,108

286

TCF to CRF 151 20 TCF to MF 59 TCF to TCF 19 TCF to SLR 50,110,114,160 TCF to PSS 6,25,38,40,58,64,70,87

,128,141 32,39,45,54,57,72, 10,35,57

TCF to GS 9 SLR to IT 131 14 SLR to TRF SLR to CRF 17,20,43,56,77,111,161

,169,171,173,175,177 105 5,25,40

SLR to MF 31 SLR to TCF 24,37,86,113,127,150,

159 8,26,38,71,92,96,100,10

8 9,34

SLR to SLR 19,23,30,35,36,47,51,54,55,97,98,115,126,13

1,155,181,182

6,7,14,25,84,89,111,114,119,126,127,128,129,13

0,135,136

13,24,31,48

SLR to PSS 48,52,61,99,103,116,132,153,156

15,21,30,41,51,85,90,112,115,120

32,49

SLR to GS 13,139,183 137,139 42,51,63 PSS to IT 146,164 1,10,22,33,63 1,44 PSS to TRF PSS to CRF 3,11,33,45,59,62,67,74

,82,89,91,93,100,104,108,117,120,123,129,13

4,136,157,166

42,46,49,61,65,74,76,79,87,121

7,11,22,46,55,58,60

PSS to MF 7,26,41,65,71,84,106,142

3,12,16,35 3,17,29,37

PSS to TCF 39,49,69 31,44,53,56,58 PSS to SLR 16,22,53,138,149,154 20,24,29,37,40,70,83,91

,95,99,104,107,110,113,118,125,134

33,50

PSS to PSS 2,81,88,133,148 52,55,69,73,81,82,86,116,117,133

16,36,54

PSS to GS GS to IT 79 67 27,52 GS to TRF GS to CRF 14 GS to MF GS to TCF 140 GS to SLR 138 GS to PSS 145,163 43 GS to GS Study Participant

4 5 6

287

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT 57 IT to TRF IT to CRF IT to MF IT to TCF 75 IT to SLR IT to PSS 1,51,114 1,43,77,91,101 1,14,20,58,88,96,104

,111,119,125 IT to GS TRF to IT TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 79,96 55,69 TRF to GS CRF to IT CRF to TRF CRF to CRF 4,5,15,22,27,35,54,81 18,19,57,71,72,119 43 CRF to MF CRF to TCF 28,31,76,117,137 5,15,73,80,105,113 31,36,39,65,81,86 CRF to SLR 6,11,16,23,36,39,46,5

5,60,82,88,91,103,106,123,131,143

11,20,28,36,46,49,58,89,93,109,115,120

4,9,23,25,34,44,47,60,74,90,99,109,114,1

31 CRF to PSS 70,98 9,13,26,34,64 51 CRF to GS 49,108 MF to IT 95 MF to TRF MF to CRF 34,75,102 8 MF to MF MF to TCF 140 MF to SLR 86 MF to PSS 32,52,84 MF to GS TCF to IT 74,76 13,87 TCF to TRF 95 68 TCF to CRF 114 TCF to MF TCF to TCF TCF to SLR 110,134 66,81

288

TCF to PSS 8,13,20,25,29,32,44,63,68,72,77,93,100,118,121,126,128,138,141

3,6,16,24,60,103,106,112,117

7,28,32,37,40,66,82,92,129

TCF to GS 145 SLR to IT 90 110,118 SLR to TRF 54 SLR to CRF 38,107,130 12 24,35,50,64,73,80 SLR to MF 83 SLR to TCF 7,12,19,24,43,92,125,

133,144 23,59,67,111,116 6,12,91,128

SLR to SLR 37,40,56,57,85,124,132

29,37,82,94,110,121 5,10,11,61,69,72,79,100,101,117,127,132

SLR to PSS 17,41,47,58,61,65,83,86,89,104,111,135

21,30,38,40,47,50,62,87,95

16,26,45,48,53,62,70,75,107,115,133

SLR to GS 122 102,121,135 PSS to IT 100 56,124 PSS to TRF 78 PSS to CRF 3,10,14,21,26,30,45,4

8,53,59,66,69,80,87,90,97,105,116,122,136,

142

4,10,14,17,25,27,33,35,45,48,56,63,70,79,88,

92,104,108,113,118

3,8,22,30,33,38,42,46,59,85,89,98,108,11

3,130

PSS to MF 33,74,102,139 7,31,51,85 94 PSS to TCF 62,67,71,94,99,120,1

27 2,65,102 27

PSS to SLR 18,42,64,84,129 22,39,53,61 15,49,52,63,68,71,78,106,116,120,126,13

4 PSS to PSS 2,9,52,73,101,115,119 44,78,96,107 2,17,21,29,41,67,83,

84,93,97,105,112 PSS to GS 112 41,97 18,54,76,122 GS to IT 50,113 42 19,103 GS to TRF GS to CRF GS to MF GS to TCF 109 GS to SLR GS to PSS 99 55,77,123 GS to GS 98 Study Participant

7 8 9

289

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR 71 IT to PSS 2,23,28,60,68 1,10,24,28,41,56 2,12,17,21,51 IT to GS TRF to IT TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 16,35,65 TRF to GS CRF to IT CRF to TRF CRF to CRF 54 CRF to MF CRF to TCF 4 14,40,78 CRF to SLR 18,30,41,78,90 7,61,64,70 24,30,36,55,66,7

5,91,105,107,112,116,119,137,139,

142 CRF to PSS 50,53,85,87 45,47,81,87,96,9

8,102,128,130,135

CRF to GS 57 73 49 MF to IT MF to TRF MF to CRF MF to MF MF to TCF 74 MF to SLR 32 MF to PSS MF to GS TCF to IT TCF to TRF TCF to CRF 65 TCF to MF

290

TCF to TCF TCF to SLR 33 TCF to PSS 21,38,75,82 5,26,37 10,15,41,43,69,7

9,94,100,126 TCF to GS 148 SLR to IT 67 40 SLR to TRF 15,34 SLR to CRF 56 60,69 29,106,134,138 SLR to MF 31 SLR to TCF 20,37,81 36 9,125 SLR to SLR 4,7,14,19,33,55,62 12,17,30,34,35,46,47,48,

49,50,51,52 27,28,62,71,84,108,109,113,120,121,122,123,124,1

45 SLR to PSS 5,8,10,12,42,46,63,72,

79,91 8,13,15,18,22,31,44,53,6

2,65,67,71 4,7,25,31,33,37,56,60,63,67,72,76,85,89,92,110,114,117,132,140,143,

146 SLR to GS 25,93 19 PSS to IT 1,22,27,59,70 9,23,27 1,11,16 PSS to TRF 64 PSS to CRF 17,29,40,49,52,77,84,8

6,89 3,6,63,72 13,23,35,39,44,4

6,48,53,74,77,80,86,90,95,97,101,104,111,115,118,127,129,136,141

PSS to MF 73 PSS to TCF 25,32 42,64,68,93,99,1

47 PSS to SLR 3,6,9,11,13,24,36,45,5

4,61,66,80,92 11,14,16,21,29,39,43,45,

59,66,68 3,6,8,18,26,32,59,61,70,83,88,131,

133,144 PSS to PSS 39,43,44,47,48,51,69,7

6,83,88 2,19,20,38,42,57,58 5,22,34,38,52,57,

58,73,82,103 PSS to GS 54 GS to IT 55 20,50 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS 26,58 GS to GS Study Participant

10 11 12

291

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,40,48,51,60,76,81,1

00,105 1,49,66 2,7,68,76,101,106,

117 IT to GS TRF to IT 75 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 6,15,27,33,44 89,94 TRF to GS CRF to IT CRF to TRF 93 CRF to CRF 62,68 13,14,17,31,38,39,43,52,

61,62,69,70,71,72,75,78,79,94,102,103,104,118,119,120,121,122,123,131,

136,141,142,150

39

CRF to MF 9 CRF to TCF 91,102,107 9,88,128,147 CRF to SLR 8,17,23,29,35,63 18,27,29,32,46,53,91,97,

105,111 14,23

CRF to PSS 46,69,85,87,98 15,40,44,59,63,73,76,80,82,85,95,109,124,125,13

2,134,137,139,143

40,53,60,78,91,112

CRF to GS 151 MF to IT MF to TRF MF to CRF MF to MF MF to TCF MF to SLR 4 4 10 MF to PSS MF to GS TCF to IT 50,75

292

TCF to TRF TCF to CRF 42,58 38,59 TCF to MF TCF to TCF TCF to SLR 57,92,94 TCF to PSS 71,89,96,103,108,112,

117 10,25,89,129,145,148 12,26,43,51,73,87,

110,115,120 TCF to GS SLR to IT 6 SLR to TRF 5,32 SLR to CRF 12,28,30,84,93,101 SLR to MF SLR to TCF 93,95 24 11,42,109,114,119 SLR to SLR 9,10,11,12,24,36,53,5

4,64 5,6,21,23,33,34,54,55,92

,98 20,31,32,45,48,56,62,63,82,96,97,122,123,124,125,126

SLR to PSS 13,18,20,25,30,42,55,58,65,73,78,110,115,1

20

7,35,56,99,106,112,114,116

15,18,21,24,28,33,36,46,49,57,64,71,

80,83,85,98,104 SLR to GS 37 19,47 127 PSS to IT 39,47,59,99,104 1,105,116 PSS to TRF 14,26,43 74,88 PSS to CRF 7,16,22,28,34,45,61,6

