Professional Documents
Culture Documents
This Content Downloaded From 130.209.6.61 On Mon, 10 Jan 2022 09:01:31 UTC
This Content Downloaded From 130.209.6.61 On Mon, 10 Jan 2022 09:01:31 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
is collaborating with JSTOR to digitize, preserve and extend access to The American Economic
Review
By DANIEL KAHNEMAN*
The work cited by the Nobel committee was hopes have been realized to some extent, giving
done jointly with Amos Tversky (1937-1996) rise to an active program of research by behav-
during a long and unusually close collaboration. ioral economists (Thaler, 2000; Colin Camerer
Together, we explored the psychology of intu- et al., forthcoming; for other examples, see
itive beliefs and choices and examined their Kahneman and Tversky, 2000).
bounded rationality. Herbert A. Simon (1955,My work with Tversky comprised three sep-
1979) had proposed much earlier that decisionarate programs of research, some aspects of
makers should be viewed as boundedly rational, which were carried out with other collaborators.
and had offered a model in which utility maxi- The first explored the heuristics that people use
mization was replaced by satisficing. Our re- and the biases to which they are prone in vari-
search attempted to obtain a map of boundedous tasks of judgment under uncertainty, includ-
rationality, by exploring the systematic biases ing predictions and evaluations of evidence
(Kahneman and Tversky, 1973; Tversky and
that separate the beliefs that people have and the
choices they make from the optimal beliefs and Kahneman, 1974; Kahneman et al., 1982). The
choices assumed in rational-agent models. Thesecond was concerned with prospect theory, a
rational-agent model was our starting point and model of choice under risk (Kahneman and
the main source of our null hypotheses, but
Tversky, 1979; Tversky and Kahneman, 1992)
Tversky and I viewed our research primarily as
and with loss aversion in riskless choice (Kah-
a contribution to psychology, with a possibleneman et al., 1990, 1991; Tversky and Kahne-
contribution to economics as a secondary ben-man, 1991). The third line of research dealt with
efit. We were drawn into the interdisciplinary
framing effects and with their implications for
conversation by economists who hoped that rational-agent models (Tversky and Kahneman,
psychology could be a useful source of assump-
1981, 1986). The present essay revisits these
tions for economic theorizing, and indirectlythree
a lines of research in light of recent ad-
source of hypotheses for economic researchvances in the psychology of intuitive judgment
(Richard H. Thaler, 1980, 1991, 1992). These
and choice. Many of the ideas presented here
were anticipated informally decades ago, but
the attempt to integrate them into a coherent
approach to judgment and choice is recent.
t This article is a revised version of the lecture Daniel
Economists often criticize psychological re-
Kahneman delivered in Stockholm, Sweden, on December
8, 2002, when he received the Bank of Sweden Prize insearch for its propensity to generate lists of
errors and biases, and for its failure to offer a
Economic Sciences in Memory of Alfred Nobel. The article
coherent alternative to the rational-agent model.
is copyright ? The Nobel Foundation 2002 and is published
here with the permission of the Nobel Foundation. This complaint is only partly justified: psycho-
logical theories of intuitive thinking cannot
* Woodrow Wilson School, Princeton University,
Princeton, NJ 08544 (e-mail: Kahneman@princeton.edu).
match the elegance and precision of formal nor-
This essay revisits problems that Amos Tversky andmative
I models of belief and choice, but this is
studied together many years ago, and continued to discuss in
just
a conversation that spanned several decades. It builds on an another way of saying that rational models
are
psychologically unrealistic. Furthermore,
analysis of judgment heuristics that was developed in col-
the
alternative to simple and precise models is
laboration with Shane Frederick (Kahneman and Frederick,
2002). A different version was published in American Psy- not
chaos. Psychology offers integrative con-
chologist in September 2003. For detailed comments on this
cepts and mid-level generalizations, which gain
version I am grateful to Angus Deaton, David Laibson,
credibility from their ability to explain ostensi-
Michael Rothschild, and Richard Thaler. The usual caveats
bly different phenomena in diverse domains. In
apply. Geoffrey Goodwin, Amir Goren, and Kurt Schoppe
provided helpful research assistance. this spirit, the present essay offers a unified
1449
treatment of intuitive judgment and choice, ations and overt behavior also goes on. We do
not express every passing thought or act on
which builds on an earlier study of the relation-
ship between preferences and attitudes (Kahne-every impulse. But the monitoring is normally
man et al., 1999) and extends a model of lax, and allows many intuitive judgments to be
judgment heuristics recently proposed by Kah- expressed, including some that are erroneous
neman and Shane Frederick (2002). The guid- (Kahneman and Frederick, 2002). Ellen J.
