Developing Visual Narrative: Designing Structural ...4258... · Developing Visual Narrative:...

64
Developing Visual Narrative: Designing Structural Composition Principles to Guide Player’s Attention in Adventure Games A Thesis Submitted to the Faculty of Drexel University by Glenn Joseph Winters in partial fulfillment of the requirements for the degree of Master of Science in Digital Media 2013

Transcript of Developing Visual Narrative: Designing Structural ...4258... · Developing Visual Narrative:...

Developing Visual Narrative: Designing Structural Composition Principles to

Guide Player’s Attention in Adventure Games

A Thesis

Submitted to the Faculty

of

Drexel University

by

Glenn Joseph Winters

in partial fulfillment of the

requirements for the degree

of

Master of Science in Digital Media

2013

c© Copyright 2013Glenn Joseph Winters.

This work is licensed under the terms of the Creative CommonsAttribution-ShareAlike license. The license is available at http://creativecommons.org/licenses/by-sa/2.0/.

ii

Dedications

This thesis is dedicated to my mom and dad, Becky and Don. Without your constant

support, I would not have made it as far as I have. Thank you for always teaching me

“There is a great big beautiful tomorrow, just a dream away”

iii

Acknowledgements

I would like to thank my thesis adviser Dr. Jichen Zhu, for all the support and construc-

tive feedback you have provided during this journey. Your passion for game research has

provided me with a new outlook on video games. I would also like to thank my commit-

tee members, Frank deSantis, and Garth DeAngelis for your contributions and providing

insights both from an architectural and gaming industry perspective. All of your feedback

and guidance have been invaluable towards this research project.

Thank you to the Digital Media faculty who provided me with multiple new creative

outlets by showcasing your passion for teaching within the classroom. Additional thanks

goes to my fellow graduate students: Nate Lapinski, Girish Balakrishnan, Kevin Gross,

and Yujie Zhu for all the constant feedback and support you have given me during our time

at Drexel. The amount I have learned from you all during our various projects cannot be

repaid. I would not have grown creatively without the constant support of the Digital Media

student body and alumni. Thank you for the countless hours you have spent helping me

grow as an artist.

Finally, special thanks go to good friends Matt Haas and Jason Graves for always being

open to conversations about game design and theory. The endless amount of support and

general discussion has made this thesis project possible.

iv

Table of Contents

List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.1 Applying Film Visual Narrative principles to Game Design . . . . . . . . . . . . . . . . . . . 5

2.1.1 Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2.1.2 Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.1.3 Staging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

2.1.4 Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.2 Game Design Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.3 Attention Not Immersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.4 Theme Park Visual Narrative Relationship to Games . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3. Principles of Visual Narrative in Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3.1 What is Visual Narrative in Games? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.2 Five Structural Visual Narrative Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3.2.1 Contrasting Shape Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3.2.2 Framed Structure Principle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.2.3 Directional Line Principle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3.2.4 Shifting Elevation Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.2.5 Structural Exaggeration Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

4. Empirical Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4.1 User Study Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

4.2.1 Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

v

4.2.2 Post Interview Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4.2.3 External to the Game Influence Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4.2.4 Self Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

5. Conclusion & Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

A. Case Study Journey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

vi

List of Tables

4.1 Zone Principle Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.3 Map A Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.4 Map B Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.2 Zone Principle Locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.5 Map A Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.6 Map B Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

vii

List of Figures

1.1 Image from Journey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1 Image from Heavy Rain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.2 Image from Bioshock Fort Frolic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82.3 Journey composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.4 View of Space Mountain at Walt Disney World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.1 Breakdown of Narrative in Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2 Breakdown of Structural Visuals in Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163.3 White outlines the contrasting shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173.4 White outline surrounds the framed structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.5 White directs the line through repetition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203.6 White line directs the path of elevation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213.7 White outlines the exaggerated structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234.1 Map A Overhead Zones Part 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.2 Map B Overhead Zones Part 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.3 Map A Overhead Zones Part 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.4 Map B Overhead Zones Part 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264.5 Map A and Map B - principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284.6 Map A - Principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284.7 Map B - Principle is not present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284.8 Map A - Principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294.9 Map B - Principle is not present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294.10 Map A - Principle is not present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.11 Map B - Principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.12 Map A - Principle is not present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.13 Map B - Principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.14 Map A and Map B - principle is present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.15 Map A Part 1 Heat Map - Red Line is Intended Path. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.16 Map B Part 1 Heat Map - Red Line is Intended Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.17 Map A Part 2 Heat Map - Red Line is Intended Path. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364.18 Map B Part 2 Heat Map - Red Line is Intended Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

viii

AbstractDeveloping Visual Narrative: Designing Structural Composition Principles to Guide

Player’s Attention in Adventure Games

Glenn Joseph WintersAdvisor: Jichen Zhu, Ph.D.

Committee: Frank DeSantis, and Garth DeAngelis,

Computer game design lacks a language for visual narrative principles, in ways simi-

lar to those in architecture, film, and theme park design. In this thesis, we develop visual

narrative methods in which spatial composition principles enhance goal direction attention

within the overarching level structure of computer adventure games. We present our work

on defining goal-directed attention structure composition patterns as a subset of visual nar-

rative principles within 3D adventure games. Based on our observation of goal-directed

attention game design patterns in existing 3D adventure games, we attempt to define a lan-

guage that game designers can utilize to prototype levels more efficiently and apply them

to a testing scenario.

1

1. Introduction

Exploring visual narrative structures as a means to provide desired experiences in games

is not a new concept. Theorists assert that early games such as Tetris have a basic narrative

structure. Interactive designer Janet Murray states that the game is about winning and

losing, casting the player as the opponent-battling or environment-battling hero. Minimally,

games want to tap the emotional residue of previous narrative experiences [24]. Events,

characters, and patterned action will further define the quality of the narrative rather than

the experience [27]. Narrative and gameplay mechanics are two of the most prominent

forms of expression that can be found in a modern video game [29].

Jenkins declares how narrative enters on two levels, in terms of broadly defined goals

or conflicts and on the level of localized incidents [33]. Game designers struggle with a

balancing act by trying to determine how much plot will create a compelling framework and

how much freedom players can enjoy at a local level without derailing the larger narrative

trajectory [33]. As inexperienced storytellers, designers often fall back on mechanical

exposition through cut scenes, much as early filmmakers were sometimes overly reliant

on intertitles rather than learning the skills of visual story telling [33]. Computer games

have evolved beyond just abstract gameplay and into generalized entertainment medium

that offers imagery, audio, and text of any kind [29]. In recent years research has led to an

increased interest in exploring narrative principles through visual means [1]. Therefore, to

understand this medium, it is important to study the effectiveness and define visual narrative

principles in computer games.

This thesis explores the principles of structural composition as a subset of visual nar-

rative in games and how each can influence attention driven goals in 3D adventure games.

We observe game patterns between popular 3D adventure games Uncharted 3 [8], Dear

Esther [32], and Journey[31]. From those, we conclude on five principles and grey box

2

them within two level designs for an empirical study to determine if they produce effective

results. It is our belief that by defining structural construction game principles we will be-

gin the formalization of visual narrative principles and provide an understanding of how to

direct player’s attention in 3D adventure games. By sculpting these concepts the designer

can support design choices in the grey boxing phase that will guide a player through visual

cues but give the sense that they are making the decisions.

Contrary to games, film is a medium that has established visual narrative principles. For

example, Jaws uses visuals to gain the audiences attention when the shark is approaching

based on the visual exposition which is established through music cues and camera angles

at the beginning of the film [5]. Furthermore, the director is able to adjust attention simply

by contrasting the visual exposition with the visual conflict. This is based on the concepts

of contrast or affinity, which states the greater the contrast in a visual component, the

more the visual intensity or dynamic increases. Where as, the greater the affinity in a

visual component, the more the visual intensity or dynamic decreases . Henry Jenkins

states that there is a tremendous amount that game designers and film theorists can learn

through making meaningful comparisons with each others methods [33]. Block agrees that

components of film visual narrative principles can be used in any video game and game

players will react to the visual contrasts and affinities.