7,83,84,86,90,97,101,106

8,16,26,37,45,51,60,68,74,77,81,87,90,96,108,110,117,125,127,130,133,1

35,138,140,146,149

8,13,22,52,77,90,92,111

PSS to MF 3 3 PSS to TCF 49,56,70,74,88,111,11

6 41,57,144 25,37,50,58,72,86

PSS to SLR 19,31,41,52,72,77,109,114,119

11,83,100,113,115 5,17,19,27,30,35,41,44,47,55,61,70,79,81,84,95,103,10

8,113,118,121 PSS to PSS 2,21,66,82,113,118,12

1,122 2,36,50,67,86,107 3,4,16,29,34,54,65

,69,102 PSS to GS 79 64 66,99 GS to IT 80 48,65 67,100 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR 20,22 GS to PSS 38 GS to GS Study Participant

13 14 15

293

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,3,6,44,46,64,72 2,50,52,84,110,155,16

5,168,173,176 1,29,34,52,69,88,106,130,132,136,143,15

1,176 IT to GS TRF to IT 28,33 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 89 71,105 45,59,79,96,112,117,

153 TRF to GS CRF to IT 167 CRF to TRF 70 44,111 CRF to CRF 6,24,25,36,54,59,80,94

,136,157,170 47,61,74,90,98,99,100,101,123,124,138,139,155,156,157,158,

159,173,178 CRF to MF 4 CRF to TCF 8,13,16,22,25,28,6

2 64,73,149 11,14,17,41,64,67,86

,91,94,108,125,167,170,174

CRF to SLR 19,59,66,77,80,93,114,118

13,17,19,26,32,37,43,45,55,81,86,88,90,95,100,113,116,118,120,129,137,139,143,147,152,1

58,162

20,36,48,55,75,81,102,114,119,134,140,14

6,160,179,182

CRF to PSS 11 7,41,60,68,107,171 62 CRF to GS MF to IT 175 MF to TRF MF to CRF 5,31 MF to MF MF to TCF 32,35 MF to SLR 38 76,98,178

294

MF to PSS 78 MF to GS TCF to IT 63 68,87,105,142,150,1

75 TCF to TRF 32,58,95,116 TCF to CRF 12 TCF to MF 30 TCF to TCF TCF to SLR 123,125,131 TCF to PSS 9,14,17,23,26,29,3

3,36,54,84,100,108,121,124

21,29,34,39,62,65,74,133,141,145,150

6,9,12,15,18,23,42,65,72,77,84,92,109,12

1,126,168,171 TCF to GS 164 SLR to IT 135 SLR to TRF 104 SLR to CRF 21,58,79,92,113,11

7 16,18,44,87,89,99,117,

119,138,148 181

SLR to MF 77,97 SLR to TCF 53,83,99,107 11,20,33,38,122,124,1

30,132,140,144 5,8,22,31,57,71,76,83,104,115,120,141,1

49,163 SLR to SLR 20,39,48,51,52,67,

68,69,78,81,82,86,91,94,95,96,97,98,103,104,105,106,11

1,112,126

46,96,103,121,126,179,180,181,182

4,21,25,37,38,49,56,82,103,147,148,161,

162,180

SLR to PSS 40,49,56,60,87,115,119

9,14,47,56,91,101,114,127,153,163

26,39

SLR to GS 70,127 27,82,159,183 50,128,183 PSS to IT 2,5,43,45 1,51,154,164,172 131 PSS to TRF 88 27,78,152 PSS to CRF 7,10,12,15,18,24,2

7,61,65,76 3,23,35,40,42,53,58,63,67,69,72,79,85,93,106,112,115,128,135,142,146,151,156,161,166,16

9

10,13,16,19,35,40,43,46,54,60,63,66,73,80,85,89,93,97,107,110,113,118,122,133,137,144,154,166,169,1

72,177 PSS to MF 31,34,37 75,174,177 PSS to TCF 120,123 61 PSS to SLR 47,50,55,57,85,90,

102,110,116,125 8,10,15,102 3,7,24,30,70,127

PSS to PSS 4,30,73,74,75,101,109,122

22,57,66,92,111,134 2,53

PSS to GS 41 48,108 GS to IT 71 49,83,109 51,129 GS to TRF GS to CRF GS to MF GS to TCF 28

295

GS to SLR GS to PSS 42 160 165 GS to GS Study Participant

16 17 18

296

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,37,81,107 2,34,48,56,74,89,104,12

0 2,54,57,62,69,82

IT to GS TRF to IT TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 33,118,135 TRF to GS CRF to IT 56 CRF to TRF CRF to CRF 40,41,42,43,44,45,

46,63,64,65,66,83,88,91,110,111,112,1

15,116,117

14,24,25,26,27,36,40,65,68,76,82,85,108,111,112,

115,125,128

21,29,65,85

CRF to MF 154 CRF to TCF 86,113 10,28,37,53,71,86 9,125 CRF to SLR 27,29,31,33,67 15,21,41,58,60,69,77,91,

94,99,106,113,116,122,126,129

4,22,30,59,66,104,107

CRF to PSS 4,47,84,89,118 66,83,109 12,45,47,71,73,86,102,121,138,150

CRF to GS 35,92 MF to IT MF to TRF MF to CRF 9,13,20,52 MF to MF MF to TCF 4 MF to SLR MF to PSS 7,50 155,161 MF to GS 31 TCF to IT 106 TCF to TRF 32,117

297

TCF to CRF 87,114 TCF to MF TCF to TCF TCF to SLR 114 TCF to PSS 24 5,11,29,38,46,54,72,87 7,10,19,98,100,123

,126 TCF to GS SLR to IT 53,81 SLR to TRF 134 SLR to CRF 28,30,32,34 59,70,81,107,114,127 28 SLR to MF SLR to TCF 6,31 SLR to SLR 10,15,16,53,56,72,

73,96,99,129 16,17,42,43,61,62,78,79,

80,95,100,117 5,23,24,25,49,50,51,52,75,76,77,78,79,80,88,89,90,94,108,109,110,111,129,132,143,144,163,1

64,167 SLR to PSS 6,8,11,13,17,19,49,

51,54,57,60,68,70,74,76,78,94,97,100,102,120,122,124,1

30

18,22,44,63,92,96,101,118,123,130

14,26,35,38,40,67,91,95,105,112,115,130,133,145,165,1

68

SLR to GS 21,104,126 60 PSS to IT 1,33,47,55,73,88,103,11

9 1,68

PSS to TRF PSS to CRF 3,26,39,62,82,85,9

0,109 23,35,39,57,64,67,75,84,90,93,98,105,110,121,12

4

3,8,11,20,44,46,55,58,64,70,72,84,101,103,106,120,124,1

37,149,153 PSS to MF 3,6,8,12,19,30,49,51 160 PSS to TCF 23 45 18,97,99,113,116,1

22 PSS to SLR 5,7,9,12,14,18,20,4

8,50,52,55,59,69,71,75,77,95,98,101,103,119,121,123,12

5,128

13,27,34,37,39,48,74,87,93,128,131,1

33,142,162,166

PSS to PSS 2,25,38,58,61,108 97,102 15,16,17,36,41,42,43,63,83,92,96,119,127,136,139,140,141,146,147,148,151,152,156,157,158,

159 PSS to GS 79,131 131 169 GS to IT 36,80 61 GS to TRF GS to CRF GS to MF

298

GS to TCF 105 GS to SLR 93 GS to PSS 22,127 32 GS to GS Study Participant

19 20 21

299

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 2,8,17,20,28,36,41,44,

50,70,100,104,109,138,144

1,35,48,54,58,64,69,89,93,112,119,129,13

4

1,35,43,50,66

IT to GS TRF to IT 92,118 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 106 29 TRF to GS CRF to IT 19,40,43 CRF to TRF CRF to CRF 4,5,10,11,12,13,146 4,5,11 45 CRF to MF CRF to TCF 6 6,74 4,28,89 CRF to SLR 23,34,38,46,52,55,57,6

0,63,66,73,75,77,79,81,86,102,111,113,116,118,120,121,130,134,140

,147

12,29,37,50,56,66,81,84,96,100,104,115,1

31,137

7,12,23,31,37,40,46,53,63,70,76,85

CRF to PSS 14 77,79 26,94 CRF to GS MF to IT MF to TRF MF to CRF 22,33 69 MF to MF MF to TCF 30 MF to SLR 92 MF to PSS MF to GS TCF to IT 7,99 34,128 42 TCF to TRF 91,117 TCF to CRF

300

TCF to MF TCF to TCF TCF to SLR TCF to PSS 26,31,89,97,149 7,16,26,75,87,102,10

8 5,21,90,92

TCF to GS SLR to IT 35,103 47,57,68,105,133 SLR to TRF SLR to CRF 39,54,56,59,74,76,78,8

0,85,112,115,117,119,121,133

10,95,136 84

SLR to MF SLR to TCF 25,96,148 33,86,91,101,116,12

7 20,41

SLR to SLR 24,53,58,67,82,83,84,93,94,95,114,122,123,124,125,131,132,135

13,18,21,30,38,41,42,60,61,67,71,85,121,122,123,124,125,126

,132

13,14,17,54,55,56,73,77,78,79,80,83

SLR to PSS 61,64,126,136 14,19,22,24,31,39,43,45,62,72,82,97

8,10,15,18,24,32,38,47,57,60,64,71,7

4,81,86 SLR to GS 47,68,87,106,141,152 51,110,138 PSS to IT 1,16,27,49,108,137,14

3 53,63,88 65

PSS to TRF PSS to CRF 3,9,18,37,42,45,51,62,

65,72,101,110,129,139,145

3,28,36,49,55,65,73,76,78,80,83,99,103,1

14,130

3,6,11,22,25,27,30,36,39,44,52,62,75,

88,93 PSS to MF 21,29,32,91 68 PSS to TCF 98 15,25,107 91 PSS to SLR 105,151 9,17,20,23,32,40,44,

