Making Management Decisions The Role of Intuition and Emotion; Herbert Simon (1987).pdf

8/14/2019 Making Management Decisions The Role of Intuition and Emotion; Herbert Simon (1987).pdf

1/8

February, 19H7

Making Management Decisions:the Role of Intuition and Emotion

By Herbert A. SimCarnegie-Mellon University

Thework of a manager includes making decisions(o r participating in their making), communicatingthem to others, an d monitoring how they are carriedout. Managers must know a great deal about the industry and social environment in which they work and thedecision-making process itself to make decisions well.Over the past 40 years, t he technique of decision making has been greatly advanced by the development of awide range of tools

in particular, the tools of opera

tions research and management science, and the technology of expert systems

But these advances have not applied to the entiredomain of decision making. They have ha d their greatest impact on decision making that is well-structured,deliberative, an d quantii ative; they have ha d less impact on decision making that is loosely structured, intuitive, an d qualitative; an d they have ha d the leastimpact on face-to- face interactions between a managerand his or her coworkers the give and take of everyda y work.

In this article, will discuss these two relativelyneglected types of decision making: intuitive decisionmaking and decision making that involves interpersonal interaction. What, sf anything, do we know aboutho w judgmental an d intuitive processes work and howthey can be made to work better? And why do managers often fail to do what they know they shoulddoeven what they have decided to do? What can bedone to bring action into closer accord with intention?

My article will therefore have the form of a diptych, with one half devoted to each of these topics.First, will discuss judgmental an d intuitive decisionmaking; then will turn to the subject of the manager'sbehavior and the influence of emotions on thatbehavior.

Sometimes the terra rational (or logical) is applied to decision making that is consciously analytic,the term nonrational to decision making that is intuitive an d judgmental, and the term irrational to decision making and behavior that responds to the emotions or that deviates from action chosen rationally.We will be concerned, then, with the nonrational an dthe irrational components of managerial decision making and behavior. Our task, you might say, is to discover the reason that underlies unreason.

Intuition and Judgment

As an appendix to the Functions of the Execut(Harvard University Press, 1938), Chester I. Barnpublished an essay, based on a talk he ha d given1936 at Princeton, entitled Mind in Everydayfairs. 1 The central motif of that essay was a contrabetween what Barnard called logical an d nonlogcal processes for making decisions. He speaks of twide divergence of opinion .. . as to what constituteproper intellectual basis for opinion or deliberatetion. And he continues:

By logical processes mean consciothinking which could be expressed in woror by other symbols, that is, reasoning. B non-logical processes mean those no t cpable of being expressed in words or as resoning, which are only made known byjudgment,

decisionor action.

Barnard's thesis was that executives, as contrastsay, with scientists, do not often enjoy the luxurymaking their decisions on the basis of orderly rationanalysis, but depend largely on intuitive or judgmentaresponses to decision-demanding situations.

Although Barnard did not provide a set of formcriteria for distinguishing between logical and jumental decision making, he did provide a phenomelogical characterization of the two styles that mthem easily recognizable, at least in their more extremforms. In logical decision making, goals an d alterntives are made explicit, the consequences of pursuindiiferent alternatives are calculated, an d these conquences are evaluated in terms of ho w close they arethe goals.

In judgmental decision making, the responsethe need for a decision is usually rapid, too rapidallow for an orderly sequential analysis of the situatioan d the decision maker cannot usually give a veridicaccount of either the process by which the decision wreached or the grounds for judging it correct. Nev

57

Copyright 2001 . All rightsreserved.


2/8

Academy of Management EXECUTIVE

theless, decision makers may have great confidence inthe correctness of their intuitive decisions an d arelikely to attribute their ability to make them rapidly totheir experience.

Most executives probably find Barnard's accountof their decision processes persuasive; it captures theirown feelings of how processes work. On the other hand,some students of management, especially those whosegoal is to improve management-decision processes,have felt less comfortable with it . It appears to vindicate snap judgments an d to cast doubt on the relevance of management-science tools, which almost allinvolve deliberation an d calculation in decisionmaking.