Although in games, designers must look at how a player traverses the space to tell

a story and direct the players attention. Game developer Ken Levine asserts the world

is the best narrator and believes that within a games level is where story should actually

be told [26]. Zimmerman insists that the game space is the field of play where players

will explore, compete and cooperate as they travel through the designers game [27]. As

a result, designers should model the possibilities of space as early as possible and invest

time up front to understand how that game would work within their game environment

[34]. In most games, designers are challenged to change a players emotion every 15 - 20

3

Figure 1.1: Image from Journey

minutes and divide gameplay moments while drawing attention to the players next goal.

This method also creates active things for the player to do during gameplay. Therefore, its

important to determine how to stage the visual complexity of these events during the grey

boxing phase or initial level layout. Grey boxing a level in 3D space allows us to determine

shapes, scale and size relationships to basic elements. Game designer Jesse Schell discusses

the value of aesthetics and how good artwork can do wondrous things for a game. It can

draw the player into a game area she may pass over [28]. This can be seen in Figure 1.1

from the popular Playstation 3 game Journey by ThatGameCompany. Different from many

other computer games, Journey provides no textual description of the story or goals. Thus

to convey the story, it relies heavily on visuals and interactions with the environment to

direct the player towards her next goal. In other words, the environment and the resulting

visual narrative are crucial narrative elements in addition to the more common storytelling

components in the game.

Schell further points out how visual design can effect where the player will look. Walt

Disney used a similar method to navigate guests throughout Disneyland Resort by hiring

members of the film production team to design the lands of the park. The stories from the

films were a driving force in order to have guests pulled into a transparent environment

[14]. Game designers dont simply tell stories; they design worlds and sculpt spaces similar

4

to the design of the Disney theme parks [33]. Game designer Jason Roher declares that

games give designers the constant challenge on how to keep players attention throughout

the dozens of hours of gameplay [17]. High-end graphics can only accomplish this for

short periods of time. Designers need to learn how to hold and direct attention to parts of

an environment during their experience. Having a clear goal motivates the player to act and

gives us a clear direction of what to do.

In the past, video games have relied on obvious visual and audio cues to direct players

attention towards certain goals within a game’s level such as a weapons targeting device

in a combat situation to lead a player forward. Milam defines this as a path target pattern

which orients the players movement towards visible targets in the level [23]. When a goal

is attached to a specific task it gives us an energizing push and a sense of purpose [20].

Exploring goals allows the player to discover different areas at her own pace [26]. Design-

ers are constructing experiences utilizing structural compositions to provide opportunities

to direct a players attention for personal exploration rather than a literal guidance towards

her goals. These techniques have not been defined like the standard cues to help formalize

the level design process and allow for efficient level construction. Therefore, this thesis

sought out how structural compositions affect attention driven decisions in 3D adventure

games. Each principle concluded with effective results and provided a foundation for future

discussions about observing and defining visual narrative in games.

This thesis is organized as follows: Chapter 2. We provide analytical framework on

previous work that discusses attention driven narrative patterns, film influences and theme

park influences in games; Chapter 3. We cross analyze game design patterns from current

adventure games; Chapter 4. an empirical study from these patterns and tested in a grey

box environment to determine if patterns can be used to efficiently prototype levels and;

Chapter 5. we conclude on our work and discuss future directions

5

2. Related Work

Game scholar, Jose Zagal discusses that understanding games can be defined as having

the ability to explain games by deconstructing them and understanding their components,

how they interact, and facilitate certain experiences in players [35]. Zagal states Knowl-

edge Building as a concept where ideas that are valuable to a community are continually

produced and improved. The central notions of knowledge building is that knowledge is

not static and given but can be improved over time. Understanding can also be gauged

by exploring the evolution and change of that discourse. Common practices are observing

game design patterns which rely general descriptions of particular areas of game play with-

out using quantitative measures [4]. Research that utilize these methods for visual narrative

concepts has received little attention but growing communities of designers are starting to

observe the space by observing game patterns, adapting theme park and film principles to

the game space.

2.1 Applying Film Visual Narrative principles to Game Design

Applying film principles to games is not a new concept and has been utilized in many

ways in the form of cut scenes or active gameplay. Different from our work, most princi-

ples take control of the players camera or provide literal guidance in order to achieve the

principles effectiveness. Game director David Cage asserts, people demand strong char-

acterization in games as demonstrated by film and literature. He challenges game design

concepts used by game creators for over 30 years and supports adapting certain narrative

structures of film within a games setting. He discusses how in film, the narrative supports

the action rather than separating the two like in a standard classic game design process [7].

Game designers normal narrative game structure differs from film by having to take into

6

account action mixed with standard cut scene [7]. That is something that is slowly changing

to be more active by incorporating game play with cinematics like the Uncharted series.

Jenkins wants people to consider what ways the structuring of game space can facilitate

different kinds of narrative experiences [33]. For the purpose of this project we take into

account principles that support defining structural composition such as composition, space,

staging, and shape.

2.1.1 Composition

David Cage’s game, Heavy Rain [9], tries to challenge the typical cut scene structures

by borrowing practices from the film industry and giving the illusion of freedom to the

player. In this game, the player’s actions are playing the story and not watching a cut

scene. For example, at one point the main character of the game is put through a trial of

cutting off one of his fingers to get information on his son’s kidnapper. Instead of relying

on a cut scene, the player is given the emotional choice of what to do. The player can

ignore the countdown of the clock and risk everything or go forward with cutting off the

finger. Players are given control movements in order to achieve this goal. The player is

more connected by these experiences since these actions are consequential and will affect

the game’s outcome [7]. The designers are sculpting a virtual space that supports the overall

Figure 2.1: Image from Heavy Rain

7

structure of the narrative. In Figure 2.1, the camera placement is adjusted to help guide the

visuals similar to film’s rule’s of thirds. The rules of third is a guideline that proposes an

image should be imagined as divided into nine equal parts by two equally placed horizontal

lines and two equally-spaced vertical lines, and that the important composition elements

should be placed along these lines [25]. At this point the designer locks the camera in the

ideal location but the player still has some camera freedom allowing the viewer to create

their experience. By adding more foreground and background objects we can increase the

tempo of the visual rhythm due to the placement of objects [5].

Hitchcock used techniques of cinematic framing to stage a scene in which game de-

signers can learn from. He has a style in which he considers where characters and objects

are placed within a frame and how other elements attract attention [13]. Designers can use

a similar Hitchcock technique like foreshadowing to excite the player about the activities

and dangers ahead. For example, Bioshock[11] uses this with staging their environments

with various poster designs. The player may not understand the significance of the posters

but they are foreshadowing future adventures [26].

2.1.2 Space

Space incorporates factors such as depth queues, deep space, and perspective all ac-

cumulate to support the characteristics of the screen. Cinematographers use techniques to

direct attention to the object of greatest interest. Techniques such as size and closeness of

an object, the sharpness of focus, spatial relationships between people, framing and how an

objects can be drawn towards objects in motion [6]. This allows the designer to keep the

attention on objects that relate to greatest dramatic significance. Designers use vanishing

points and lines of perspective to guide the player along by drawing the audience’s atten-

tion to certain areas. This can either be done through one, two or three point perspective

viewpoints. By having more vanishing points we can create more illusion of depth. Game

8

designer, Luke McMillain discusses how greater the player’s line of sight, the more they

will feel empowered and the easier a particular scenario may become. The confined spaces

and use of flashlight in Doom 3 are an excellent example of how the mood is tense due to

the limited line of sight offered by tight, constricted corridors. The player’s line of sight

will increase depending on the camera position, frustum and perspective whilst occluding

objects will limit the line of site [21].

2.1.3 Staging

Size variations in level geometry should be used in a way in which the player can

observe contrasts in their environment [22]. By providing a sense of awe designers can

create something bigger than us. Psychologists say it’s the single most overwhelming and

gratifying positive emotion and inspires us to do well [20]. Stage management was very

important to Bioshock. Lead level designer Bill Garnder stated that when you’re building

a space like Fort Frolic you have a lot of different ideas that come into a designer’s head.

You have to have the introduction into the space and establish what Fort Frolic is about.

Following that, the designer needs to draw the player into the quest and make sure the

gameplay is interesting every time you turn [15]. Therefore, supporting the shape of the

picture is the tone and color. This can be observed in Figure 2.2.