46,59,70,90,94,109,120,135

9,16,19,59,72,82

PSS to PSS 15,71,90,127,128,150 2,8,27,98,113 2,51,58,61,67,87 PSS to GS 33,48,95 GS to IT 69 111 34,49 GS to TRF GS to CRF GS to MF GS to TCF 88 GS to SLR GS to PSS 48,107,142 52 GS to GS Study Participant

22 23 24

301

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 2,56,71 2,6,26,45,52,55,59,77,

91 1,17,22,26,34,62,1

07 IT to GS TRF to IT 21,33 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 68 10,100 TRF to GS CRF to IT CRF to TRF 99 CRF to CRF 22,26,40,41,59,74,75 70,76,82,97,98,102

,110 CRF to MF CRF to TCF 9 CRF to SLR 5,12,17,23,27,37,42,4

5,48,52,60,76,81 9,11,30,35,37,48,66,73,75,79,84,94,99,101,

104

4,12,28,65,71,77,83,87,92,103,111,11

3,115 CRF to PSS 7,33,35,50 4,57,63 85 CRF to GS MF to IT MF to TRF MF to CRF 4 8 MF to MF MF to TCF 24 MF to SLR 19,42 MF to PSS 17 MF to GS TCF to IT 25 TCF to TRF 20 TCF to CRF 29 TCF to MF TCF to TCF

302

TCF to SLR 79 TCF to PSS 10,20 14,21,40,71,86 15,95 TCF to GS 89 SLR to IT 54,76 SLR to TRF 67 SLR to CRF 6,47,49,80 10,36,62,65,74,100 81,84,112,114 SLR to MF SLR to TCF 13,20,28,39,70,85,88 14,19,78,94 SLR to SLR 13,14,28,46,61,62,63,

64,65,66,67,68,77,78,79

12,38,49,61,80,81,95,96

5,6,7,13,29,30,36,37,38,39,40,41,46,47,50,51,52,53,54,55,56,57,58,59,66,6

7,72,73,80,93 SLR to PSS 15,18,24,29,38,43,53,

69,82 31,33,50,82,97,102 8,24,31,42,44,48,6

8,74,88,90,104 SLR to GS 43,105 60,116 PSS to IT 1,70 1,5,51,58 16,25 PSS to TRF 9,32 PSS to CRF 8,11,16,21,25,32,34,3

6,39,44,51,58,73 3,34,47,56,72,78,83,9

3,98,103 3,11,27,64,69,75,86

,91,96,101,109 PSS to MF 3 7,16,18,23,41 PSS to TCF 19 PSS to SLR 27,32,53,60,64,69,87 18,23,35,43,45,49,

89 PSS to PSS 30,31,57,72 15,22,46,92 2,63,108 PSS to GS 54,83 105 GS to IT 55 44,90 61,106 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS GS to GS Study Participant

25 26 27

303

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,71,87,94,105,120,160,

166 1,18,21,28,36,50 1,3,6,11,26,32

IT to GS TRF to IT 86 17,20,27 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 34,53,61,114,156 TRF to GS CRF to IT CRF to TRF CRF to CRF 4,55,63,74,89,96,97,107,

108,124,136,149,150,151,168,169

30 46

CRF to MF CRF to TCF 132,142,170 16 CRF to SLR 5,13,15,22,29,36,38,40,4

4,46,56,75,90,98,109,116,125,137,152

5,9,14,31,42,44,46,53,71

14,35,47

CRF to PSS 64,130 39,67 20 CRF to GS MF to IT MF to TRF MF to CRF 4 MF to MF MF to TCF 146 42 MF to SLR 162 MF to PSS MF to GS TCF to IT 93,104,165 TCF to TRF 33,85,155 TCF to CRF TCF to MF

304

TCF to TCF TCF to SLR 50 TCF to PSS 27,133,143,147,158,171 25,60,65 9,17,40,43 TCF to GS SLR to IT SLR to TRF 60 19 SLR to CRF 12,14,37,39,45,123 8,13,43,45,70 13,15 SLR to MF SLR to TCF 92,103,154,164 24,64 8,39 SLR to SLR 6,7,8,11,16,17,18,19,23,

24,30,47,68,76,77,78,79,80,81,91,99,100,101,102

,110,126,127,153,163

6,7,10,11,12,23,32,54

22,29,48

SLR to PSS 9,20,25,31,41,48,51,57,66,82,111,117,128,138,14

0

15,33,47,55,62,72 23,36

SLR to GS 69 58 30,49 PSS to IT 159 2,5,10 PSS to TRF 52,113 16,26 PSS to CRF 3,21,28,35,43,54,62,73,8

8,95,106,115,129,131,135,141,148,167

29,38,41,52,66 19,34,45

PSS to MF 145,161 3 41 PSS to TCF 26,32,49,84,157 PSS to SLR 10,59,65,67,122,139 18,22,57,61,63,69,74 7,12,21,28,38 PSS to PSS 2,42,58,72,83,112,121,1

34,144 2,37,40,51,56,68,73 4,18,27,33,37,44

PSS to GS 118,172 34,48 24 GS to IT 70,119 35,49 25,31 GS to TRF GS to CRF GS to MF GS to TCF 59 GS to SLR GS to PSS GS to GS Study Participant

28 29 30

305

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,16,31,36,60,77 1,11,15,85,129 1,61,106 IT to GS TRF to IT 35,59 14 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 44 32,75,79,113,144,18

3 12,80,112,124,135,

140,150,156 TRF to GS CRF to IT CRF to TRF 149 CRF to CRF 92,93,114,123,146,1

53,156 42,43,63,64,65,88,108,126,139,153,154,1

66,167

17,37,50,51,52,67,74,75,88,97,98,99,

100,148 CRF to MF CRF to TCF 129,163 17,115 5,47,118,121,129,1

42 CRF to SLR 4,6,8,11,19,23,40,46,

48,62,67,80,87,94,100,104,106,110,115,13

5,140,147,157,166

4,6,19,26,29,36,38,44,46,49,51,60,66,69,81,89,92,96,99,103,106,109,121,124,132,136,140,149,155,157,163,168,171,174,17

6,178,180,185

8,18,21,38,53,64,68,76,82,101,109,114,126,145,152,160

CRF to PSS 38,108,124,126,154 34,40,77,161 24,44,58,89,132,154

CRF to GS 75 127 MF to IT MF to TRF MF to CRF 162 4 MF to MF MF to TCF 28 MF to SLR 53,85,117 MF to PSS 121

306

MF to GS TCF to IT TCF to TRF 34,43,58 31,74,182 134,139 TCF to CRF 18 TCF to MF TCF to TCF TCF to SLR TCF to PSS 26,29,65,90,98,102,1

30,133,144,151,164 9,111,116,142 6,10,48,72,78,93,9

5,119,122,130,143 TCF to GS SLR to IT SLR to TRF SLR to CRF 5,7,22,47,86,105,107 5,25,37,39,45,50,91,

102,107,125,135,152,156,175,177,179

20,36,153

SLR to MF 116 SLR to TCF 25,33,42,57,64,97,10

1,143 8,30,73,110,141,181 9,77,92,138

SLR to SLR 12,20,21,24,4149,54,55,56,63,68,69,81,95

,96,111,141,142

7,20,52,57,72,90,146,158

14,19,39,54,69,83,84,91,102,115

SLR to PSS 9,13,50,70,73,82,88,112,118,136,138,148,

158,167

21,23,27,47,53,55,58,61,67,70,82,93,97,100,104,118,122,133,137,147,150,159,164

,169,172,186

15,22,40,42,55,65,70,85,103,110,116,127,146,158,161

SLR to GS PSS to IT 30 10 PSS to TRF 13,78,112,143 11,79,111,123,155 PSS to CRF 3,10,18,37,39,45,61,

66,74,79,91,99,103,109,113,122,125,128,134,139,145,152,155

,165

3,16,28,33,35,41,48,59,62,68,76,80,87,95,98,105,114,120,123,131,138,148,160,162

,165,170,173,184

7,16,23,43,46,49,57,63,66,73,81,87,96,108,113,117,120,125,128,131,141,1

44,147,151,159 PSS to MF 27,52,84,120,161 3 PSS to TCF 89,132,150 71,94,133 PSS to SLR 32,72,137 22,24,54,56,71,101,1

17,134,145,151 13,35,41,90,137,15

7 PSS to PSS 2,17,51,71,78,83,119

,127,131,149,159,160

2,12,86,94,119,130 2,45,56,62,86,107,136

PSS to GS 14,168 83,187 59,104,162 GS to IT 15,76 84,128 60,105 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS

307

GS to GS Study Participant

31 32 33

308

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,11,19,41,54,67,70,

77 1,8,34 1,20,28,31,39,53,57,

84,93 IT to GS TRF to IT 53 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 11 TRF to GS CRF to IT 30 CRF to TRF CRF to CRF 4,13,30,44 4,47,61,62,70,90,95 CRF to MF 13,96 CRF to TCF 9,31 11,14,17,22,28 CRF to SLR 5,7,14,16,22,24,26,3

4,38,45,49,57,60,62,72,79,81,85,87,89,93

5,22,24,33,42,48,58,63,81,87,91

CRF to PSS 55,71,75 CRF to GS 37 MF to IT MF to TRF MF to CRF 21,27 46 MF to MF MF to TCF MF to SLR 4 14,97 MF to PSS 25 MF to GS TCF to IT 10,66,69 TCF to TRF 52 TCF to CRF TCF to MF TCF to TCF TCF to SLR 16

309

TCF to PSS 32,74,83,91 12,15,18,23,29 9,18,44,51,78 TCF to GS SLR to IT 76 27,92 SLR to TRF SLR to CRF 6,8,15,23,25,37,48,5

6,59,61,80,86,88 23

SLR to MF SLR to TCF 51,65,73,82,90 8,15,17,43,50,77 SLR to SLR 27,36,46,47,50,58,63