ing ideas are (i) that most judgments and most Langer et al. (1978) provided a well-known
choices are made intuitively; (ii) that the rules example of what she called "mindless behav-
that govern intuition are generally similar to theior." In her experiment, a confederate tried to
rules of perception. Accordingly, the discussion cut in line at a copying machine, using various
of the rules of intuitive judgments and choices preset "excuses." The conclusion was that state-
will rely extensively on visual analogies. ments that had the form of an unqualified re-
Section I introduces a distinction between quest were rejected (e.g., "Excuse me, may I use
two generic modes of cognitive function, corre- the Xerox machine?"), but almost any statement
sponding roughly to intuition and reasoning. that had the general form of an explanation was
Section II describes the factors that determine accepted, including "Excuse me, may I use the
the relative accessibility of different judgments Xerox machine because I want to make cop-
and responses. Section III relates prospect the-ies?" The superficiality is striking.
ory to the general proposition that changes and Frederick (2003, personal communication)
differences are more accessible than absolute has used simple puzzles to study cognitive self-
values. Section IV explains framing effects in monitoring, as in the following example: "A bat
terms of differential salience and accessibility. and a ball cost $1.10 in total. The bat costs $1
Section V reviews an attribute substitution more than the ball. How much does the ball
model of heuristic judgment. Section VI de-cost?" Almost everyone reports an initial ten-
scribes a particular family of heuristics, called dency to answer "10 cents" because the sum
prototype heuristics. Section VII discusses the $1.10 separates naturally into $1 and 10 cents,
interactions between intuitive and deliberate and 10 cents is about the right magnitude. Fred-
thought. Section VIII concludes. erick found that many intelligent people yield to
this immediate impulse: 50 percent (47/93) of a
I. The Architecture of Cognition: Two Systems group of Princeton students and 56 percent
(164/293) of students at the University of Mich-
The present treatment distinguishes two igan gave the wrong answer. Clearly, these re-
modes of thinking and deciding, which corre- spondents offered their response without first
spond roughly to the everyday concepts of rea- checking it. The surprisingly high rate of errors
soning and intuition. Reasoning is what we do in this easy problem illustrates how lightly the
when we compute the product of 17 by 258, fill output of effortless associative thinking is mon-
an income tax form, or consult a map. Intuition itored: people are not accustomed to thinking
is at work when we read the sentence "Bill hard, and are often content to trust a plausible
Clinton is a shy man" as mildly amusing, or judgment that quickly comes to mind. Re-
when we find ourselves reluctant to eat a piece markably, Frederick has found that errors in
of what we know to be chocolate that has been this puzzle and in others of the same type
formed in the shape of a cockroach (Paul Rozin were significant predictors of high discount
and Carol Nemeroff, 2002). Reasoning is done rates.
deliberately and effortfully, but intuitive thoughts In the examples discussed so far, intuition
seem to come spontaneously to mind, without was associated with poor performance, but in-
conscious search or computation, and without tuitive thinking can also be powerful and accu-
effort. Casual observation and systematic re- rate. High skill is acquired by prolonged
search indicate that most thoughts and actions practice, and the performance of skills is rapid
are normally intuitive in this sense (Daniel T. and effortless. The proverbial master chess
Gilbert, 1989, 2002; Timothy D. Wilson, 2002; player who walks past a game and declares
Seymour Epstein, 2003). "white mates in three" without slowing is per-
Although effortless thought is the norm, forming intuitively (Simon and William G.