Figure 2.2: Image from Bioshock Fort Frolic

9

2.1.4 Shape

Shape is a component that is one of the strongest visual techniques in game design.

Silhouette design is a very powerful way of designing characters, props, and environments

[10]. It’s a great way to make designs have strong shape and memorable visual appeal to

stand out from the rest. Having an environment silhouetted allows the designer to focus

on the form and visual appeal of the object. Like in film, with the combination of line and

shape, designers can stage movement carefully or evaluate certain pictures. Block discusses

how certain emotions occur when a line is straight vs. curved. A straight line is direct, and

aggressive where as a curved line is more passive and organic. These shapes help create an

emotional mood [5].

2.2 Game Design Patterns

Essentially, game designers are architects creating meaningful and powerful spaces

similar to film. Although, in games the virtual world is free of constraints; it offers de-

signers a universe where anything is possible [28]. In some instances, designers limit the

freedom of what the player can achieve to add to a more controlled experience. These tech-

niques allow designers to continue to sculpt a space for both the designer and the player.

The spatial features of a game have a strong impact on creating the narrative space of pos-

sibilities. The organization of spatial features in a game is critical to the design of a game’s

narrative space of possibilities[27].

Similar to our work, researchers are starting to observe the game space by analyzing

games rather than adapting principles from other mediums such as theme parks and film.

Milam and El Nasr focus on one of the fundamental building blocks of the narrative in

games: goals [23]. The player is constantly making decisions when engaged in a story-

line which should work in tangent with the games visual narrative. Goals not only help

players judge their progress but also guide the significance of their actions [27]. They

10

continue to introduce concepts regarding design patterns that guide a players movement

in 3D games. This is accomplished by interviewing designers and analyzing four games:

Bioshock, F.E.A.R: Persueus Mandate, LOST: Via Domus, and Medal of Honor 2: Heroes.

More specifically these patterns focus on the game mechanic elements to draw the player’s

attention throughout the space.

Different from our work, their pattern results focus on literal visual guidance. For

example, the path target pattern orients and directs the players movement towards visible

targets in level such as a targeting device. The path movement pattern is one that can be

with or without resistance or explicit environmental barriers. For example, a path resistance

could be a barrier such as a locked door or a forced detour. Our work takes into account

these patterns and takes into consideration how they could be complementary to spatial

composition principles due to the interactions the player must have with her environment.

Bjrk et al. discuss several hundred of game design patterns but lack focus on specific

level design construction terminology. Patterns are generalized by: the form of using the

pattern, consequences, and relations. For example, immersion patterns are briefly discussed

as games require players’ attention and as such can make players focus on gameplay to the

extent that they feel immersed in the games. Furthermore, they discuss one of the uses

of this pattern as spatial immersion. This is a result of extensive maneuvering in the game

world in real-time games. Although he discusses this concept he negates discussing specific

effects of varying compositions patterns to understand what draws players’ attention to

areas of a level’s construction [4]. He continues about how immersion is one of the most

difficult patterns to instantiate in game design. The consequence is that immersion can

easily be lost as soon as its gained. Our work defines principles to understand the specifics

of attention driven patterns vs. immersive qualities.

11

Figure 2.3: Journey composition

2.3 Attention Not Immersion

In his 2012 GDC talk, Richard Lemarchand discusses the power of attention over im-

mersion. Lemarchand discusses that designer’s focus more on the topic of immersion vs.

talking about what the player is choosing to pay attention to on the screen. He discusses

how people focusing on the same image for a long time eventually become distracted or

impatient. Therefore switching to another activity, especially one that is low stress can

restore the player’s interest in her goals. He relates this too attention restoration theory,

which asserts that people can concentrate better after spending time in nature or even look-

ing at scenes of nature. For example, in the Uncharted franchise designers would spaces

out intense periods of combat with less intensive exploration. Players of games have the

the freedom to choose what to pay attention to before taking an action [17].

He discusses the “attention design” concept adopted by ThatGameCompany while de-

veloping Journey. He specifically talks about the compositions developed for when the

player enters the space that can be seen in Figure 2.3. The player is dropped into a game

space and is isolated around her environment. She is presented with a short-term goal with

12

Figure 2.4: View of Space Mountain at Walt Disney World

a single shape at the top of the hill. As she climbs to the top of the short-term goal the

camera composes a long term goal of the mountain that can be observed in the distance.

These type of goals are similar to what theme park designers utilize to draw a guest towards

the location of different areas of the parks design.

2.4 Theme Park Visual Narrative Relationship to Games

Theme park design has a few standard practices that relate to designers can sculpt en-

vironments. Senior Show Designer for Walt Disney Imagineering Don Caron has argued

that game designers can learn a great deal by studying techniques of “environment story-

telling” which Disney employs in designing amusement park attractions [33]. Similar to

games, theme park design is a 3D storytelling art that places guests or the player in the story

of the environment and in some cases become a linear experience. Environmental story-

telling creates the preconditions for an immersive narrative experience in several different

ways. Disney Imagineers, who design these spaces, are concerned about how to move

guests from place to place. As designers, Imagineers created spaces that take place in care-

13

fully structured environments, allowing the guests to see and hear in new ways. Ultimately

Imagineers give the guests a place to play [14].

Designers from game developer Naughty Dog have already adopted some of the Imag-

ineers philosophies by giving the player a weaning point. This is a term that Imagineers

have coined to give guests a way to navigate their way within the theme park by having a

structure sit just above the horizon line. For example, Figure 2.4 shows the top of Space

Mountain sitting just above the horizon and gives the guest a goal to head towards To-

morrowland, where it resides. Paths towards the weaning point are usually curved around

obstructions but we are always able to see the main objective [16].

Techniques like this were used when building the Indiana Jones adventure at the Dis-

neyland Resort. The queue line leading up to the ride is a specifically structured path that

tells the story of a jungle into ruins. Graphics such as maps are posted on the walls to show

off false directions and suspense builds due to the visual contrast of the rest of the park.

This structure confuses the perception of the actual length of the trip into the rides main

attraction due to the way the que line is constructed [18]. The sculptural direction carries a

story that unfolds in space and time as the guest moves on throughout the que line. Similar

to this, game designers can sculpt epic visual moments and allow players to describe their

most memorable, gratifying game experiences [20].

In summary, design principles are being adapted into games from other mediums such

as film and theme park design. Several researchers are starting to observe the game space

and cross-analyzing game design patterns to formulate results on the affects certain in-

stances have on games. Research has been formalized on the certain mechanical design

aspects but lacks focus the specific principles that guide the geometry construction within

the 2D game space. Therefore, it’s important for designers to understand the relationships

of how spatial construction can affect player’s attention within the design process.

14

3. Principles of Visual Narrative in Games

Narrative in games enters in the form of both visuals and content. Figure 3.1 showcases

how these categories can be broken down into to subsections. First being the visuals, which

involve design concepts such as structural, light, camera, color, etc. Separately, the content

can be broken down into standard narrative structures such as the exposition, climax, and

resolution. For the purpose of this thesis we will be mainly focusing on the visual aspects

of narrative in games. Visual narrative in games affects various aspects of game design

but is ultimately constructed within the games environment. The environment constrains

the goal and will guide the players movements through physical properties. It communi-

cates boundaries and reinforces and shapes the players identity while providing a narrative

context [34]. Game companies divide the construction of this environment differently but

every game needs management, designers, programmers and artists. Game designer Bob

Bates breaks this down further into categories such as programmers, VFX artists, environ-

ment lighters, character riggers, character animators, level designers, concept artists and

more [2].

Figure 3.1: Breakdown of Narrative in Games

15

Furthermore, Bates states that artists now affect every aspect of game design, which

can range from the user interface to the representation of the game world on the screen.

Creating art has become increasingly complex through the years for game designers. Along

side the artists, programmers need to work within technical limitations of the game engine

to produce visuals running in real-time. Bates further discusses how artists need to not

only work on creating believable images but also an understanding of the limitations of the

target platform. Ninety percent of the feedback in a game is visual.