,64 5,31 34,49,64,65,98

SLR to PSS 28 6,25,35,59,67,88,99 SLR to GS 17,39,94 6,32 37,82 PSS to IT 19,52,56 PSS to TRF 10 PSS to CRF 3,12,21,29,33,43,71,

78,84,92 10,13,16,36 3,12,21,29,32,41,54,

58,60,69,74,80,86,89,94

PSS to MF 3,20,24,26 45 PSS to TCF 68 PSS to SLR 55,75 30 7,26,36,76 PSS to PSS 2,20,42 2,9,35 2,40,68,72,73,79,85 PSS to GS 100 GS to IT 18,40 7,33 38,83 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS GS to GS Study Participant

34 35 36

310

Appendix F. (Continued) Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR IT to PSS 1,54,73,105,128 4,35,40,58,65,80,90,98

,102,124,133,143 1,52,67

IT to GS TRF to IT 34 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 116 TRF to GS CRF to IT CRF to TRF CRF to CRF 5,59,60,66,67,68,76,

77,118 10,11,120 5,96,97,102,105,10

6,118,127,128,129,130

CRF to MF 56 CRF to TCF 17,31,85,114 15,18,24,29,32,49,126,

137 6,17,34,94

CRF to SLR 20,22,27,45,69,100,110,119,121,123,131,

135,138,141,144

7,12,37,42,53,61,82,92,104,115,121,130,135,

145

13,21,23,27,29,55,71,79,83,85,88,98,

107,110,116 CRF to PSS 6,10,13,15,25,29,61,

78,80,82,89,93,95,98,108,147

47,100,118 46,48,74,103,119,131,133,135

CRF to GS 65,137 MF to IT 53 MF to TRF MF to CRF 4,44,65 4,12,70,78,126 MF to MF 125 MF to TCF 37,41,50 9,40,43,62 MF to SLR 113 MF to PSS 35,57 27 38 MF to GS TCF to IT 132,142

311

TCF to TRF 115 33,79 TCF to CRF 95,109 TCF to MF TCF to TCF TCF to SLR 21,73,84 TCF to PSS 8,18,32,38,42,48,51,

63,86,91 16,19,25,30,45,50,69,9

6,127,138 7,10,18,32,35,41,44,60,63,76,91,100,

121,123 TCF to GS 63 SLR to IT SLR to TRF SLR to CRF 21,28,120,122 9,114,117,136 20,22,28,84,117 SLR to MF SLR to TCF 44,62,68,78,83,131 99,108 SLR to SLR 101,102,132 8,43,54,55,67,74,75,76

,77,85,93,116 24,80,111,112,113

SLR to PSS 23,46,70,103,111,124,133,136,139,142,14

5

2,13,22,38,71,86,94,105,107,109,122,140

14,25,30,56,58,72,81,86,89,114

SLR to GS 56,146 50 PSS to IT 104 3,39,89,97,101,123 PSS to TRF PSS to CRF 9,12,14,16,19,24,26,

30,55,58,75,79,81,84,88,92,94,97,99,107,109,113,117,130,134,

137,140,143,146

6,14,17,23,28,31,36,41,46,48,52,60,81,91,99,103,119,125,129,134,1

44

16,26,33,45,47,54,64,73,82,87,93,101,104,115,132,134,1

36

PSS to MF 3,34,36,40,43,49,52,64

26,112 3,8,11,37,39,42,61,69,77,124

PSS to TCF 7,47,62,90 20,72,95,141 31,59,75,90,120,122

PSS to SLR 1,66,70,106,108,139 19,49,57 PSS to PSS 2,11,33,39,74,83,87,

96,106,112,125,129,148,149

5,51,59,110,111,128 2,15,36,53,68,92

PSS to GS 71,126,150 87 GS to IT 72,127 57,64 51,66 GS to TRF GS to CRF GS to MF GS to TCF GS to SLR GS to PSS 88 GS to GS Study Participant

37 38 39

312

Appendix F. (Continued)

Level of Cognitive Coordination Sequences (From —> To)

Step by Step Transition per Study Participant

IT to IT IT to TRF IT to CRF IT to MF IT to TCF IT to SLR 73,75,84 IT to PSS 1,57,79 2,36,48,62,70 2,26,37,44,56,61,6

8 IT to GS TRF to IT 36 TRF to TRF TRF to CRF TRF to MF TRF to TCF TRF to SLR TRF to PSS 31 72 TRF to GS CRF to IT 72 60 CRF to TRF CRF to CRF 4,32,63,70 19 CRF to MF 33,83 CRF to TCF 44,47 16,33 5,10,39,42,53,70,7

4,77,80 CRF to SLR 5,10,16,21,27,29,33,38,4

2,52,64,66,71,75,82,85,87,90

5,11,50,66 8,17

CRF to PSS 50 20,22,38,40,42,44,58,64

14,47,58

CRF to GS 49 MF to IT MF to TRF MF to CRF 4 4,16 MF to MF MF to TCF 8 34 MF to SLR 60 MF to PSS 66 MF to GS 84 TCF to IT 35 43 TCF to TRF 35,71 TCF to CRF TCF to MF

313

TCF to TCF 34 TCF to SLR TCF to PSS 7,14,40,45,48 9,14,17,25,52,56 6,11,40,51,54,75,78

,81 TCF to GS SLR to IT 74,83 SLR to TRF SLR to CRF 28,31,37,43,62,65,84,86,

89 9,46

SLR to MF 65 SLR to TCF 6,39 13,24,51,55 SLR to SLR 11,17,18,22,23,24,30,34,3

5,36,53,61,67,72,76,83,88,91

12,76,77,80,81,82 28,29,30

SLR to PSS 12,19,25,54,68,73,92 6,29,46,78 18,20,22,31,63 SLR to GS 77 67,85 PSS to IT 1,47,61,69 1,25,55,67 PSS to TRF 30 PSS to CRF 3,9,15,20,26,41,46,49,51,

69,74,81 10,15,18,21,32,37,39,41,43,49,57,63,65,7

1

7,13,32,38,41,48,52,57,59,69,73,76,7

9,82 PSS to MF 59 3,7 3,15 PSS to TCF 13 PSS to SLR 23,28,45,54,79 19,21,27,45,62,64 PSS to PSS 2,8,58,80 26,27,53 12 PSS to GS 55,93 59 23 GS to IT 56,78 GS to TRF GS to CRF GS to MF GS to TCF 50 GS to SLR GS to PSS 60,68 24 GS to GS Study Participant

40 41 42

314

Appendix G

Glossary

315

Glossary

The following table collects all the terms used throughout this thesis.

Terms Description Cognitive Coordination

Cognitive coordination has been defined as the management of dependencies, or conflicts, between goals, tasks, and resources of various agents and physical attributes (Du & Spink, 2009; Spink, Park, & Cole, 2006).

Cognitive Shifts Cognitive science defines a cognitive shift or shift in cognitive focus as triggered by the brain's response and change due to some external force (Jacobs, 2002). It is a human ability to handle the demands of complex and often multiple tasks resulting from changes due to external forces (Spink & Dee, 2007).

Information Cognitive Coordinating Behaviour

During information behaviour, humans cognitively coordinate a number of elements, including their cognitive state, level of domain knowledge, and understanding of their information problem, into a coherent series of activities that may include seeking, foraging, sense-making, searching, interactive browsing and retrieving, and constructing information (Spink & Du, 2007).

Interactive Information Retrieval (IR)

“The interactive communication processes that occur during retrieval of information by involving all major participants in information seeking and retrieval, i.e., the searcher, the socio-organisational context, the IT setting, interface and information space” (Ingwersen & Jarvelin, 2005, p. 21).

Information Problem

Information problem is a more specific term to describe the task the user needs to carry out during Web searching. Information problems can evolve and change over time (Spink, Park & Koshman, 2006).

Information Task Information tasks refer to the tasks that require information behaviour. “Information tasks involve conducting actions related to information seeking, organising or information use” (Spink, Albertorio, Narayanan, Brumfield & Park, 2007, p. 176).

Multitasking Multitasking is a critical human behaviour that allows people to cope with complex environments by handling multiple tasks simultaneously through task switching (Spink, Park, Jansen & Pedersen, 2006).

Multiple Web Search Sessions

Multiple Web search sessions refer to the users’ practice of submitting the entire sequence of queries through the interactions with several Web search systems over time in several windows/tabs when searching on a particular information problem.

Web Searching “Web searching can be conceptualized as a complex behaviour embedded within an individual’s everyday social, cultural, political, and information-seeking activities” (Spink & Zimmer, 2008, p. 3).

Web Search Episode

A Web search episode consists of a sequence of Web search sessions.

Web Search Session

A Web search session is a period devoted to a particular Web information problem searching.

316

Bibliography Adler, P. A., & Adler, P. (1994). Observational techniques. In N. K. Denzin & Y. S.

Lincoln (Eds.), Handbook of Qualitative Research (pp. 377-392). Thousand

Oaks, CA: Sage Publications.

Adya, K. (2005). An implicit-feedback based ranking methodology for Web search

engines. Unpublished M.S., Southern Illinois University at Carbondale, United

States -- Illinois.

Aguilar, Francis J. (1967). Scanning the business environment. New York, N.Y.:

Macmillan Co.

Al Nazer, A. A. (2006). Collaborative autonomous interface agent for personalized

Web search. Unpublished M.S., King Fahd University of Petroleum and

Minerals (Saudi Arabia), Saudi Arabia.

Allport, A., Styles, E. A., & Hsieh, S. (1994). Shifting intentional set: Exploring the

dynamic

control of tasks. In C. Umilta & M. Moscovitch (Eds.), Attention and

Performance (pp. 421-52). Cambridge, MA: MIT Press.