Barnard did not regard the nonlogical processesof decision as magical in any sense. On the contrary, hefelt they were grounded in knowledge and experience:

The sources of these non-logical processes lie

in physiological conditions or factors, or inthe physical and social environment, mostlyimpressed upon us unconsciously or withoutconscious effort on our part. They also consist of the mass of facts, patterns, concepts,techniques, abstractions, and generally whatwe call formal knowledge or beliefs, whichare impressed upon our minds more or lessby conscious effort and study. This secondsource of non-logical mental processesgreatly increases with directed experience,

study and education. P 302)

At the time wrote Administrative Behavior(1941-42), was troubled by Barnard's account of intuitive judgment (see the footnote on p. 51 of AB),largely, think, because he left no clues as to what subconscious processes go on while judgments are beingmade. 2 was wholly persuaded, however, that a theoryof decision making ha d to give an account of both conscious and subconscious processes (see the end of p. 75to the top of p. 76). finessed the issue by assumingthat both the conscious an d the unconscious parts ofthe process were the same, that they involve drawingon factual premises an d value premises, and operatingon them to form conclusions that became the decisions.

Because used logic (drawing conclusions frompremises) as a central metaphor to describe the decision-making process, many readers of AdministrativeBehavior have concluded that the theory advancedthere applies only to logical decision making, not todecisions that involve intuition and judgment. Thatwas certainly not my intent. But now, after nearly 50years, the ambiguity can be resolved because we haveacquired a solid understanding of what the judgmentaland intuitive processes are. will take up the new evidence in a moment; but first, a word must be saidabout the two brains hypothesis, which argues that

rational and intuitive processes are so different thathey are carried out in different parts of the brain.

Split Brains and Forms of Thought

Physiological research on split brains brainin which the corpus callosum, which connects the twhemispheres of the cerebrum, has been severed haprovided encouragement to the idea of two qualitatively different kinds of decision making the analytcal, corresponding to Barnard's logical, and the intuitive or creative, corresponding to his non-logical.The primary evidence behind this dichotomy is thathe two hemispheres exhibit a division of labor:right-handed people, the right hemisphere plays a special role in the recognition of visual patterns, and thleft hemisphere in analytical processes an d the us elanguage.

Other evidence in addition to the split-brain rsearch suggests some measure of hemispheric specialzation. Electrical activity in the intact brain can bmeasured by EEG techniques. Activity in a braihemisphere is generally associated with partial or totasuppression in the hemisphere of the alpha system,salient brain wave with a frequency of about ten vibrtions per second. When a hemisphere is inactive, thalpha rhythm in that hemisphere becomes strong. Fomost right-handed subjects, when the brain is engagein a task involving recognition of visual pattern, the apha rhythm is relatively stronger in the left than in thright hemisphere; with more analytical tasks, the alphrhythm is relatively stronger in the right hemispher(See Doktor and Hamilton, 1973, an d Doktor, 1975, fosome experiments an d a review of the evidence.3)

The more romantic versions of the split-braindoctrine extrapolate this evidence into the two polaforms of thought labeled above as analytical an d creative. As an easy next step, evaluative nuances creepinto the discussion. The opposite of creative, afteall, is pedestrian. The analytical left hemisphere,this story goes, carries on the humdrum, practical, evryday work of the brain, while the creative right hemsphere is responsible for those flights of imaginationthat produce great music, great literature, great argreat science, and great management. The evidence fthis romantic extrapolation does not derive from thphysiological research. As indicated above, that rsearch has provided evidence only for some measurespecialization between the hemispheres. It does notany way imply that either hemisphere (especially thright hemisphere) is capable of problem solving, decsion making, or discovery independent of the otheTh e real evidence for two different forms of thoughtessentially that on which Barnard relied: the observat ion that, in everyday affairs, men an d women oftemake competent judgments or reach reasonable decsions rapidly without evidence indicating that thehave engaged in systematic reasoning, and withoutheir being able to report the thought processes thatook them to their conclusion.