3.1 What is Visual Narrative in Games?

Film producer Bruce Block, discusses how every picture is comprised of story, visuals,

and sounds [5]. The three communicate the meaning of the picture to the viewer. If under-

stood, a designer can control their impact and have the viewer become emotionally invested

in what they are viewing. Therefore, visual narrative in games can be broken down into

similar components such as space, line, shape, tone, color, movement, and rhythm which

all accumulate to story and visual structure. Several factors come into account, which in-

fluence a players decision-making process on what to choose to go towards next. Game

designer Chris Solarski states that with the gentle push of a player’s thumb we can now

interact with these visual worlds [30]. Contrary to other mediums, a game’s visual nar-

rative must be constructed in a 3D space while considering player’s agency. Therefore

designers influence the story and the way the player’s navigation methods. A key challenge

in computer-based interactive narrative is the conflict between user agency and authorial

control of the story qualities [36].

Game designer Steve Gaynor states, “In games that don’t utilize a ton of cut scenes,

visual communication is done via the environment. A lot of similar concepts are used

from film such as framing, emphasis, etc. but in games the visual language lives in three-

dimensional space around the player. Things like continuing to ascend or descend means

16

Figure 3.2: Breakdown of Structural Visuals in Games

you’re going the right way, or cresting a hill or turning a corner to reveal a surprising vista

serve similar purposes. The best cases are subtle enough that the player isn’t conscious of

how they’re being directed to understand the space in the intended way [12].” Therefore,

by understanding the concepts the designer can craft certain paths but also give the player

the illusion of full control.

3.2 Five Structural Visual Narrative Principles

For the purpose of this project we will be focusing on structural composition principles.

More specifically how does shape, space, and form relate to one another and create atten-

tion to certain areas of a level. The following are five principles that were derived from

cross analyzing play sessions and observing game design patterns between popular 3D ad-

venture games Uncharted 3, Dear Esther, and Journey. These principles do not necessarily

stand-alone and can work in tangent with each other to create a stronger statement. By

utilizing these principles the designer can shift the players attention by taking into consid-

eration of her surroundings.

17

Figure 3.3: White outlines the contrasting shapes

3.2.1 Contrasting Shape Principle

Definition: A silhouette contrasts with the adjacent environments overall shape struc-

ture often the picturesque versus the rigid line. Through the use of such contrast, designers

can employ silhouette to emphasize/de-emphasize certain processional goals. This Princi-

ple focuses attention by the figural quality of the silhouette shape and also by affecting the

18

environmental boundary around the shape.

Examples: a) Journey - In Figure 3.3.1, the angular structure contrasts its organic sur-

roundings to draw players attention towards certain locations within the Desert sequence;

b) Uncharted 3 - In Figure 3.3.2, the player is directed to the top of a building in the Citadel

sequence due to its contrasting curve from the horizontal plane created on its top level and;

c) Dear Esther - In Figure 3.3.3, the player is directed towards a building that protrudes

from the side of a mountain during the beach side moment of The Beacon chapter and

bringing focus to various pictorial elements of the narrative.

Figure 3.4: White outline surrounds the framed structure

19

3.2.2 Framed Structure Principle

Definition: A frame may occlude foreground information, focusing attention on a

midground or background viewpoint. More simply, a framing device may simply enframe

a view without occlusion suggesting an inside and outside to this figure. Dominant vertical

elements may define a subset of the horizontal plane. Similarly, horizontal elements can

be used to enframe a vertical plane. This principle focuses attention by heightening the

legibility of the subset view.

Examples: a) Journey - In Figure 3.4.1, the player approached strategically placed struc-

tures during the Water Caves sequence that bring focus to collectibles and also a path to

the player next goal; b) Uncharted 3 - In Figure 3.4.2, designers utilized windows and arch

ways in Greatness from Small Beginnings level to bring focus to specific path goals such

as stairs and important narrative elements and; c) Dear Esther - In Figure 3.4.3, the player

approached an abandoned shipyard filled with open containers during The Lighthouse se-

quence. Upon exiting these containers, there is a specific framed compositions that shows

the player next goal.

20

Figure 3.5: White directs the line through repetition

3.2.3 Directional Line Principle

Definition: A patterning of repetitive lines and/or edges defines a visual, actual and/or

metaphoric pathway(s) linking foreground, midground, and background through foreshort-

ening. The principle focuses attention by defining an implied perspective view thought the

diminishing size of repetitive patterning.

Examples: a) Journey - In Figure 3.5.1, the player is directed along a path from the repeti-

tious placement of bridge like structures to showcase the end goal of The Bridge sequence;

b) Uncharted 3 - In Figure 3.5.2, the player traversed through the Historical Research chap-

21

ter on strategically placed awning that blend in with the side of the buildings to guide the

player in a chase sequence and; c) Dear Esther - In Figure 3.5.3, the player has subtle

carved out paths surrounded by organic rock shapes as she traversed through the mountains

landscapes during The Lighthouse sequence.

Figure 3.6: White line directs the path of elevation

22

3.2.4 Shifting Elevation Principle

Definition: A spatial relationship between ground plane and line of sight. The pro-

cessional sequence is characterized as below, in line, or above an implied horizontal often

with diagonal or parallel relationships of succeeding ground planes. The principle focuses

attention by manipulating the relationship of the ground plane in relation to foreground,

mid-ground and background elements.

Examples: ) Journey - In Figure 3.6.1, designer’s directed players towards the various

goal locations by the shifting the elevation between the fore, mid, and background during

the Descent sequence; ) Uncharted 3 - In Figure 3.6.2, designer’s shifted the elevation of

the level during the Mirage of the Desert sequence which focuses on no action elements

but requires player navigating form one mirage to the next by navigating through elevated

desert planes and; ) Dear Esther - In Figure 3.6.3, designer’s structured path ways within

the Cave sequence and direct players towards the next location based on the contrast in

elevation from her previous path.

23

Figure 3.7: White outlines the exaggerated structure

3.2.5 Structural Exaggeration Principle

Definition: Exaggerated structures contrast their surrounding to showcase emphasis

and direct the players attention towards her next goal. A spatial relationship between

boundaries, edges and centers understood primarily as an interruption of the legibility/continuity

of either a horizontal or vertical plane. This principle results in implied axes of symmetry

and/or asymmetry about dominant horizontals or verticals and uses visual weight, position,

balance and depth to determine focus. The pattern creates attention on the area by breaking

the horizontal plane created when structures are similar in scale in Y coordinates. This can

24

be either used to create the end goal or a subset of goals within the designer’s path.

Examples: a) Journey - In Figure 3.7.1, the player is directed towards her goal from the

exaggerated size of the entrance of the cave that contrast the surrounding environments as

she travels up during The Mountain sequence; b) Uncharted 3 - In Figure 3.7.2, players

are presented with the Chateau building several times early on in the Chateau sequence

to bring focus to it’s importance by constructing it above the horizon line and contrasting

it’s height from the surrounding forest and; c) Dear Esther - In Figure 3.7.3, the player is

directed towards the phone tower during certain points of her play through. This element

is given prominence due to being structurally taller than its surrounding and its placement

as one of the furthest locations on the level.

25

4. Empirical Study

To this point we have discussed our theoretical backing and related work that lead

us to define structural composition principles in 3D adventure games. Similar to previous

approaches we cross-analyzed popular 3d adventure games to determine common structural

designs that support attention driven goals. Based on our observations we developed a user

study to analyze the defined structural composition principles, the effectiveness of each and

answer the following research questions:

1. Do our choices of structural composition principles influence players attention and

guide them to certain goals within a level?

2. Can principles stand on their own or are they in need of assistance by other literal

guidance patterns such as those from Milam and El Nasr to influence players naviga-

tion?

4.1 User Study Design

Our user study was targeted toward individuals between the ages of 18-35 with a min-

imum of 40 participants. Limited information was provided to the player to avoid influ-

encing her decision making process. Player’s were prefaced with the following: “This is a

navigation experiment. You will be presented with path options throughout the level. There

is no right or wrong way to go. Please choose what feels most natural to you and once you

choose that direction please continue to move forward. There will be a post interview

process following your play through.”