Amichai-Hamburger, Y. (2002). Internet and personality. Computers in Human

Behaviour, 18(1), 1-10.

Ashby, F. G., Isen, A. M., & Tucker, A. U. (1999). A neuropsychological theory of

positive effect and its influence on cognition, Psychological Review, 106(3),

529-50.

Au, A. K. (2005). Interfering effects of multitasking on muscle activity in the upper

extremity. Unpublished M.Sc., York University (Canada), Canada.

Aula, A., & Nordhausen, K. (2006). Modeling successful performance in Web

searching. Journal of the American Society for Information Science and

Technology, 57(12), 1678-1693.

317

Aurelio, D. N. (2002). The effects of Web search engine query ambiguity and

results sorting method on user performance and preference. Unpublished

Ph.D., Northeastern University, United States -- Massachusetts.

Bates, M. J. (1989). The design of browsing and berrypicking techniques for the

online search interface. On-line Review 13, 407-431.

Belkin, N., J. (1980). Anomalous states of knowledge as a basis for information

retrieval. Canadian Journal of Information Science, 5, 133-143.

Belkin, N., J. (1984). Cognitive models and information transfer. Social Science

Information Studies, 4, 111-129.

Belkin, N. J. (1996). Intelligent information retrieval: Whose intelligence? In: ISI ’96:

Proceedings of the Fifth International Symposium for Information Science (pp.

25–31). Konstanz: Universtaetsverlag Konstanz.

Belkin, N. J., Brooks, H. M., & Daniels, P. J. (1987). Knowledge elicitation using

discourse analysis. International Journal of Man-Machine Studies, 27, 127-

144.

Belkin, N. J., & Vickery, A. (1985). Information in Information Retrieval Systems: A

Review of Research from Document Retrieval to Knowledge-Based Systems.

London: British Library.

Blume, A., & Gneezy, U. (2000). An experimental investigation of optimal learning in

coordination games. Journal of Economic Theory, 90(1), 161-172.

Brumby, D. P., Salvucci, D. D., & Howes, A. (2007). An empirical investigation into

dual-task trade-offs while driving and dialing. In D. Ramduny-Ellis & D.

Rachovides (Eds.), Volume 2 HCI 2007: Proceedings of the 21st British

Computer Society HCI Group Conference. Swindon, UK: British Computer

Society.

318

Burgess, P. W. (2000). Real-world multitasking from a cognitive neuroscience

perspective. In S. M. J. Driver (Ed.), Control of Cognitive Processes: Attention

and Performance XVIII (pp. 465-472). Cambridge, MA: MIT Press.

Busch, D., & Strauss, B. (2005). Qualitative differences in performing coordination

tasks. Measurement in Physical Education and Exercise Science, 9(3), 161-

180.

Bystrom, K. (1999). Task complexity, information types and information sources:

Examination of relationships. Unpublished Ph.D., University of Tampere,

Finland -- Tampere.

Carlson, R. A., Sohn, M. Y. (2000). Cognitive control of multistep routines:

Information processing and conscious intentions. In S. M. J. Driver (Ed.),

Control of Cognitive Processes: Attention and Performance XVIII (pp. 443-

464). Cambridge, MA: MIT Press.

Case, D. O. (2002). Looking for information: A survey of research on information

seeking, needs and behavior. Amsterdam: Academic Press.

Catledge, L., & Pitkow, J. (1995). Characterizing browsing strategies in the World

Wide Web. Proceedings of the 3rd International World Wide Web Conference

on Technology, Tools and Applications, Darmstadt, Germany. Retrieved

October 1, 2007, from http://www.viktoria.se/~dixi/BISON/resources/catledge-

pitkow%201995.pdf.

Cheng, S-H. (1998). Effects of cognition-emotion processes on shifts in conflict

management strategies. Unpublished Ph.D., University of Minnesota, United

States -- Minnesota.

Choo, C., Detlor, B., & Turnbull, D. (2000a). Web work: Information seeking and

knowledge work on the World Wide Web. Boston: Kluwer Academic

Publishers.

319

Choo, C., Detlor, B., & Turnbull, D. (2000b). Information seeking on the Web: An

integrated model of browsing and searching. Retrieved January 14, 2008,

from

http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/

80/16/15/6.pdf.

Choo, C., & Marton, C. (2003). Information seeking on the web by women in IT

professions. Internet Research: Electronic Networking Applications and Policy,

13(4), 267-280.

Clark, H. H. (1985). Language use and language users. In G. L. E. Aronson (Ed.),

Handbook of Social Psychology (II): Special Fields and Applications (pp. 179-

232). New York: Random House.

Clark, H. H. (1992). Arenas of language use. Chicago: The University of Chicago.

Clark, H. H. (1996). Using language. Cambridge: Cambridge University.

Clark, H. H. & Carlson, T. B. (1981). Context for comprehension. In J. Long & A.

Baddeley (Eds.) Attention and Performance IX. Hillsdale, NJ: Erlbaum.

Clark, H. H. & Marshall, C. R. (1981). Definite reference and mutual knowledge. In

A. K. Joshe, B. Webber, & A. Sag (Eds.) Elements of Discourse

Understanding. (pp. 96-128). Cambridge: Cambridge University Press.

Creswell, J. W. (1994). Research design: Qualitative & quantitative approaches.

Thousand Oaks, CA: Sage Publications.

Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed

methods approaches (2nd ed.). Thousand Oaks, CA: Sage Publications.

Crowston, K. (1997). A coordination theory approach to organizational process

design. Organization Science, 8(2), 157-176.

Dervin, B. (1992). From the mind's eye of the 'user': The sense-making qualitative-

quantitative methodology. In J. D. Glazier & R. R. Powell (Eds.), Qualitative

Research in Information Management (pp. 61-84). Englewood, CO: Libraries

Unlimited.

320

Dervin, B., & Nilan, M. (1986). Information needs and uses. In M. Williams (Ed.),

Annual Review of Information Science and Technology, 21, 3-33.

Du, J. T. & Spink, A. (2009). Web searching model: Integrating multitasking,

cognitive coordination and cognitive shifts. In ASIST’09: Annual Meeting of

the American Society for Information Science and Technology, November 6–

11, Vancouver, BC, Canada.

Dudziak, K. Q. (2000). The psychological fidelity of Web search engines.

Unpublished M. A., Rice University, United States -- Texas.

Diaper, D. & Stanton, N. (Eds.) (2003). The Handbook of Task Analysis for Human-

Computer Interaction. Hillside, NJ: Lawrence Erlbaum Associates.

Ellis, D. (1989). A behavioural approach to information retrieval design. Journal of

Documentation, 45(3), 171–212.

Ellis, D. (1993). Modeling the information-seeking patterns of academic researchers:

A grounded theory approach. Library Quarterly, 63(4), 469-486.

Ellis, D., Cox, D., & Hall, K. (1993). A comparison of the information seeking

patterns of researchers in the physical and social sciences. Journal of

Documentation, 49(4), 356–369.

Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis: Verbal report as data.

Cambridge, MA: The MIT Press.

Farand, L. (1996). Cognitive multitasking in situated medical reasoning.

Unpublished Ph.D., McGill University (Canada), Canada.

Fidel, R. (1985). Moves in online searching. Online Review, 9(1), 61-74.

Foehr, U. G. (2006). Media multitasking among American youth: Prevalence,

pairings and predictors. Unpublished Ph.D., Stanford University, United States

-- California.

Ford, N. (2000). Improving the darkness to light ratio in information retrieval

research. Journal of Documentation, 56(6), 624-643.

321

Ford, N., & Miller, D. (1996). Gender differences in Internet perception and use.

Paper presented at the Third ELVIRA Conference on Electronic Library and

Visual Information Research (ASLIB), April-May, London, UK.

Ford, N., Miller, D., & Moss, N. (2001). The role of individual differences in Internet

searching: An empirical study. Journal of the American Society for Information

Science and Technology, 52(12), 1049-1066.

Ford, N., Miller, D., & Moss, N. (2002). Web search strategies and retrieval

effectiveness: An empirical study. Journal of Documentation, 58(1), 30-48.

Ford, N., Miller, D., & Moss, N. (2003). Web search strategies and approaches to

study. Journal of the American Society for Information Science and

Technology, 54(6), 473-489.

Ford, N., Miller, D., & Moss, N. (2005). Web search strategies and human individual

differences: Cognitive and demographic factors, Internet attitudes, and

approaches. Journal of the American Society for Information Science and

Technology, 56(7), 741-756.

Ford, N., Wilson, T. D., Foster, A., Ellis, D., & Spink, A. (2002). Information seeking

and mediated searching. Part 4. Cognitive styles in information seeking.

Journal of the American Society for Information Science and Technology,

53(9), 728-735.

Ford, N., Wood, F., & Walsh, C. (1994). Cognitive styles and searching. Online &

CD-ROM Review, 18(2), 79-86.

Freitas, P. B., Krishnan, V., & Jaric, S. (2007). Force coordination in static

manipulation tasks: effects of the change in direction and handedness. Exp

Brain Res, 183, 487-497.

Fuse, A. (2006). Flexible coordination of spatial cognition and language.

Unpublished Ph.D., City University of New York, United States -- New York.

322

Gay, L. R., & Airasian, P. (2003). Educational research: Competencies for analysis

and applications. Upper Saddle River, NJ: Prentice Hall.

Glass, J. M. B. (1996). Aging and multiple-task-performance: Task-coordination

strategies in young and old Adults. Unpublished Ph.D., University of Michigan,

United States -- Michigan.

Glasser, B. G. & Strauss, A. L. (1967). The discovery of grounded theory. New York,

NY: Aldine de Gruyter.