some monitoring of the quality of mental oper- Chase, 1973), as is the experienced nurse who
Fast Slow
Parallel Serial
Automatic Controlled
O Effortless Effortful
Associative Rule-governed
Slow-learning Flexible
PLr
Emotional Neutral
z
Percepts Conceptual representations
Current stimulation Past, Present and Future
Stimulus-bound Can be evoked by language
L)
I
detects subtle signs of impending heart failureneither cause nor suffer much interference when
(Gary Klein, 1998; Atul Gawande, 2002). combined with other tasks. For example, a driv-
The distinction between intuition and reason-er's ability to conduct a conversation is a sen-
ing has recently been a topic of considerable sitive indicator of the amount of attention
interest to psychologists (see, e.g., Shelly currently demanded by the driving task. Dual
Chaiken and Yaacov Trope, 1999; Gilbert,tasks have been used in hundreds of psycholog-
2002; Steven A. Sloman, 2002; Keith E. ical experiments to measure the attentional de-
Stanovich and Richard F. West, 2002). There is mands of different mental activities (for a
substantial agreement on the characteristics thatreview, see Harold E. Pashler, 1998). Studies
distinguish the two types of cognitive processes,using the dual-task method suggest that the self-
for which Stanovich and West (2000) proposed monitoring function belongs with the effortful
the neutral labels of System 1 and System 2.operations of System 2. People who are occu-
The scheme shown in Figure 1 summarizespied by a demanding mental activity (e.g., at-
these characteristics. The operations of Systemtempting to hold in mind several digits) are
1 are fast, automatic, effortless, associative, andmuch more likely to respond to another task by
often emotionally charged; they are also gov- blurting out whatever comes to mind (Gilbert,
erned by habit, and are therefore difficult to 1989). The phrase that "System 2 monitors the
control or modify. The operations of System 2activities of System 1" will be used here as
are slower, serial, effortful, and deliberatelyshorthand for a hypothesis about what would
controlled; they are also relatively flexible andhappen if the operations of System 2 were dis-
potentially rule-governed. rupted. For example, it is safe to predict that the
The difference in effort provides the most percentage of errors in the bat-and-ball question
useful indications of whether a given mentalwill increase, if the respondents are asked this
process should be assigned to System 1 or Sys- question while attempting to keep a list of
tem 2. Because the overall capacity for mental words in their active memory.
effort is limited, effortful processes tend to dis-
In the language that will be used here, the
rupt each other, whereas effortless processesperceptual system and the intuitive operations
intuitive evaluations of outcomes are also expected utility of wealth (the moral expecta-
reference-dependent. tion). The language of Bernoulli's essay is pre-
The role of prior stimulation is familiar scriptive-it
in the speaks of what is sensible or
domain of temperature. Immersing the hand reasonable
in to do-but the theory was also in-
water at 20?C will feel pleasantly warm tended after as a description of the choices of reason-
prolonged immersion in much colder water, able men
and (Gerd Gigerenzer et al., 1989). As in
pleasantly cool after immersion in much most modem treatments of decision-making,
warmer water. Figure 5 illustrates reference- Bernoulli's essay does not acknowledge any
dependence in vision. The two enclosed squares tension between prescription and description.
have the same luminance, but they do not ap- The proposition that decision makers evaluate
pear equally bright. The point of the demonstra- outcomes by the utility of final asset positions
tion is that the brightness of an area is not a has been retained in economic analyses for al-
single-parameter function of the light energy most 300 years. This is rather remarkable, be-
that reaches the eye from that area, just as the cause the idea is easily shown to be wrong; I
experience of temperature is not a single-param- call it Bernoulli's error.
eter function of the temperature to which one is Tversky and I constructed numerous thought
currently exposed. An account of perceived experiments when we began the study of risky
brightness or temperature also requires a param- choice that led to the formulation of prospect
eter for a reference value (often called adapta- theory (Kahneman and Tversky, 1979). Exam-
tion level), which is influenced by the context of ples such as Problems 1 and 2 below convinced
current and prior stimulation. us of the inadequacy of the utility function for
From the vantage point of a student of per- wealth as an explanation of choice.