58



3/8

February, 19H7

There is also some evidence for the very plausiblehypothesis that some people, confronted with a particular problem, make more use of intuitive processes insolving it , while other people make relatively more useof analytical processes (Doktor, 1978). 3

For our purposes, it is the differences in behavior,and not the differences in the hemispheres, that areimportant. Reference to the two hemispheres is a redherring that can only impede our understanding of intuitive, non-logical thought. The important questions for us are What is intuition? an d How is itacccomplished? not I n which cubic centimeters ofthe brain tissue does it take place?

New Evidence on the Processes ofIntuition

In the 50 years since Barnard talked about themind in everyday affairs, we have learned a great dealabout the processes human beings use to solveproblems, to make decisions, and even to create worksof art an d science. Some of this ne w knowledge hasbeen gained in the psychological laboratory; some hasbeen gained through observation of the behavior ofpeople who are demonstrably creative in some realm ofhuman endeavor; and a great deal has been gainedthrough the use of the modern digital computer tomodel human thought processes and perform problem-solving and decision-making functions at expert levels.

should like to examine this body of research,which falls under the labels of cognitive science and artificial intelligence, to see what light it casts on intuitive, judgmental decision making in management.We will see that a rather detailed account can be givenof the processes that underlie judgment, even thoughmost of these processes are not within the consciousawareness of the actor using them.

The Expert's IntuitionIn recent years, the disciplines of cognitive sci

ence and artificial intelligence have devoted a greatdeal of attention to the nature of expert problem solving and decision making in professional-level tasks.The goal of the cognitive science research has been togain an understanding of the differences between thebehavior of experts and novices, and possibly to learnmore about how novices can become experts. The goalof the artificial intelligence research has been to buildcomputer systems that can perform professional tasksas competently as human experts can. Both lines of research have greatly deepened our understanding ofexpertise.4

Intuition in Chessplaying

One much studied class of experts isgrandmasters in the game of chess. Chess is usuallylieved to require a high level of intellect, agrandmasters are normally full-time professionals whave devoted many years to acquiring their masterythe game. From a research standpoint, the advantaof the game is that the level of skill of players cancalibrated accurately from their official ratings, baon their tournament success.

From the standpoint of studying intuitive thining, chess might seem (at least to outsiders) anpromising research domain. Chess playing is thoughtinvolve a highly analytical approach, with playeworking out systematically the consequences of movand countermoves, so that a single move may takemuch as a half hour's thought, or more. On the othand, chess professionals can play simultaneous gamesometimes against as many as 50 opponents, an dhibit only a moderately lower level of skill thangames playing under tournament conditions. In simtaneous play, the professional takes much less thanminute, often only a few seconds, for each move. Theis no time for careful analysis.

When we ask the grandmaster or master ho wor she is able to find good moves under these circustances, we get the same answer that we get from othprofessionals who are questioned about rapid decisioIt is done by intuition, by applying one's professio judgment to the situation. A few seconds' glancethe position suggests a good move, although the playhas no awareness of how the judgment was evoked

Even un r tournament conditions, good movusually come to a player's mind after only aseconds' consideration of the board. The remainderthe analysis time is generally spent verifying thatmove appearing plausible does not have a hiddweakness. We encounter this same kind of behavioother professional domains where intuitive judgmentare usually subjected to tests of various kinds befthey are actually implemented. The main exceptiare situations where the decision has to be made befa deadline or almost instantly. Of course we know tunder these circumstances (as in professional chwhen the allowed time is nearly exhausted), mistaare sometimes made.

How do we account for the judgment or intuitithat allows the chess grandmaster usually to find gmoves in a few seconds? A good deal of the answerbe derived from an experiment that is easily repeatFirst, present a grandmaster and a novice with a potion from an actual, but unfamiliar, chess game (wabout 25 pieces on the board). After five orseconds, remove the board an d pieces and ask the s

51



4/8


jects to reproduce it . Th e grandmaster will usually reconstruct the whole position correctly, an d on averagewill place 23 or 24 pieces on their correct squares. Thenovice will only be able to replace, on average, about 6pieces.