González, V. M., & Mark, G. (2004). Constant, constant, multi-tasking craziness:

Managing multiple working spheres. Proceedings of the SIGCHI Conference

on Human Factors in Computing Systems (pp. 113-120). New York, USA:

ACM.

Hawk, W. B., & Wang, P. (1999). Users’ interaction with the World Wide Web:

Problems and problem solving. In L. Woods (Ed.). Proceedings of the 62nd

American Society for Information Science Annual Meeting (pp. 256-270).

Medford, NJ: Information Today.

Heinström, J. (2000). The impact of personality and approaches to learning on

information behavior. Information Research, 5(3). Retrieved February 28,

2008, from http://informationr.net/ir/5-3/paper78.html.

Heinström, J. (2003). Five personality dimensions and their influence on

information behaviour. Information Research, 9(1). Retrieved August 27, 2007,

from http://informationr.net/ir/9-1/paper165.html.

Hider, P. M. (2007). Constructing an index of search goal redefinition through

transaction log analysis. Journal of Documentation, 63(2), 175-187.

Hill, J. R., & Hannafin, M. J. (1997). Cognitive strategies and learning from the

World Wide Web. Educational Technology Research and Development, 45(4),

37-64.

323

Hochstadt, J. F. (2004). From gaze shifts to set shifts: Using eye-tracking during

sentence-picture matching to link deficits in language comprehension and

cognition in parkinson's disease. Unpublished Ph.D., Brown University, United

States -- Rhode Island.

Hodkindon, C., & Kiel, G. (2003). Understanding web information search behaviour:

An exploratory model. Journal of End User Computing, 15(4), 27-48.

Hoelscher, C., & Strube, G. (2000). Web search behaviour of Internet experts and

newbies. In H. Maurer & R.G. Olson (Eds.), Proceedings of the 9th

International World Wide Web Conference (pp. 337-346). New York, USA:

Elsevier North-Holland, Inc.

Holsti, O. R. (1969). Content analysis for the social sciences and humanities.

Reading, MA: Addison-Wesley.

Horng, H-C. (1996). Optimizing the multitasking of workers in just-in-time systems.

Unpublished Ph.D., Arizona State University, United States -- Arizona.

Huang, M. H. (1992). Pausing behavior of end-users in online searching.

Unpublished Ph.D., The University of Maryland, United States.

Hunt, E., Joslyn, S. (2000). A functional task analysis of time-pressured decision

making. In S. F. C. J. M. Schraagen, & V. L. Shalin (Ed.), Cognitive Task

Analysis. (pp. 119-134). Mahwah, NJ: Erlbaum.

Iani, C., & Wickens, C. D. (2004). Factors affecting task management in aviation.

Moffett Field, C.A.: NASA AMES Research Center.

Ingwersen, P. (1992). Information retrieval interaction. London: Taylor Graham.

Ingwersen, P. (1996). Cognitive perspective of information retrieval interaction:

Elements of a cognitive IR theory. Journal of Documentation, 52(11), 3-50.

Ingwersen, P. (2001). Cognitive information retrieval. In: M. Williams (Ed.), Annual

Review of Information Science and Technology, 34, 3-51.

Ingwersen, P., & Jarvelin, K. (2005). The turn: Integration of information seeking

and retrieval in context. Dordrecht: Springer.

324

Jansen, B. J., & Pooch, U. (2001). A review of web searching studies and a

framework for future research. Journal of the American Society for Information

Science, 52(3), 235-246.

Jong, H-S. (1991). The sub-goal structure as a cognitive control mechanism in a

human-computer interaction framework. Unpublished Ph.D., University of

Michigan, United States -- Michigan.

Ju, S. (1988). A Real Time Multitasking Kernel for the IBM Personal Computer.

Unpublished M.S., The University of Arizona, United States -- Arizona.

Judt, C. P. (1988). Development of a multitasking control environment for a flexible

manufacturing system. Unpublished M.Sc., The University of Manitoba

(Canada), Canada.

Just, M. A., Carpenter, P. A., Keller, T. A., Emery, L., Zajac, H., Thulborn, K. R.

(2001). Interdependence of non-overlapping cortical systems in dual cognitive

tasks. Neuroimage, 14, 417-426.

Katzeff, C. (1989). Cognitive aspects of human-computer interaction: Mental

models in database query writing. Unpublished Fil.Dr., Stockholms Universitet

(Sweden), Sweden.

Keeling, K. R. (1973). Hypnosis and adaptive regression: A study using wild's

cognitive shift measure. Unpublished Ph.D., University of Waterloo (Canada),

Canada.

Kellogg, W. A., & Richards, J. T. (1995). The human factors of information on the

Internet. In J. Neilsen (Ed.), Advances in Human-Computer Interaction, 5, 1-

36. Norwood, NJ: Ablex.

Kim, K-S. (2008). Effects of emotion control and task on web searching behavior.

Information Processing & Management, 44(1), 373-385.

Kim, K-S., & Allen, B. (2002). Cognitive and task influences on web searching

behavior. Journal of the American Society for Information Science and

Technology, 53(2), 109-119.

325

Kling, R., Kraemer, K. L., Allen, J. P., Bakros, Y., Gurbaxani, V., & Elliott, M. (2001).

Transforming coordination: The promise and problems of information

technology in coordination. In G. M. Olsen, T. W. Malone & J. B. Smith (Eds.),

Coordination Theory and Collaboration Technology (pp. 507-534). Mahwah,

NJ: Lawrence Erlbaum Associates.

Knight, S. A. & Spink, A. (2008). Toward a Web search information behavior model.

In A. Spink & M. Zimmer (Eds.), Web Searching: Interdisciplinary

Perspectives (pp. 209-234). Springer.

Koshman, S., Spink, A., & Jansen, B. J. (2006). Web searching on the Vivisimo

search engine. Journal of the American Society for Information Science and

Technology, 57(14), 1875-1887.

Krippendorf, K. (1980). Content analysis: An introduction to its methodology.

Beverly Hills: CA: Sage Publications.

Krathwohl, D. R. (1997). Methods of educational and social science research: An

integrated approach. New York: Longman.

Kuhlthau, C. C. (1993). Seeking meaning: A process approach to library and

information services. Norwood, NJ: Ablex Publishing.

Kuhlthau, C. C. (2004). Seeking meaning: A process approach to library and

information services (2nd ed.). Westport., Connecticut: Libraries Unlimited.

Kuhlthau, C. C., Spink, A., & Cool, C. (1992). Exploration into stages in the

information search process in online information retrieval: Communication

between users and intermediaries. Proceedings of the 55th Annual Meeting of

the American Society for Information Science, 29th October, Pittsburgh, PA

(pp. 67-71).

Kushleyeva, Y., Salvucci, D. D., & Lee, F. J. (2005). Deciding when to switch tasks

in time-critical multitasking. Cognitive Systems Research, 6, 41-49.

326

Lau, A. Y. S., & Coiera, E. W. (2006). A Bayesian model that predicts the impact of

Web searching on decision making. Journal of the American Society for

Information Science and Technology, 57(7), 873-880.

Lee, F. J., & Taatgen, N. A. (2002). Multitasking as skill acquisition. In W. D. Gray &

C. D. Schunn (Eds.), Proceedings of the 24th Annual Conference of the

Cognitive Science Society (pp. 572-577). Hillsdale, NJ: Lawrence Erlbaum

Associates.

Lieberman, M. D. (1998). Why introverts can't always tell who likes them: Social

multitasking and nonverbal decoding. Unpublished Ph.D., Harvard University,

United States -- Massachusetts.

Lincoln, Y.S., & Guba, E.G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage

Publications.

Long, X. (2006). Efficient query processing in large Web search engines.

Unpublished Ph.D., Polytechnic University, United States -- New York.

Ma, H, Y. (2008). User-system coordination in unified probabilistic retrieval:

Exploiting search logs to construct common ground. Unpublished Ph.D.,

University of California, United States -- Los Angeles.

Malone, T. W., & Crowston, K. (1994). The interdisciplinary study of coordination.

ACM Computing Surveys, 26(1), 87-119.

Martin, P. Y., & Turner, B. A. (1986). Grounded Theory and Organizational Research.

The Journal of Applied Behavioral Science, 22 (2), 141-157.

Meyer, D. E., Kieras, D. E., Lauber, E., Schumacher, E., Glass, J., Zurbriggen, E.,

Gmeindl, L., & Apfelblat, D. (1995). Adaptive executive control: Flexible

multiple-task performance without pervasive immutable response-selection

bottlenecks. Acta Psychologica, 90, 163-190.

Meyer, R. E. (1977). Thinking and Problem Solving: An Introduction to Human

Cognition and Learning. Glenview, IL: Scott Foresman.

327

Miyata, Y., & Norman, D. A. (1986). Psychological issues in support of multiple

activities. In D. A. Norman & S. W. Draper (Eds.), User Centered Design (pp.

265-284) NJ: Lawrence Erlbaum.

Monsell, S. (2003). Task switching. Trends in Cognitive Sciences, 7, 134-140.

Montgomery, A., & Faloutsos, C. (2001). Identifying Web browsing trends and

patterns. IEEE Computer, 34(7), 94-95.

Morgan, A. (2008). An analysis of the influence of human individual differences on

web searching behavior among Blacks and Whites: A case of health

information searching. Unpublished Ph.D., The Pennsylvania State University,

United States -- Pennsylvania.

Navarro-Prieto, R., Scaife, M., & Rogers, Y. (1999). Cognitive Strategies in Web

Searching. Retrieved October 1, 2007, from

http://eidetic.ai.ru.nl/egon/education/cogerg/literature/usability/Navarro-

Prieto99-Cognitive_strategies_web_searching.html.