ception, it is quite surprising that in standard
economic analyses the utility of decision out-
comes is assumed to be determined entirely by Problem 1
the final state of endowment, and is therefore
Would you accept this gamble?
reference-independent. In the context of risky
choice, this assumption can be traced to the 50% chance to win $150
brilliant essay that first defined a theory of ex- 50% chance to lose $100
pected utility (Daniel Bernoulli, 1738). Ber-
noulli assumed that states of wealth have a Would your choice change if your
specified utility, and proposed that the decision overall wealth were lower by $100?
rule for choice under risk is to maximize the
Problem 2
Which would you choose?
lose $100 with certainty
or
poor, but these preferences are incoherent.blocks, and they do not spontaneously trans-
Schelling's problem highlights an important form the representation of puzzles or decision
problems. Obviously, no one is able to recog-
point. Framing effects are not a laboratory cu-
riosity, but a ubiquitous reality. The tax tablenize "137 x 24" and "3,288" as "the same"
must be framed one way or another, and eachnumber without going through some elaborate
frame will increase the accessibility of some computations. Invariance cannot be achieved by
responses and make other responses less likely. a finite mind.
There has been considerable interest among The impossibility of invariance raises signif-
behavioral economists in a particular type of
icant doubts about the descriptive realism of
framing effect, where a choice between two rational-choice models (Tversky and Kahne-
options A and B is affected by designatingman, 1986). Absent a system that reliably gen-
either A or B as a default option. The option
erates appropriate canonical representations,
designated as the default has a large advantage intuitive decisions will be shaped by the factors
in such choices, even for decisions that have that determine the accessibility of different fea-
considerable significance. Eric J. Johnson et al. tures of the situation. Highly accessible features
(1993) described a compelling example. The will influence decisions, while features of low
states of Pennsylvania and New Jersey both accessibility will be largely ignored-and the
offer drivers a choice between an insurance correlation between accessibility and reflective
policy that allows an unconstrained right to sue,judgments of relevance in a state of complete
and a less expensive policy that restricts the information is not necessarily high.
right to sue. The unconstrained right to sue is A particularly unrealistic assumption of the
the default in Pennsylvania, the opposite is the rational-agent model is that agents make their
default in New Jersey, and the takeup of fullchoices in a comprehensively inclusive context,
coverage is 79 percent and 30 percent in the two which incorporates all the relevant details of the
states, respectively. Johnson and Daniel G. present situation, as well as expectations about
Goldstein (2003) estimate that Pennsylvania all future opportunities and risks. Much evi-
drivers spend 450 million dollars annually on dence supports the contrasting claim that peo-
full coverage that they would not purchase if ple's views of decisions and outcomes are
their choice were framed as it is for New Jersey normally characterized by "narrow framing"
drivers. (Kahneman and Daniel Lovallo, 1993), and by
Johnson and Goldstein (2003) also compared the related notions of "mental accounting"
the proportions of the population enrolled in (Thaler, 1985, 1999) and "decision bracketing"
organ donation programs in seven European (Daniel Read et al., 1999).