Participants were randomly selected to test one of two variations of the constructed

level. A standard first person perspective navigation controller similar to the one in Dear

Esther was used as a control scheme. Each map constructed resembled an adventure game

26

style layout to evoke the sense of exploration. Minimal visual enhancements were con-

structed to increase the players interest levels between pattern locations such as ambient

occlusion light maps, colored shading, non-attention driven lighting and set dressing ob-

jects. Initial levels were designed with 2d sketches and later moved into a 3D grey boxing

environment within Autodesk Maya. Grey boxing involves creating basic geometry shape,

which lacks detail such as textures and shading. The geometry is created in a way to allow

for easy iteration and alteration of the levels layout. Artists are able to use the grey box

environment as a basis for placement of final 3D models. This provided a platform for the

investigator to iterate informal user testing to determine the effectiveness and overall length

of the study.

Figure 4.1: Map A Overhead Zones Part 1 Figure 4.2: Map B Overhead Zones Part 1

Figure 4.3: Map A Overhead Zones Part 2 Figure 4.4: Map B Overhead Zones Part 2

27

Table 4.1: Zone Principle Locations

Zone Principle Map A Map B

1 Contrasting Shape Present Present

2 Framed Structure Not Present Present

3 Elevation Present Not Present

4 Exaggeration Not Present Present

5 Directional Line Not Present Present

6 Goal Present Present

The study was organized further within two separate levels. Each contained six zones,

which evaluated if the principles hypotheses had any influence on the user’s decision mak-

ing process. Figure 4.1 and 4.3 showcases the six specific zones and how it contrasts Figure

4.2 and 4.3. Two levels allowed for variation and observation on how users interacted dif-

ferently in each zone within the environment. It was important to constrict the players view

when approaching each principle in order to have a non-biased decision making process.

Therefore, the viewing frustum was limited as the player approached the principles testing

zone. In between zones we created small areas where player’s can minimally explore to

avoid players feeling over constricted which we define as visual breathing points.

In some instances, the principles influence may not be present in the zone to test the

effectiveness of their hypotheses. For example, a framed structure is presented in Figure

4.2 and not in Figure 4.1 because it is our belief that the structure will draw attention to

certain path goals within the map. Table 4.1 details which zones had patterns present or not

present within the levels construction. The following figures detail out each zone in which

the green arrow is the designer’s intended path vs. the red arrow which is the designers

28

non-intended path. These zones are detailed as follows:

Figure 4.5: Map A and Map B - principle is present

1)Zone 1 - Contrasting Shape - This zone allowed us to compare organic shaped vs rigid

shaped structures and observe what player’s gravitate towards. Player’s are presented with

the same option within both Map A and Map B. In Figure 4.5 one door frame is curvature

shaped similar to the entrance to a cave where as the other is more linear like a standard

door way. It was the goal of this zone to observe if the player would recognize and gravitate

towards organic shaped vs rigid shaped entrance.

Figure 4.6: Map A - Principle is present Figure 4.7: Map B - Principle is not present

29

2)Zone 2 - Framed Structure - This zone allowed us to determine if a framed structure

can guide the player towards a specific goal upon exit. In Map A player’s are presented

with no archway which can be seen in Figure 4.6. Where as Map B contained an archway

that initially points the player towards the right while occluding the left which can be seen

in Figure 4.7. Upon exiting the archway the player is able to see both options and is able to

choose her direction. It was the goal for the player to gravitate towards the right hand side

of the map upon exiting.

Figure 4.8: Map A - Principle is present Figure 4.9: Map B - Principle is not present

3)Zone 3 - Shifting Elevation - This zone determined if a shift in elevation will direct

the player interest level towards an intended path. In Map A player’s are presented with an

elevated set of stairs to the left which can be observed on the left side of Figure 4.8. Where

as in Map B both directions display similar elevation to the current path the player was

previously traversing. This observation can be seen on the right hand side of Figure 4.9. It

was the goal of this zone to see if the elevated stairs presented on the left would navigate

the players direction.

30

Figure 4.10: Map A - Principle is not present Figure 4.11: Map B - Principle is present

4)Zone 4-Structural Exaggeration - The following zone allowed us to determine if an

exaggerated structure provided more interest to a player based on scale. In Map A a flat

horizontal plane is created when both buildings are similar in scale. This can be seen on

the left hand side of Figure 4.10. Where as in Map B the horizontal plane is broken by the

right building being taller than its counterpart which can be observed on the right hand side

of Figure 4.11. It was the goal of this zone to observe if the player chose an exaggerated

structure over one similar in size when placed side by side.

Figure 4.12: Map A - Principle is not present Figure 4.13: Map B - Principle is present

31

5)Zone 5-Directional Line - The following zone determined if a series of similar struc-

tures in a repeatable pattern would guide a player towards her next goal. Player’s were

presented with the option of going left or right of the elevated shape in the open ”sand

dune” area of the map. In Map A structures are placed in a repeatable curvature line to

guide a player down a specific path which can be seen on the left hand side of Figure 4.12.

Where as Map B contains no structures which can be seen on the right hand side of Figure

4.13. It was the goal of this zone to have the players be drawn towards the right hand side

of the map based on the placement of the structures and follow a specific path.

Figure 4.14: Map A and Map B - principle is present

6)Zone 6-End Goal - The following zone determined if the user becomes familiar with

the end goal at the beginning of the level. One structure is immediately visible at the

beginning of the map vs. one that is side by side at the end of the map. Both structures

are the same in Map A and Map B. At the beginning the player is able to see a pyramid

structure but as she enters the open ”sand dune” area she is presented with the tower object

which can be seen in Figure 4.14. These two objects were shaped differently in order for

the viewer to have a distinct choice at the end and not think of them as similar goals. It was

the goal of this zone to see if players would remember the pyramid shape from her entrance

32

into the space and choose it at completion.

4.2 Results

As mentioned earlier players would approach each zone and make her decision on what

path she would like to take. The investigator notated each decision during the play through

of the grey box environment. From these decisions, the investigator was able to interview

each player and discuss their decision making process. Player’s sessions were tracked uti-

lizing heat mapping technology and a screen capture software in order to visually observe

her experience during the post interview process. 3D heat maps allow for a more robust

interpretation of data and avoid any errors when testing areas of the map that may be higher

or lower than a steady Y value. This also gave us a visual representation of what areas of

the map may be underutilized for future iterations of the testing procedures. From these

results, we categorized each map into a table where 1 represented that the player did nav-

igate towards the designers intended path where as 0 was not the designers intended path.

We were then able to utilize both the statistical data and post-interview questionnaire to

support the results of our empirical study.

4.2.1 Statistical Analysis

We were able to formulate some basic demographic information but unfortunately we

were not able to see any common trends between demographics and our results. Between

both maps the players age ranged from 18 - 48 with the average age being 21. Players were

asked how many hours each played video games weekly ranging from under 1 hour to 20

+ hours per week. The average between both maps played 2-5 hours per week.

Table 4.3 details that Zone 1 had 55% of player’s taking the intended path vs Table 4.4

where 65% chose to go towards the non intended path. The variety in these results may be

due to what we discuss in Section 4.2.3. Furthermore in Table 4.4 displayed that Zone was

33

5% stronger than Table 4.4 when a framed structure is present in the level. Zone 3 had the

strongest statistical response with 90% of users in Table 4.3 going in the intended direction.

Where as in Table 4.4 had 75% of users going towards the right hand side when a view of

elevation was not present, which was expected.

Zone 4 showed similar results in both Table 4.3 and Table 4.4 which we discuss further

in section 4.2.2. Statistically Zone 5 proved influential as it can be seen in Table 4.3 on page

34 with a 55% response going towards the right hand side when entering the zone. Where

as in Table 4.4 80% of users were not influenced to go towards the right at all. Finally in

Zone 6, both Table 34 and Table 34 statistics showed that most player’s chose to go towards

the tower vs the pyramid shape.

Unfortunately, Figure 4.15 and Figure 4.16 illustrate that users explored every corner of

the map. Data becomes clustered when users are not prevented from back tracking which

is not necessarily a bad action. The investigator utilized the post interview responses to

resolve any questions about the zones with overlapping data. We were able to observe

large amounts of dense data for the elevation principles influence in Zone 3. Data are

much stronger in Figure 4.18 within this Zone than 4.16. Figure 4.17 does however help

investigators visualize the influence of structures as driving attention in Zone 5. Where as

in Figure 4.18 showed minimal data going towards the right hand side of the map.