Nguyen, Q. T. (2006). Using the genetic algorithm to optimize Web search: Lessons

from biology. Unpublished M.Sc., University of Ottawa (Canada), Canada.

Ozmutlu, S., Ozmutlu, H. C., & Spink, A. (2003a). Multitasking web searching:

Implications for design. ASIST’03: Annual Meeting of the American Society

for Information Science and Technology, October 18–22, Long Beach CA.

Ozmutlu, S., Ozmutlu, H. C., & Spink, A. (2003b). A study of multitasking Web

searching. In IEEE ITCC’03: International Conference on Information

Technology: Coding and Computing, April 28–30, Las Vegas, NV.

Palmquist, R. A., & Kim, K. S. (2000). Cognitive style and on-line database search

experience as predictors of Web search performance. Journal of the

American Society for Information Science, 51(6), 558–566.

Park, M. (2008). Human prioritizing and coordinating information behavior during

information seeking and retrieval in the Web environment. Unpublished Ph.D.,

University of Pittsburgh, United States -- Pennsylvania.

328

Parolia, N., Goodman, S., Li, Y., & Jiang, J. J. (2007). Mediators between

coordination and IS project performance. Information & Management, 44(7),

635-645.

Pashler, H. (2000). Task switching and multitask performance (tutorial). In S. M. J.

Driver (Ed.), Control of Cognitive Processes: Attention and Performance XVIII

(pp. 277-309). Cambridge, MA: MIT Press.

Pesante-Santana, J. A. (1997). The effects of multitasking on quality inspection in

advanced manufacturing systems. Unpublished Ph.D., Virginia Polytechnic

Institute and State University, United States -- Virginia.

Pezas, N. J. (1992). A real-time, multitasking approach to automatic monitoring and

control of industrial processes. Unpublished M.S., California State University,

Long Beach, United States -- California.

Phillips, W. A., & Silverstein, S. M. (2003). Convergence of biological and

psychological perspectives on cognitive coordination in schizophrenia.

Behavioral and Brain Sciences, 26(1), 65-82.

Rapoport, A., & Fuller, M. A. (1998). Coordination in noncooperative three-person

games under different information infrastructures. Group Decision and

Negotiation, 7(4), 363-382.

Rieh, S. Y., & Xie, H. (2001). Patterns and sequences of multiple query

reformulations in Web searching: A preliminary study. In E. Aversa & C.

Manley (Eds.), Proceedings of the 64th American Society for Information

Science Annual Meeting (pp. 246-255). Medford, NJ: Information Today.

Rieh, S. Y., & Xie, H. I. (2006). Analysis of multiple query reformulations on the web:

The interactive information retrieval context. Information Processing &

Management, 42(3), 751-768.

Robins, D. (2000). Shifts of focus on various aspects of user information problems

during interactive information retrieval. Journal of the American Society for

Information Science, 51(10), 913-928.

329

Rose, D. E. (2006). Reconciling information-seeking behavior with search user

interfaces for the Web. Journal of the American Society for Information

Science and Technology, 57(6), 797-799.

Roy, M., Taylor, R., & Chi, M. T. H. (2004). Searching for information on-line and off-

line: Gender differences among middle school students. Journal of

Educational Computing Research, 29(2), 229-252.

Rubinstein, J. S., Meyer, D. E. & Evans, J. E. (2001). Executive control of cognitive

processes in task switching. Journal of Experimental Psychology: Human

Perception and Performance, 27(4), 763-797.

Ruthven, I. (2006). Interactive information retrieval. In: B. Cronin (Ed.), Annual

Review of Information Science and Technology, 42, 43-91. Medford, N.J:

Information Today.

Salvucci, D. D., Taatgen, N. A., & Kushleyeva, Y. (2006). Learning when to switch

tasks in a dynamic multitasking environment. In D. Fum, F. D. Missier & A.

Stocco (Eds.), Proceedings of the Seventh International Conference on

Cognitive Modeling (pp. 268-273) Trieste, Italy: Edizioni Goliardiche.

Sangster, C. A., Beninger, C., Polatajko, H. J., & Mandich, A. (2005). Cognitive

strategy generation in children with developmental coordination disorder. The

Canadian Journal of Occupational Therapy, 72(2), 67-77.

Santon, J. (2003). Shifts between search stages in mediated information seeking

interaction. Unpublished Ph.D., Rutgers The State University of New Jersey -

New Brunswick, United States -- New Jersey.

Saracevic, T. (1996). Modeling interaction in information retrieval (IR): A review and

proposal. Proceedings of the 59th Annual Meeting of the American Society for

Information Science, October 21-24, Baltimore, MD.

Saracevic, T., Kantor. P. (1988) A study of information seeking and retrieving. III.

Searchers, searches and overlap. Journal of the American Society for

Information Science, 39 (3), 197-216.

330

Saracevic, T., Kantor, P., Chamis, A. & Trivison, D. (1988). A study of information

seeking and retrieving. Ⅰ. Background and methodology. Journal of the

American Society for Information Science, 39(3), 161-176.

Saracevic, T., Mokros, H., & Su, L. (1990). Nature of interaction between users and

intermediaries in online searching: A qualitative analysis. In: Henderson, D.

(Ed.), Proceedings of the 53rd American Society for Information Science

Annual Meeting (pp. 47-54). Medford, NJ: Learned Information.

Sarmah, S. P., Acharya, D., & Goyal, S. K. (2007). Coordination and profit sharing

between a manufacturer and a buyer with target profit under credit option.

European Journal of Operational Research, 182(3), 1469-1478.

Schamber, L. (2000). Time-line interviews and inductive content analysis: Their

effectiveness for exploring cognitive behaviours. Journal of the American

Society for Information Science, 51(8), 734-744.

Simon, H. A. (1981). Information-processing models of cognition. Journal of the

American Society for Information Science, 32(5), 364-377.

Slone, D. J. (2007). The impact of time constraints on internet and web use. Journal

of the American Society for Information Science and Technology, 58(4), 508-

517.

Smith, L. A. (2006). Think-aloud mysteries: Using structured, sentence-by-sentence

text passages to teach comprehension strategies. The Reading Teacher,

59(8), 764-773.

Spink, A. (1996). Multiple search sessions model of end-user behavior: An

exploratory study. Journal of the American Society for Information Science

and Technology, 47(8), 603-609.

Spink, A. (1997). Study of interactive feedback during mediated information retrieval.

Journal of the American Society for Information Science, 48(5), 382-394.

331

Spink, A. (2002). A user-centered approach to evaluating human interaction with

Web search engines: An exploratory study. Information Processing &

Management, 38(3), 401-426.

Spink, A. (2004). Multitasking information behavior and information task switching:

an exploratory study. Journal of Documentation, 60(3), 336-345.

Spink, A., Alvarado-Albertorio, F., Narayanan, B., Brumfield, J., & Park, M. (2007).

Multitasking information behavior in public libraries: A survey study. Journal of

Librarianship and Information Science, 39(3), 177-186.

Spink, A., Bateman, J., & Greisdorf, H. (1999). Successive searching behavior

during mediated information seeking: An exploratory study. Journal of

Information Science, 25(6), 439-449.

Spink, A., & Cole, C. (2004). A human information behavior approach to a

philosophy of information. Library Trends, 52(3), 617-628.

Spink, A., & Cole, C. (2005). A multitasking framework for cognitive information

retrieval. In A. Spink & C. Cole (Eds.), New Directions in Cognitive Information

Retrieval (pp. 99-112). Dordrecht: Springer.

Spink, A., Cole, C., & Waller, M. (2008). Multitasking behavior. In B. Cronin (Ed.),

Annual Review of Information Science and Technology, 42, 93-118. Medford,

N.J: Information Today.

Spink, A., & Dee, C. (2007). Cognitive shifts related to interactive information

retrieval. Online Information Review, 31(6), 845-860.

Spink, A., & Du, J. (2007). Information behavior as cognitive shifting, cognitive

coordination and multitasking. RAILS 2007: Seminar of Research Applications

in Information and Library Studies, November 30, Melbourne, Australia.

Spink, A., & Jansen, B. J. (2004). Web search: Public searching of the Web.

Dordrecht: Kluwer Academic Publishers.

332

Spink, A., Ozmutlu, H. C., & Ozmutlu, S. (2002). Multitasking information seeking

and searching processes. Journal of the American Society for Information

Science and Technology, 53(8), 639-652.

Spink, A., & Park, M. (2005). Information and non-information task interplay. Journal

of Documentation, 61(4), 548-554.

Spink, A., Park, M., & Cole, C. (2006). Multitasking and coordinating framework for

human information behavior. In A. Spink & C. Cole (Eds.), New Directions in

Human Information behavior (pp. 137-154). Dordrecht: Springer.

Spink, A., Park, M., Jansen, B. J., & Pedersen, J. (2006). Multitasking during Web

search sessions. Information Processing & Management, 42(1), 264-275.

Spink, A., Park, M., & Koshman, S. (2006). Factors affecting assigned information

problem ordering during Web search: An exploratory study. Information

Processing & Management, 42(5), 1366-1378.

Spink, A., Wilson, T. D., Ford, N. A., Foster, A., & Ellis, D. (2002a). Information

seeking and mediated searching. Part 1. Background and research design.

Journal of the American Society for Information Science and Technology,

53(9), 695–703.

Spink, A., Wilson, T. D., Ford, N. A., Foster, A., & Ellis, D. (2002b). Information

seeking and mediated searching study. Part 3. Successive searching. Journal

of the American Society for Information Science and Technology, 53(9), 716-

727.

Sran, B. S. (1999). Exploring Web search strategies in technical environments.

Unpublished Ph.D., Stevens Institute of Technology, United States -- New

Jersey.

Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory

procedures and techniques. Newbury Park, CA: Sage Publications.