I t might seem that we are witnessing remarkableskill in visual imagery an d visual memory, but we caneasily dismiss that possibility by carrying out a secondexperiment. The conditions are exactly the same as inthe first experiment, except that now the 25 pieces areplaced on the board at random. The novice can stillreplace about 6 pieces and the grandmaster about 6The difference between them in the first experimentdoes not lie in the grandmaster's eyes or imagery, butin his knowledge, acquired by long experience, of thekinds of patterns and clusters of pieces that occur onchessboards in the course of games. For the expert,such a chess board is not an arrangement of 25 piecesbut an arrangement of a half dozen familiar patterns,recognizable old friends. On the random board thereare no such patterns, only the 25 individual pieces inan unfamiliar arrangement.

The grandmaster's memory holds more than a setof patterns. Associated with each pattern in his or hermemory is information about the significance of thatpattern what dangers it holds, and what offensive ordefensive moves it suggests. Recognizing the patternbrings to the grandmaster's mind at once moves thatmay be appropriate to the situation. It is this recognition that enables the professional to play very strongchess at a rapid rate. Previous learning that has storedthe patterns and the information associated with themin memory makes this performance possible. This,then, is the secret of the grandmaster's intuition orjudgment.

Estimates have been made, in a variety of ways,of the number of familiar patterns (which psychologistsnow call chunks) that the master or grandmaster mustbe able to recognize. These estimates fall in the neighborhood of 50,000, give or take a factor of two. Is this alarge number? Perhaps not. Th e natural language vocabularies of college graduates have been estimated tobe in the range of 50,000 to 200,000 words, nearly thesame range as the chess expert's vocabularies of patterns of pieces. Moreover, when we recognize a word,we also get access to information in our memoriesabout the meaning of the w ord a nd to other information associated with it as well. So our ability to speakand unders tand language has the same intuitive orjudgmental flavor as the grandmaster's ability to playchess rapidly.

Intuition in Computerized ExpertSystems

A growing body of evidence from artificial intelligence research indicates that expert computer systems,capable of matching human performance in some limited domain, can be built by storing in computer memory tens of thousands of productions. Productions are

computer instructions that take the form of if-thenpairs. Th e if is a se t of conditions or patterns to brecognized; the then is a body of information associated with the if an d evoked from memory whenevethe pattern is recognized in the current situation.

Some of our best data about this organization o

expert knowledge come from the areas of medical diagnosis. Systems like Caduceus an d Mycin consist oflarge number of such if-then pairs, together with an inference machine of modest powers. These systems arcapable of medical diagnosis at a competent clinicallevel within their respective limited domains. Theirrecognition capabilities, the if-then pairs, representtheir intuitive or judgmental ability; their inferencingpowers represent their analytical ability.

Medical diagnosis is just one of a number of domains for which expert systems have been built. Fomany years, electric motors, generators, an d transformers have been designed by expert systems developed blarge electrical manufacturers. These computer programs have taken over from professional engineermany standards and relatively routine design taskThey imitate fairly closely the rule-of-thumb procedures that human designers have used, the result oflarge stock of theoretical an d practical informationabout electrical machinery. Recognition also playslarge role in these systems. For example, examinatioof the customer's specifications reminds the programof a particular class of devices, which is then used athe basis for the design. Parameters for the design arthen selected to meet the performance requirements othe device.

In chemistry, reaction paths for synthesizing oganic molecules can be designed by expert systems. I

these systems, the process appears relatively analytic,for it is guided by reasoning in the form of means-endanalyses, which work backward from the desired molecule, via a sequence of reactions, to available rawmaterials. But the reasoning scheme depends on a largestore of knowledge of chemical reactions an d the abiity of the system to recognize rapidly that a particularsubstance can be obtained as the output of one or morfamiliar reactions. Thus, these chemical synthesis programs employ the same kind of mixture of intuitionand analysis that is used in the other expert systems,and by human experts as well.