34

Table 4.3: Map A Statistics

Zone 1 2 3 4 5 6

Intended 55% 55% 90% 70% 55% 55%Not Intended 45% 45% 10% 30% 45% 45%

Table 4.4: Map B Statistics

Zone 1 2 3 4 5 6

Intended 35% 60% 25% 70% 15% 35%Not Intended 65% 40% 75% 30% 80% 60%

Table 4.2: Zone Principle Locations

Zone Principle Map A Map B

1 Contrasting Shape Present Present

2 Framed Structure Not Present Present

3 Elevation Present Not Present

4 Exaggeration Not Present Present

5 Directional Line Not Present Present

6 Goal Present Present

35

Figure 4.15: Map A Part 1 Heat Map - Red Line is Intended Path

Figure 4.16: Map B Part 1 Heat Map - Red Line is Intended Path

36

Figure 4.17: Map A Part 2 Heat Map - Red Line is Intended Path

Figure 4.18: Map B Part 2 Heat Map - Red Line is Intended Path

37

Table 4.5: Map A Responses

User Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 601 0 0 1 1 0 002 1 1 1 1 0 103 0 0 1 1 0 004 0 1 1 1 1 005 1 1 1 0 1 106 0 1 1 1 1 107 1 0 0 1 0 108 0 0 0 1 1 109 0 0 1 0 0 110 1 1 1 1 1 011 1 0 1 0 0 112 1 0 1 1 1 113 1 1 1 0 1 014 0 1 1 1 0 115 1 0 1 1 1 116 0 1 1 1 0 017 0 0 1 1 1 018 1 0 1 0 1 019 1 1 1 1 1 020 1 0 1 0 0 1

Male 16 Female 41 = Yes / 0 = No

38

Table 4.6: Map B Responses

User Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 602 0 0 0 0 0 003 0 1 1 0 N/A N/A04 0 1 0 0 0 005 0 0 1 1 0 006 0 1 0 1 1 007 0 1 0 1 0 008 1 0 0 1 0 009 1 1 1 0 1 110 0 0 0 1 0 011 1 1 0 0 0 012 0 1 1 0 0 113 0 0 0 1 0 014 0 1 0 1 0 115 1 1 0 1 0 016 1 0 0 1 0 017 1 1 1 1 0 118 0 1 0 1 0 019 0 1 0 1 0 120 0 0 0 1 1 121 1 0 0 1 0 1

Male 18 Female 21 = Yes / 0 = No

39

4.2.2 Post Interview Analysis

To this point we were able to observe a statistical analysis that formalized the hard

intended vs. intended decision-making process. Moving forward we were able to support

or clarify those results based on the post-interview held immediately following the play

through. Zone 1 provided interesting results revolving around not only the shape of the

entrance but also participants habits. One of the biggest results was that users who chose

the organic shape door specifically mentioned that the shape was more interesting than the

standard entrance doorway. User 007 and 010 of map A related the organic shaped entrance

to the terms ”exploration and adventurous.” Several users mentioned that they chose the

rigid door over the organic shape entrance due to clarity of depth in one hallway vs. the

other. In map B, user 004 specifically mentioned that they observed more ”viewable depth”

in the right hand side vs. the left. In some instances it was considered an instinct decision

to choose the right over the left due to it being the first path the user saw as they navigated

down the hallway. Interestingly, early on user 004 in Map B discussed that he is right

predominant and felt that he mainly gravitates towards the right when making his choices.

We discuss this further in section 4.2.3. In map B, user 019 mentioned that the straight

edges caught his attention more than the curvature shaped presented to the left. Although,

one is presently not stronger than the other, we are able to recognize a pattern that organic

shapes are specifically being described more interesting where as the rigid doorway being

chosen due to it being on the right side.

No users specifically stated that the framed structure in Zone 2 guided them towards the

right hand goal. One of the biggest influences was the pillar and fire escapes that players

were presented with on the right hand side as they panned upward when making their

decisions. User 008 of map B mentioned that upon exiting the frame structure he initially

gravitated towards the left but changed his decision as soon as he saw the pole structure

to his right. User 006 in map A mentioned that the fire escapes at the beginning of the

40

map caught his attention and that being the significant reason for choosing the right hand

side vs. the left. Once again, several users mentioned that they generally gravitate towards

the right hand side when making navigation decisions. In map A, user 008 explored the

left side due to it being “smaller in size” and felt that the ladders made the right feel more

expansive. We still were able to see stronger results of players gravitating towards the right

hand side of the map when a frame structure was present. Alternatively, in further testing

scenarios it would be interesting to restructure the framed element with the desire point

within the frame and as the user exits the frame structure she is presented with a similar

desire point on the left hand side.

An intriguing observation from Zone 3 arose during the testing phase of the elevation

principle. As the player approached the L shaped hallway she would try to immediately

go right despite it being a small alcove. This is once again a reoccurring pattern that we

discuss in Section 4.2.3. Furthermore, Zone 3 proved to be the most successful scenario.

The elevation was not present for users in map B and in 75% of cases users continued down

their normal right hand decision-making process. In map B, user 011 talked about how he

didn’t go left because nothing seemed to influence his decision to go in that direction.

In map A, user 009 specifically mentioned that he saw the stairs and usually wants to go

higher when playing. Similarly, user 002 felt that the stairs would lead him to a place where

he could see everything. Furthermore, user 012 specifically mentions that the stairs drew

interest vs. his other option, which had a similar floor plane he was presently on. User

013 mentions that the right hand side would most likely be more flat and the left would

allow him to go upwards. Finally, user 017 states that the right hand side felt like he was

on the same level where as going up was someplace different. From these results we are

able to determine that elevation was an influence to motivate and draw players attention in

a certain direction.

Surprisingly, both cases for Zone 4 provided the same results regarding the exaggeration

41

principle. Although several observations were made from the results of the participants in

map B. Out of all 40 testing members, 45% of the users had no specific reason why the

chose the right over the left but 20% specifically mention that they chose the right direction

vs. the left due to it being larger than the other building. In map A, use 007 mentioned

that due to it being the same size he went with the closer option. User 016 in map A

specifically mentioned that he thought they were both the same size and it was a coin flip.

We decided to take this opportunity to ask the user if anything would have influenced his

decision if he were presented with another options. The user responded that he would have

gone to a much taller building over a smaller one. Previous to this pattern in map B, the

intended path was towards the left. In a few cases, players mentioned that there choice was

influenced based on their what direction they chose previously. For example, one being

that they decided to go left because they had been constantly going right up until this point

and vice versa. Therefore, we can conclude that players are noticing exaggerated structures

as a point of focus. In the future it would be beneficial to swap the exaggerated structure to

the left hand said which is the contrary to participants standard directional choice.

It was clear that the repeated lined structures added to the right hand side of the map

influenced 55% of the users in that direction for Zone 5. Users in map A, specifically

mentioned that there intentions of going right was decided by the curiosity of what may

be placed in structures. One user explained that he likes to follow things in the distance

and found the tunnels interesting. User 015 brought up that normally the user is rewarded

if they spend the time to explore. Continuing on, User 006 discussed that the ring created

from the placement of the boxes was very intriguing and felt it would eventually lead him

back to the tower. Two users in map B felt that by turning right it would loop back and

gravitated in that direction because they were concerned they would miss something. In

map B, most of the 80% statistical results discussed that the main reason they went left is

that nothing drew their attention to the right of the map. As soon as they saw the tower

42

structures in the distance they ended up going in that direction.

Finally, players in most cases gravitated towards the tower structure vs. the pyramid

structure in Zone 6 when testing if the player remembered the goal from the beginning of

the game. Player’s responded that the reason they chose the pyramid over the tower was

due to the fact that it was a more interesting shape. Where as, the tower was mainly chosen

due to it being the furthest location in the players path. No player’s mentioned that they

remembered seeing the pyramid at the beginning of their play through. A stronger testing

scenario for future work would cause the player to see the pyramid structure several times

early on in the map before being presented with the option at the end. It will be important

to the testing scenario that other locations are not hindered by this set up.