333

Swamy, S. (2005). Multitasking on social networks: Role of culture, status and

group membership in computer-mediated communication. Unpublished Ph.D.,

Stanford University, United States -- California.

Tao, C.-C. (2006). Cognitive processing during Web search: The role of working

memory load in selective attention and inhibitory control. Unpublished Ph.D.,

Indiana University, United States -- Indiana.

Tauscher, L., & Greenberg, S. (1997). How people revisit Web pages: Empirical

findings and implications for the design of history system. International

Journal of Human-Computer Interaction, 47(1), 97-138.

Thompson, S. (1996). Paying respondents and informants. Social Research Update,

14. Retrieved May 11, 2008, from http://sru.soc.surrey.ac.uk/SRU14.html.

Tillotson, J., Cherry, J. M., & Clinton, M. (1995). Internet use through the university

of Toronto library: Demographics, destinations and user’s reactions.

Information Technology and Libraries, 14(3), 190-198.

Waller, M. J. (1995). Multitasking in work groups: Coordination processes in work

groups with multiple tasks. Unpublished Ph.D., The University of Texas at

Austin, United States -- Texas.

Waller, M. J. (1997). Keeping the pins in the air: How work groups juggle multiple

tasks. In M. M. Beyerlein, D. A. Johnson, & S. T. Beyerlein (Eds.),

Advances in interdisciplinary studies of work teams, Vol. 4: Team

implementation issues (pp. 217-247). Greenwich, CT: JAI.

Wang, P., Hawk, W. B., & Tenopir, C. (2000). Users’ interaction with World Wide

Web resources: An exploratory study using a holistic approach. Information

Processing & Management, 36(2), 229–251.

Weber, R. P. (1990). Basic content analysis. Newbury Park, CA: Sage.

White, M. D., & Iivonen, M. (2002). Assessing level of difficulty in Web search

strategies. Library Quarterly, 72(2), 205-233.

334

Wickens, C. D. (1989). Attention and skilled performance. In D. Holding (Ed.),

Human Skills (pp. 71-105). New York: Wiley & Sons.

Wickens, C. D. (1991). Processing resources and attention. In D. L. Damos (Ed.),

Multiple-task performance (pp. 3-34). London: Taylor & Francis.

Wickens, C. D., & Gopher, D. (1977). Control theory measures of tracking as

indices of attention allocation strategies. Human Factors, 19(4), 349-365.

Wilson, T. D., Ford, N. J., Ellis, D., Foster, A., & Spink, A. (2002). Information

seeking and mediated searching: Part 2. Uncertainty and its correlates.

Journal of the American Society for Information Science and Technology,

53(9), 704-715.

Xie, H. (2000). Shifts of interactive intentions and information-seeking strategies in

interactive information retrieval. Journal of the American Society for

Information Science, 51(9), 841-857.

Yang, S. (1997). Information seeking as problem-solving using a qualitative

approach to uncover the novice learner’s information-seeking process in a

Perseus hypertext system. Library & Information Science Research, 19(1),

71-92.

Yeung, N., Nystrom, L. E., Aronson, J. A., & Cohen, J. D. (2006). Between-task

competition and cognitive control in task switching. The Journal of

Neuroscience, 26(5), 1429-1438.

Zhang, L. (2006a). Contextual Web search based on semantic relationships: A

theoretical framework, evaluation and a medical application prototype.

Unpublished Ph.D., The University of Arizona, United States -- Arizona.

Zhang, Y. (2006b). The effects of monochronicity and polychronicity on multitasking

strategy and performance. Unpublished Ph.D., Hong Kong University of

Science and Technology (People's Republic of China), Peoples Republic of

China.

335

Zhang, Y., & Wildemuth, B. M. (2009). Qualitative analysis of content. In B.

Wildemuth (Ed.), Applications of Social Research Methods to Questions in

Information and Library. Retrieved April 22, 2010, from

http://www.ischool.utexas.edu/~yanz/Content_analysis.pdf

336

Curriculum Vita Jia Tina Du

Education Ph.D. Information Science. Thesis title: Multitasking, Cognitive

Coordination and Cognitive Shifts During Web Searching. Faculty of Science and Technology, Queensland University of Technology, Brisbane, Australia. 2010.

M.L.I.S Information Science. Thesis title: The Research of Evaluation on Search Engine. Department of Information Management, Nanjing University, Nanjing, China. 2006.

B.S. (Honours) Information Management and Information System. Thesis title: The Application of Information Architecture (IA) in Websites Architecting and Evaluation. Department of Information Management, Nanjing University, Nanjing, China. 2004.

Professional Experience 2007-10 Research Assistant. Professor Amanda Spink. Faculty of Science and

Technology. Queensland University of Technology. Brisbane, Australia. 2004-07 Research Assistant. Professor Qinghua Zhu. Department of Information

Management. Nanjing University. Nanjing, China.

Publications Du, J.T. (2010). Multitasking, cognitive coordination and cognitive shifts during Web searching. Unpublished Ph.D., Queensland University of Technology, Australia -- Queensland. Du, J.T. (2010). The role of cognitive coordination in Web searching. RAILS 2010: Seminar of Research Applications in Information and Library Studies, January, Canberra, Australia. Du, J.T., & Spink, A. (2009). Modeling Web search: Preliminary results. In Proceedings of the 72nd Annual Meeting of the American Society for Information Science and Technology (ASIS&T), 46, Vancouver, BC, Canada. Du, J.T., & Spink, A. (2009). Cognitive web search model as multitasking, cognitive coordination and cognitive shifts. In ASCS 2009: The 9th Conference of the Australasian Society for Cognitive Science, September 30–October 2, Macquarie University, Sydney, Australia.

337

Du, J. T. (2009). Book review of " Personalized Information Retrieval and Access: Concepts, Methods, and Practices" R. A. Gonzalez, N. Chen & A. Dahanayake (Ed.). Online Information Review, 33(1), 211-212. Du, J. T., & Spink, A. (2008). Web searching model: Integrating multitasking, cognitive coordination and cognitive shifts. In Proceedings of the 71st Annual Meeting of the American Society for Information Science and Technology (ASIS&T), 45, (pp. 416-418). Columbus, USA. Spink, A., Alvarado-Albertorio, F., & Du, J. T. (2008). Beyond relevance judgments: Cognitive shifts and gratification. In Beyond Binary Relevance Workshop. ACM SIGIR2008: 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, July 29, Singapore. Liu, Y. H., Qi, A. H., Du, J., Zhu, Q. H., Huang, Q., & Yue, Q. (2008). The establishment and application of evaluation criteria systems for academic Websites. Information Science, 26(1), 64-68. Zhu, Q. H., & Du, J. (2007).The establishment and application of evaluation criteria system for search engines. Journal of the China Society for Scientific and Technical Information, 26(5), 684-690. Zhu, Q. H., & Du, J. (2007). The research of evaluation on government Websites in China and overseas. E-Government, 7, 31-39. Zhu, Q. H., Han, X. J., Du, J., & Qi, A.H. (2007). The establishment and application of evaluation criteria systems for Chinese e-government Websites. Library and Information Service, 51(11), 67-70. Spink, A., & Du, J. (2007). Information behaviour as cognitive shifting, cognitive coordination and multitasking. RAILS 2007: Seminar of Research Applications in Information and Library Studies, November 30, Melbourne, Australia. Zhu, Q. H., Du, J., & Han, X. J. (2007). The establishment and application of evaluation criteria systems for Chinese e-government Websites. In Management Track within WiCOM: 2007 International Symposium on Information System & Management, (pp. 3783-3786), 26-28July, Shanghai, China. Du, J. (2006). Quantitative analysis on Enterprise Resource Planning (ERP) dissertations from the database of PQDD. Information Science, 24(5), 748-753. Zhu, Q. H., & Du, J. (2006). The influence of information disclosure system on the governmental information resource management. Information Studies: Theory & Application, 29(2), 150-152,226. Du, J., & Zhu, Q. H. (2004). Application of information architecture (IA) on website evaluation. Information and Documentation Services, 6, 3-16. Wang, L. J., & Du, J. (2004). Discussion on fostering mode of graduate students in top universities abroad. Academia Bimestris, 5, 172-174.

338

Teaching Experience Queensland University of Technology (QUT), Australia Guest Lecturer 2009 Postgraduate Course INN540 Information and User Behaviour Teaching Assistant/Tutor 2009 Undergraduate Courses INB302 Capstone Project INB272 Interaction Design 2008 Undergraduate Courses ITB254 Interaction Design ITB009 IT Project Management Project Supervisor 2009 Project One: Centralised Complaint Management System (12 weeks) Project Two: Game Design (12 weeks) Invited Reviewer • Online Information Review • ITNG 2009: The 6th International Conference on Information Technology - New

Generations, Las Vegas, Nevada, USA, April 27-29, 2009. Awards & Honors 2010 Postgraduate Research Write Up Scholarship, QUT, Australia 2010 ALIA RAILS Award for Research Presentation, Australia 2009 Postgraduate Grant-in-Aid Scholarship, QUT, Australia 2008 NGS08 Full Travel Grant, Australia 2008 Joint HCSNet-EII Full Travel Grant, Australia 2006 Excellent Graduate Student Scholarship, Nanjing University, China 2005 Excellent Graduate Student Leader, Nanjing University 2003 Excellent Student, Nanjing University 2001-03 Excellent Undergraduate Scholarship (First-rank Prize), Nanjing

University Professional Memberships 2009+ Australian Library and Information Association (ALIA), Australia. 2009+ Association for Library and Information Science Education (ALISE), USA. 2007+ American Society for Information Science and Technology (ASIS&T), USA. 2007+ ASIST SIG/USE (Information Needs, Seeking and Use), USA. 2004+ National Center for Information Resource Management, Nanjing, China.

339