Other examples of expert systems can be citean d all of them exhibit reasoning or analytic processes

combined with processes for accessing knowledbanks with the help of recognition cues. This appearsto be a universal scheme for the organization of expertsystems an d of expert human problem solving as wel

Notice that there is nothing irrational about intuitive or judgmental reasoning based on productions.The conditions in a production constitute a setpremises. Whenever these conditions are satisfied, thproduction draws the appropriate conclusionevokes from memory information implied by these conditions or even initiates moto r responses. A personlearning to drive a car may notice a red light, be awarethat a red light calls for a stop, and be aware that stop

60



5/8

February, 1987

ping requires applying the brakes. For an experienceddriver, the sight of the red light simply evokes the application of brakes. How conscious the actor is of theprocess inversely, how automatic the response is, maydiffer, but there is no difference in the logic beingapplied.

Intuition in Management

Some direct evidence also suggests that the intuitive skills of managers depend on the same kinds ofmechanisms as the intuitive skills of chessmasters orphysicians. It would be surprising if it were otherwise.The experienced manager, too, has in his or he r memory a large amount of knowledge, gained from trainingand experience and organized in terms of recognizablechunks and associated information.

Marius J. Bouwman has constructed a computerprogram capable of detecting company problems froman examination of accounting statements.5 Th e program was modeled on de lailed thinking-aloud protocolsof experienced financial analysts interpreting suchstatements, an d it captures the knowledge that enablesanalysts to spot problems intuitively, usually at a veryrapid rate. When a comparison is made between the responses of the program and the responses of an experthuman financial analyst, a close match is usuallyfound.

In another study, R. Bhaskar gathered thinking-aloud protocols from business school students and experienced businessmen, wh o were all asked to analyze abusiness policy case The final analyses produced bythe students and the businessmen were quite similar.What most sharply discriminated between the novicesan d the experts was the time required to identify thekey features of the case. This was done very rapidly,with the usual appearances of intuition, by the experts;it wa s done slowly, with much conscious and explicitanalysis, by the novices.

These two pieces of research are just drops ofwater in a large bucket that needs filling. The description, in detail, of the use of judgmental an d analyticalprocesses in expert problem solving and decision making deserves a high priority in the agenda of management research.

Can Judgment Be Improved?

From this an d other research on expert problemsolving and decision making, we can draw two mainconclusions. First, experts often arrive at problem diagnoses and solutions rapidly and intuitively without being able to report how they attained the result. Second,this ability is best explained by postulating a recognition and retrieval process that employs a large numbergenerally tens of thousands or even hundreds ofthousands of chunks or patterns stored in long termmemory.

W hen the problems to be solved are more thantrivial, the recognition processes have to fie organized

in a coherent way an d they must be supplied with resoning capabilities that allow inferences to be drawfrom the information retrieved, and numerous chunkof information to be combined. Hence intuition is notprocess that operates independently of analysis; ratherthe two processes are essential complementary compo

nents of effectivedecision-making systems.

When thexpert, is solving a difficult problem or making a complex decision, much conscious deliberation may be ivolved. But each conscious step may itself constituteconsiderable leap, with a whole sequence of automateproductions building the bridge from the premisesthe conclusions. Hence the expert appears to take gianintuitive steps in reasoning, as compared with the tinsteps of the novice.

It is doubtful that we will find two types of maagers (a t least, of good managers), one of whom relialmost exclusively on intuition, the other on analytictechniques. More likely, we will find a continuumdecision -making styles involving an intimate combin

tion of the two kinds. of skill. We will likely also finthat the nature of the problem to be solved will beprincipal determinant of the mix.

With our growing understanding of the organiztion of judgmental and intuitive processes, of the spcific knowledge that is required to perform particularjudgmental tasks, and of the cues that evoke suknowledge in situations in which it is relevant, we haa powerful new tool for improving expert judgmenWe can specify the knowledge and the recognition cpabilities that experts in a domain need to acquire asbasis for designing appropriate learning procedures.