4.2.3 External to the Game Influence Analysis

During the user testing several factors arose that would influence further iterations of

the testing scenario. The most interesting was the how certain players brought up the fact

they normally choose to go towards the right when making there directional decisions.

The investigator informally asked the participants if they were predominately right or left

handed. Unfortunately, the right-handed testers did outweigh the left-handed player’s to

form conclusive data. Two testers specifically notated that they normally follow the hand

on the wall rule, which is commonly used for maze algorithms. Interestingly left hand

predominate user mentioned it as the left hand on the wall rule vs. the right hand users

referring to it as the right. Furthermore, one user mentioned he enters a room in games

similar to his military training, which is normally done by scanning the room from left to

right. It will be important to consider these factors in future iterations even though our

study can not provide data to support these concerns.

43

4.2.4 Self Analysis

It is important that designers recognize that most players inadvertently choose to go

right when organizing zones in the future. Designer’s can take this into consideration early

on in the development process and make sure that the guided path is varied throughout the

design of the level. Consideration also should be taken as to what the designers intended

previous path was and possibly alternate it with her next intention. When constructing

future levels one must consider the steady flow of constantly making one direction is equal

with the other. It proved valuable to present both options equally to the player. Minor

adjustments should be made so that the depths of entrances are presented equally.

44

5. Conclusion & Future Work

In this thesis we have presented and evaluated an approach to define structural composi-

tion principles as a subset of visual narrative to guide goal driven attention in 3D computer

games. The breakdown of our approach is: define structural composition principles by

cross analyzing patterns in popular 3d adventure games, design a level to test the effec-

tiveness of our principles, and then utilize the results to test the effectiveness of structural

principles for visual narrative in games.

The study does confirm our hypotheses: structural composition patterns can inuence

the players attention to certain areas of a level. Furthermore, it pushes for the discussion

on how certain principles can stand on their own or if their need support of others to make

a stronger attention driven statement.

The empirical study provided us with data about future work. On a low level it would

be beneficial to run this testing scenario through other iterations by swapping the desired

paths in each zone and determine if the structures are still effective. Small mistakes such

as the depth of the hallway made a larger impact than expected on the player’s decision-

making process. Its important to take into consideration how the player views the end of a

hallway just as much as the entrance to that hallway. Furthermore, it would be beneficial to

take into consideration more external decision making factors such as cultural influences.

As already mention, Milam and El Nasr work proved to be beneficial and we feel that

combining their patterns with our results may provide stronger results on defining what

visual narrative means to 3d computer adventure games. For example, by combining the

directional line principle with collection pattern may result in an opaque decision in the

players direction towards her next goal.

On a higher level, a more procedural approach would provide a more detailed analysis.

These principles essentially are variables for a level that is procedurally generated and

45

provide more variation for user testing scenarios. By doing this, the investigator can have

quick iteration times and test a larger amount of subjects and level variations to further

define the effectiveness of the patterns. As a whole, visual narrative in games is as extensive

as other mediums such as film and theme park design. Areas of specific interest are how

lighting and camera design influence players decisions. By defining these other variables

designers can understand the effectiveness of how to draw the player towards her goal

without taking complete control of her experience.

On broader scale, game design can become more of a procedural approach, which

would allow for more variation in game play styles. Game designers, artists and program-

mers can work together to create systems that will dynamically change the environment

and possibly adapt to the habits of the players game play style. This would allow for a

more robust system that would drive storytelling within the 3D space. Though, in order

to accomplish this, games must be studied as its own medium for creators to understand

the effectiveness of design principles such as structural composition, have on the player’s

attention towards her goals. Therefore, it may not be effective to try to adapt film principles

into games but try to understand how those style of principles effect the player during game

play.

Games as a medium have matured into a form of expression that provides opportu-

nities for players to encounter new experiences. We discuss this further in Appendix A.

where we break down Journey and discuss how the visuals support the narrative context

but also provided an opportunity for the player to create her own story. ThatGameCom-

pany created a structure that would allow players to enter open minded and build her own

narrative alongside the underlying story. They challenged themselves to do this visually

and it was accomplished by taking into account how the visual narrative is constructed and

supports the overall classic Monomyth narrative pattern. This thesis presents the first step

in understanding these variables and treat games as their own medium

46

Appendix A. Case Study Journey

A close study of ThatGameCompany’s Journey analyzes elements that showcase how

structural composition methods work to support the games narrative structure and goals.

The overarching narrative structure of Journey follows classic Monomyth Pattern of a heros

journey. The hero begins with a call for an adventure that will include challenges, rebirth,

transformation, and atonement. Helpers and mentors assist him along his quest and ulti-

mately lead him back to the start of the adventure. In this full cycle of the heros journey,

the Journey player learns about the story of the rise and fall of a civilization. As mentioned

above, thanks to the games open-ended structure and characterization, the player is able to

interpret the finer details of the heros tale on her own.

In Understanding Comics [19], Scott McCloud argued that when images of cartoon

characters become more abstract, it is easier for the readers to identify themselves with

the characters. In other words, the language of abstraction leads to greater universality.

The characters in Journey, like its overarching narrative, are purposefully designed to be

abstract for similar effect. As the player progresses through each stage of the heros journey,

she can draw on her own background and complete her own version of the story through

the faceless characters and the open-ended pictograms.

Unlike many other instances of a heros journey story where text or spoken words are

used to communicate the end goal, Journey communicates its story and the emotional tale

by intertwining the abstract visual narrative of the players experience with the implied story

arcs. By visual narrative, I mean how visual composition within level design communicates

the goals that are intertwined with the games narrative arc structure.

For example, Journeys ancestral narrative is implied through pictograms, a form of writ-

ing that uses representational, pictorial drawings and is the basis for hieroglyphic writings.

These pictograms depict the beginning of time and the birth of a civilization. A natural

47

energy source comes from the sky in the form of a shooting star and creates life, also refer-

encing the beginning of the player experience. This is the birth of animals and vegetation

and ultimately leads to the birth of the first red robe travelers. They are given the gift of

another energy source in the form of cloth that is used to start building their architectural

structures. While this occurs the travelers ancestors focus less on natural energy and focus

more on urbanization. Natural energy sources such as plants and wildlife slowly disappear

as buildings overshadow nature.

Because of this, energy becomes sparse and panic ensues until no other energy sources

remain. Rock like creatures are used to gather the remaining energy sources in a war

between the travelers. Ultimately, this leads to the demise of the traveler species and the

civilization begins to fall. The important architecture structures begin to crumble and sand

covers this once thriving civilization. The natural energy source once again starts to emit

from the earth and drifts over the tombs of the travelers. Ultimately, it gives birth in the

form of a shooting start, the player.

This story is fully unveiled before the players final climb towards the top of the moun-

tain. This is the moment when the player also comes to the realization that she has traveled

through all these different scenarios displayed in the pictograms but in their fallen environ-

mental state. These visuals serve as the backbone of the players story and describe what

she must learn before she reaches the atonement arc of the Monomyth pattern. Therefore,

its important to discuss how the player views her story through each sequence.

The player is introduced to the tombs of the past in an open desert and begins her call

to adventure. The sequence is followed by architectural structures that contain remaining

scarf like energy creatures trapped within the fallen structures that once were the main focus

of her ancestors. These creatures provide support to the player and give a sense of comfort

throughout the rest of her journey. She continues her quest and discovers the beauty of

the once thriving civilization by observing and interacting with protruding architectural

48

structures. The player begins the descent into the tunnels during the threshold arc of the

Monomyth pattern and meets the rock like creatures. These were once considered war

machines but now act as guardians of the past and are the only element throughout the

game that can hurt the player. Eventually the player slowly rises back to the surface and

encounters her ancestors one last time.

As mentioned earlier, ancestors display the heros history at the end of the sequences

and act as a mentor for the player. However the ancestor reveals to the player that her story

is a continuation of the ancestors history and foreshadows what is to come. This is the

moment of revelation and notifies the player of her death in order to achieve atonement.

The player makes one last struggle towards the top of mountain as she witnesses the death

of her red robed traveler. The ancestors meet the player one last time and rocket her towards

atonement by giving her their ultimate energy source. It also serves as a moment of peace

and rest before her ultimate return to where she began. This results in the player returning

to the beginning of her journey in the spirit of a shooting star.

Journeys visual narrative structure of story, visuals, and soundtrack work in conjunction

with the travelers history and speak to the player. Key elements are important in working

in conjunction with players story in order to provide a sense of emotion and exploration.

They are color, light, structural composition, and visual pacing.

The use of color serves as a huge factor in the way images relate to the narrative and

immediately conveys an emotional tone. The narrative quality of color sequences and the

emotions they evoke work in parallel with the storys key dramatic moments. Lighting

methods enhancing the color value, which add to the players sense of awe and astonish-

ment. For example, during the narratives threshold crossing the player begins her descent

into the tunnels, which results in one of the most beautiful parts of the game due to the way

light interacts with the structural composition and creates a warm color palette. This palate

slowly declines into a shade of dark blue as the player gets closer towards the end of the

49

sequence. This conveys the time shift from daytime to the night, but also causes the players

emotions to change from a sense of comfort to a sense of cautiousness. This correlation of

story and emotion corresponds with the challenge arc of the Monomyth pattern in order to

reach revelation.

As the player learns more about the red robed travelers heritage she becomes more chal-

lenged by the environment and its structural composition. Structural composition relates

to how buildings are constructed, laid out and serve as the incentive for the players goals.

For instance, throughout the desert, buildings are sparsely laid out in order to show the

desertedness of the fallen city. Hidden in the corners are structures that showcase bonus

pictograms that deepen Journeys narrative. The environment becomes dense as the player

moves through each sequence. This creates several different emotions due to its interaction

with color values. Its important to note how structural composition (i.e., level design) is

used as a goal driven device due to the lack of spoken word. For example, the player starts

in an open desert and makes an initial climb up a sand dune. As the player reaches the crest

of the dune she is able to see the summit in the distance. Due to its prominent placement

just above the horizon line, the player can sense its narrative importance, even though she

may not realize immediately it is the final destination of the story.

In Journey visual pacing is created by combining several visual narrative elements. It

functions by affecting not only the players attention within a level, but also how it causes

a player to slow down or speed up. A perfect example comes after the players revelation

when she approaches the mountain during her final transformation. Wind pushes the player

down the mountain if she does not stand behind strategically placed goals shaped similar

to the tombs at the beginning of the game. This enforces the storys final struggle and the

challenges that the player needs to take before the revelation arc of her story. The player is

able to feel this due to the way she interacts with the environment. Contrasting this, is the

visual pacing as the player begins her return to where the journey started after the state of

50

atonement. At this point the player returns to a comfort level familiar to that at the begin-

ning of the game. Players are free roaming and engulfed in a bright atmosphere that cause

no visual restrictions on how fast or slow they can proceed throughout the environment.

Ultimately, the player is able to take one last walk into the summit and once again become

a shooting star to restart the storys cycle.

Finally, along with these two styles of storytelling, Journey allows the player to create

stories written by her game play experience. This is the result of what Chris Bell, designer

of Journey discusses in his 2012 GDC talk Designing for Friendship. By engaging in the

experience with other players and sharing a wide range of emotions, a narrative is created

[3]. This can be attributed to the combination of story structure, visual narrative elements,

and how the players communicate.

Players are able to develop these bonds by having a simple language mentioned above.

During a recent play through, I was exploring the environment to locate certain collectibles

in order to achieve white robe status achievement. My fellow traveler had already achieved

this status and realized my intentions based on the actions I was making within the envi-

ronment, drawing attention to certain areas through the use of singing. From that point my

companion was able to lead me to specific areas of the map and I was able to collect the

final pieces. If I fell behind, the player would wait for me and vice versa. Its a unique way

for the player to create a friendship story alongside your heros quest and cross the finish

line together.

52

Bibliography

[1] Ernest Adams. Fundamentals of Game Design. New Riders Publishing, ThousandOaks, CA, USA, 2nd edition, 2009.

[2] B. Bates. Game Design. Course/Premier Press, 2004.

[3] Chris Bell. Designing for friendship. Presented at the 2012 Game Developers Con-ference.

[4] S. Bjork and J.A. HOLOPAINEN. Patterns In Game Design. Charles River MediaGame Development Series. Charles River Media, 2005.

[5] Bruce A. Block. The visual story : seeing the structure of film, TV, and new media.Focal Press, Boston, 2001. 00068126 Bruce Block. ill. (some col.) ; 26 cm. Includesbibliographical references (p. [255]-257) and index.

[6] J.M. Boggs. The Art of Watching Films: A Guide to Film Analysis. Benjamin, 1978.

[7] David Cage. Experiencing stories with/in digital games. presented at Technoculture,Art and Games (TAG) research centre.

[8] Naughty Dog. Uncharted 3. [Blu-ray Disc], 2011.

[9] Quantic Dream. Heavy rain. [Blu-Ray Disc], 2010.

[10] Alex Galuzin. Silhouette design game environments, 2012.

[11] Irrational Games. Bioshock. [Blu-Ray Disc], 2007.

[12] Steve Gaynor. private interview, Jan. 24 2013.

[13] Tom Gunning. Hitchcock and the picture in the frame. New England Review (1990-),28(3):pp. 14–31, 2007.

[14] J. Hench, P. Van Pelt, and M.A. Sklar. Designing Disney. Disney Editions, 2009.

[15] Ratana Huot. From the vault: Evolution, July 2012.

[16] R. Lemarchand, J. MinKoff, and J. Richmond. Naughty dog live and unleashed gamedesign with the dogs. Presented at PAX East 2010.

[17] Richard Lemarchand. Attention, not immersion:making your games better with psy-chology and playstesting, the uncharted way. presented at the Game Developers Con-ference 2012.

[18] K.A. Marling and Centre canadien d’architecture. Designing Disney’s theme parks:the architecture of reassurance. Canadian Centre for Architecture, 1997.

53

[19] S. McCloud. Understanding Comics. A Kitchen Sink book. HarperCollins, 1994.

[20] Jane McGonigal. Reality Is Broken: Why Games Make Us Better and How They CanChange the World. Penguin Group , The, 2011.

[21] Luke McMillan. A theory of compression and funneling, May 2011.

[22] Luke McMillan. The metrics of space: Tactical level design, September 2012.

[23] David Milam and Magy Seif El Nasr. Design patterns to guide player movement in3d games. In Proceedings of the 5th ACM SIGGRAPH Symposium on Video Games,Sandbox ’10, pages 37–42, New York, NY, USA, 2010. ACM.

[24] Janet Horowitz Murray. Hamlet on the Holodeck: The Future of Narrative in Cy-berspace. The Free Press, New York, NY, USA, 1997.

[25] B. Peterson. Learning to See Creatively: Design, Color and Composition in Photog-raphy. Crown Publishing Group, 2003.

[26] S. Rogers. Level up!: The guide to great video game design, 2010.

[27] K Salen and E. Zimmerman. Rules of play: game design fundamentals. UniversityPress Group Limited, 2004.

[28] J. Schell. The art of game design: a book of lenses, 2008.

[29] Ara Shirinian. The uneasy merging of narrative and gameplay, January 2010.

[30] C. Solarski and T. Donavan. Drawing Basics and Video Game Art: Classic to CuttingEdge Art Techniques for Winning Game Design. Watson-Guptill Publications, 2012.

[31] ThatGameCompany. Journey. [Download], 2012.

[32] thechineseroom. Dear esther. [Steam], 2012.

[33] Noah Wardrip-Fruin and Pat Harrigan. First Person: New Media as Story, Perfor-mance, and Game. The MIT Press, 2006.

[34] M. Worch. Player stories and designer stories. Presented at the 2012 Game Develop-ers Conference.

[35] Jose P. Zagal. A framework for games literacy and understanding games, 2008.

[36] Jichen Zhu, Kenneth Ingraham, and J. Michael Moshell. Back-leading through char-acter status in interactive storytelling. In Proceedings of the 4th international confer-ence on Interactive Digital Storytelling, ICIDS’11, pages 31–36, Berlin, Heidelberg,2011. Springer-Verlag.