We can also, in more an d more situations, desiexpert systems capable of automating the expertise,

alternatively, of providing the human decision makwith an expert consultant. Increasingly, we will see dcision aids for managers that will be highly interactive,with both knowledge and intelligence being shared btween the human and the automated componentsthe system.

A vast research an d development task of etracting an d cataloging the knowledge and cues usby experts in different kinds of managerial tasks lahead. Much has been learned in the past few yearabout how to do this. More needs to be learned abohow to update and improve the knowledge sourcesexpert systems as new knowledge becomes available

Progress will be most rapid with expert system

that have a substantial technical component. It isaccident that the earliest expert systems were built fsuch tasks as designing motors, making medical dianoses, playing chess, an d finding chemical synthespaths. In the area of management, the analysis of company financial statements is a domain where some prgress has been made in constructing expert systemsThe areas of corporate policy and strategy are excellecandidates for early development of such systems.

What about the aspects of executive work that ivolve Ihe managing of people? What help can we epec , in improving this crucial component of the maagement task?

61



6/8


Knowledge and Behavior

What managers know they should dowhetherby analysis or intuitively is very often different from

whatthey actually

do.A common failure of managers,

which all of us have observed, is the postponement ofdifficult decisions. What is it that makes decisions difficult and hence tends to cause postponement? Often,the problem is that all of the alternatives have unde-sired consequences. When people have to choose thelesser of two evils, they do not simply behave likeBayesian statisticians, weighing the ba d against theworse in the light of their respective possibilities. Instead, they avoid the decision, searching for alternatives that do not have negative outcomes. If such alternatives are not available, they are likely to continue topostpone making a choice. A choice between un desirables is a dilemma, something to be avoided or

evaded.Often, uncertainty is the source of the difficulty.Each choice may have a good outcome under on e set ofenvironmental contingencies, but a bad outcome underanother. When this occurs, we also do not usually observe Bayesian behavior; the situation is again treatedas a dilemma.

The ba d consequences of a manager's decision areoften bad for other people. Managers sometimes haveto dismiss employees or, even more frequently, have tospeak to them about unsatisfactory work. Dealing withsuch matters face to face is stressful to many, perhapsmost, executives. The stress is magnified if the employee is a close associate or friend. If the unpleasant

task cannot be delegated, it may be postponed.Th e manager wh o has made a mistake (that is tosay, all of us at one time or another) also finds himselfor herself in a stressful situation. The matter must bedealt with sooner or later, but why not later instead ofsooner? Moreover, when it is addressed, it can be approached in different ways. A manager may try toavoid blame It wasn't my fault A different way isto propose a remedy to the situation. know of no systematic data on ho w often the on e or the other courseis taken, but most of us could probably agree thatblame-avoiding behavior is far more common thanproblem-solving behavior after a serious error has beenmade.

The Consequences of Stress

What all of these decision-making situations havein common is stress, a powerful force that can divertbehavior from the urgings of reason. They are examples of a much broader class of situations in whichmanagers frequently behave in clearly nonproductiveways. Nonproductive responses are especially commonwhen actions have to be made under time pressure.The need to allay feelings of guilt, anxiety, an d embarrassment may lead to behavior that produces tempo

rary personal comfort at the expense of ba d long-ruconsequences for the organization.

Behavior of this kind is intuitive in the senthat it represents response without careful analysis acalculation. Lying, for example, is much more often tresult of panic than of Machiavellianscheming. Th etuition of the emotion-driven manager is very differenfrom the intuition of the expert whom we discussearlier. The latter's behavior is the product of learninand experience, and is largely adaptive; the former'behavior is a response to more primitive urges, andmore often than not inappropriate. We must not cofuse the nonrational decisions of the experts tdecisions that derive from expert intuition an d judment with the irrational decisions that stressful emtions may produce.

have made no attempt here to produce a coprehensive taxonomy of the pathologies of organiztional decision making, but simply have given someamples of the ways that stress interacts with cognitioto elicit counterproductive behavior. Such responsescan become so habitual for individuals or even forganizations that they represent a recognizable managerial style.

Organizational psychologists have a great dealsa y about ways of motivating workers an d executivesdirect their efforts toward organizational goals. Thhave said less about ways of molding habits so thatecutives can handle situations in a goal-directed maner. When it comes to handling situations, two dimesions of behavior deserve particular attention: tresponse to problems that arise, an d the initiationactivity that looks to the future.

Responding to Problems

Th e response of an organization to a problemdifficulty, whether it results from a mistake or soother cause, is generally one that looks both backwan d forward. It looks backward to establish responbility for the difficulty an d to diagnose it , an d forwato find a course of action to deal with it.

Th e backward look is an essential part of theganization's reward system. The actions that haveto difficulties, an d the people responsible for thosetions, need to be identified. But the backward look calso be a source of serious pathologies. Anticipation

it particularly

anticipationthat

it willbe

actedon

a punitive way is a major cause for the concealmof problems until they can no longer be hidden. It calso be highly divisive, as individuals point fingerstransfer blame to others. Such outcomes can hardlyeliminated, but an organization's internal reputationfor fairness an d objectivity can mitigate them. So canpractice of subordinating the blame finding to a dianosis of causes as a first step toward remedial action

Most important of all, however, is the forwarlook: the process of defining the problem and identifying courses of action that ma y solve it . Here also treward system is critically important. Readiness

62



7/8

February, 1987

search for problem situations an d effectiveness in finding them need to be recognized and rewarded.

Perhaps the greatest influence a manager canhave on the problem-solving style of the organizationas a role model is making the best responses toproblems. Th e style the manager should aim for rests

onthe following

principles:1. Solving the problem takes priority over lookingbackward to its causes. Initially, backward looks shouldbe limited to diagnosing causes; fixing responsibility formistakes should be postponed until a solution is beingimplemented.

2. The manager accepts personal responsibilityfor finding and proposing solutions instead of seekingto shift that responsibility either to superiors or to subordinates, although the search for solutions may, ofcourse, be a collaborative effort involving many people.

3. The manager accepts personal responsibilityfor implementing action solutions, including securingthe necessary authority from above if required.

4. When it is time to look backward, fixing blamemay he an essential part of the process, but the primary focus of attention should be on what can belearned to prevent similar problems from arising in thefuture.

These principles are as obvious as the Ten Commandments and perhaps not quite as difficult to obey.Earlier, indicated that stress might cause departuresfrom them, but failure to respond effectively toproblems probably derives more from a lack of attention an d an earlier failure to cultivate the appropriatehabits. The military makes much use of a procedurecalled Estimate of the Situation. Its value is not thatit teaches anything esoteric, but that through continual

training in its use, commanders become habituated toapproaching situations in orderly ways, using thechecklists provided by the formal procedure.

Habits of response to problems are taught an dlearned both in the manager's one-on-one conversations with subordinates and in staff meetings. Is attention brought back repeatedly to defining the problemsuntil everyone is agreed on just what the problem is? Isattention then directed toward generating possible solutions an d evaluating their consequences? The leastoften challenged an d most reliable base of managerialinfluence is the power to set the agenda, to focus attention. It is one of the most effective tools the managerhas for training organization members to approach

problems constructivelyby

shaping their own habits ofattention.The perceptive reader will have discerned that

shaping habits of attention is identical to acquiringintuitions. The habit of responding to problems bylooking for solutions can an d must become intuitive cued by the presence of the problem itself. Aproblem-solving style is a component of the set of intuitions that the manager acquires, one of the key components of effective managerial behavior.

Looking to the Future

With respect to the initiation of activity, the orgnizational habit we would like to instil is responsivness to cues that signal future difficulties as well asthose that call attention to the problems of the m

ment. Failureto

give sufficient attention to the futurmost often stems from two causes. The first is interuption by current problems that have more proximatedeadlines and hence seem more urgent; the secondthe absence of sufficient scanning activity that cpick up cues from the environment that long-run forcno t impinging immediately on the organization haimportance for it in the future.

In neither case is the need for sensitivity to tfuture likely to be met simply by strengthening intitions. Rather, what is called for is deliberate an d sytematic allocation of organizational resources to dewith long-range problems, access for these resourcesappropriate input from the environment that will a

tract their attention to new prospects, an d protectionof these planning resources from absorption in currenproblems, however urgent they may be. Attentionthe future must be institutionalized; there is no simpleway to incorporate it into managerial style or habi

It is a fallacy to contrast analytic an d intuitive styles of management. Intuition an d judmentat least good judgmentare simply analysfrozen into habit and into the capacity for rapid rsponse through recognition. Every manager needs toable to analyze problems systematically (and with taid of the modern arsenal of analytical tools providedby management science an d operations research).Every manager needs also to be able to respond to sit

ations rapidly, a skill that requires the cultivationintuition and judgment over many years of experienand training. Th e effective manager does not have tluxury of choosing between analytic an d intuitiveapproaches to problems. Behaving like a managermeans having command of the whole range of managment skills an d applying them as they become apprpriate'.

Herbert A. Simon is Richard King Mellon University Professor of computer scienceand psychology at Carnegie-Mellon University,where he has taught since 1949. For the past 30years he has been studying decision-makingand problem-solving processes, using computersto simulate human thinking.

Dr Simon was educated at the Universityof Chicago (where he received his Ph.D.). Hewas elected to the National Academy of Sciences in 1967, and ha s received awards for hisresearch from the American Psychological Association, the American Economic Association,and the Institute of Electrical and Electronic

63

Copyright 2001 . Allrightsreserved.


8/8


Engineers, among others. He received the Alfred Nobel Memorial Prize in Economics in1978 and the National Medal of Science in1986. Dr. Simon ha s been Chairman of theBoard of Directors of the Social Science Research Council and of the Behavioral ScienceDivision of the National Research Council, andwas a member of the President's Science Advisory Committee.

Dr. Simon ha s published over 60 0 papersand 20 books and monographs.

ENDNOTES1. Chester I. Barnard's (1938) The Functions of the

Executive (Cambridge, Mass.: Harvard University Press),contains the essay on the contrast between logical an dnonlogical processes as bases for decision making.

2. Simon, H. A. (1978) Administrative Behavior, 2nded. Ne w York: Free Press.

For a review of the artificial intelligence research on expert systems, see A. Barr an d E. A. Figenbaum's (eds.) Th e

Handbook of ArtificialIntelligence, Vol. 2, Los Alamos, CWilliam Kaufmann, 1982, pp. 77-294.

3. Two works that examine the split brain theoryforms of thought are R. H. Doktor's Problem Solving Styof Executives an d Management Scientists, in A. CharW. W. Cooper, an d R. J. Niehaus's (eds.) Managementence Approaches to Manpower Planning and OrganizatioDesign (Amsterdam: North-Holland, 1978); an d R. H. Dokan d W. F. Hamilton's Cognitive Style and the AcceptancManagement Science Recommendations (Managementence, 19:884-894, 1973).

4. For a survey of cognitive science research on probsolving an d decision making, see Simon, H. A. (1979)Sciences of the Artificial, 2n d ed., Cambridge, Mass.:MIT Press, Chapters 3 and 4.

5. Marius J. Bouwman's doctoral dissertation, Fincial Diagnosis (Graduate School of Industrial Administion, Carnegie-Mellon University, 1978).

6. R. Bhaskar's doctoral dissertation, Problem Solvin Semantically Rich Domains (Graduate School of Intrial Administration, Carnegie-Mellon University, 1978).

64

Making Management Decisions The Role of Intuition and Emotion; Herbert Simon (1987).pdf

Documents

Transcript of Making Management Decisions The Role of Intuition and Emotion; Herbert Simon (1987).pdf