countries in which enrollment was the default The following are some examples of the
and four in which nonenrollment was the de- prevalence of narrow framing. The decision of
fault. Averaging over countries, enrollment in whether or not to accept a gamble is normally
donor programs was 97.4 percent when this considered as a response to a single opportunity,
was the default option, 18 percent otherwise. not as an occasion to apply a general policy
The passive acceptance of the formulation (Gideon Keren and Willem A. Wagenaar, 1987;
given has significant consequences in this Tversky and Donald A. Redelmeier, 1992; Kah-
case, as it does in other recent studies where neman and Lovallo, 1993; Shlomo Benartzi and
the selection of the default on the form that Thaler, 1999). Investors' decisions about partic-
workers completed to set their 401(k) contri- ular investments appear to be considered in
butions dominated their ultimate choice isolation from the remainder of the investor's
(Brigitte Madrian and Dennis Shea, 2001; portfolio (Nicholas Barberis et al., 2003). The
James J. Choi et al., 2002). time horizon that investors adopt for evaluating
The basic principle of framing is the passive their investments appears to be unreasonably
acceptance of the formulation given. Because of short-an observation that helps explain the
this passivity, people fail to construct a canon- equity-premium puzzle (Benartzi and Thaler,
ical representation for all extensionally equiva- 1995). Finally, the prevalence of the gain/loss
lent descriptions of a state of affairs. They do framing of outcomes over the wealth frame,
not spontaneously compute the height of a which was discussed in the previous sec-
tower that could be built from an array oftion, can now be seen as an instance of narrow
(a) (b)
Tom W. Linda
9 7
8
RI
0 7 0
0
0
m
- 5
6 -
.0
id
._9 5 _ 4
ec 4
3
c
(a 3
0 (0
1 1
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7
in economics (see, e.g., Peter A. Diamond, ple's decisions often express affective evalua-
1977; David M. Grether, 1978; Howard Kun- tions (attitudes), which do not conform to the
reuther, 1979; Arrow, 1982). There has also logic of economic preferences. To understand
been some discussion of the role of specific preferences, then, we may need to understand
judgment biases in economic phenomena, espe- the psychology of emotions. And we cannot
cially in finance (e.g., Werner F. M. De Bondt take it for granted that preferences that are con-
and Thaler, 1985; Robert J. Shiller, 2000; An- trolled by the emotion of the moment will be
drei Shleifer, 2000; Matthew Rabin, 2002). Re- internally coherent, or even reasonable by the
cent extensions of the notion of heuristics to the cooler criteria of reflective reasoning. In other
domain of affect may be of particular relevance words, the preferences of System 1 are not
to the conversation between psychology and necessarily consistent with the preferences of
economics, because they bear on the core con- System 2. The next section will show that some
cept of a preference. As was noted earlier, af- choices are not appropriately sensitive to vari-
fective valence is a natural assessment, which is ations of quantity and cost-and are better de-
automatically computed and always accessible. scribed as expressions of an affective response
This basic evaluative attribute (good/bad, like/ than as economic preferences.
dislike, approach/avoid) is therefore a candidate
for substitution in any task that calls for a fa- VI. Prototype Heuristics
vorable or unfavorable response. Slovic and his
colleagues (see, e.g., Slovic et al., 2002) intro- The results summarized in Figure 8 showed
duced the concept of an affect heuristic. They that the judgments that subjects made about the
showed that affect (liking or disliking) is the Tom W. and Linda problems substituted the
heuristic attribute for numerous target at- more accessible attribute of similarity (repre-
tributes, including the evaluation of the costs sentativeness) for the required target attribute of
and benefits of various technologies, the safe probability. The goal of the present section is to
concentration of chemicals, and even the pre- embed the representativeness heuristic in a
dicted economic performance of various indus- broader class of prototype heuristics, which
tries. In an article aptly titled "Risk as share a common psychological mechanism-
Feelings," Loewenstein et al. (2001) docu- the representation of categories by their proto-
mented the related proposition that beliefs about types-and a remarkably consistent pattern of
risk are often expressions of emotion. biases.
If different target attributes are strongly in- In the display of lines in Figure 3, the average
fluenced by the same affective reaction, the (typical) length of the lines was highly accessi-
dimensionality of decisions and judgmentsble, but the sum of their lengths was not. Both
about valued objects may be expected to beobservations are quite general. Classic psycho-
unreasonably low. Indeed, Melissa L. Finucane logical experiments have established the fol-
et al. (2000) found that people's judgments lowing of proposition: whenever we look at or
the costs and benefits of various technologiesthink about a set (ensemble, category) which is
are negatively correlated, especially when thesufficiently homogeneous to have a prototype,
judgments are made under time pressure. A information about the prototype is automati-
technology that is liked is judged to have lowcally accessible (Michael I. Posner and Stephen
costs and large benefits. These judgments are W. Keele, 1968; Eleanor Rosch and Carolyn B.
surely biased, because the correlation betweenMervis, 1975). The prototype of a set is char-
costs and benefits is generally positive in theacterized by the average values of the salient
world of real choices. In the same vein, Kahne- properties of its members. The high accessibil-
man et al. (1997) presented evidence that dif-ity of prototype information serves an important
ferent responses to public goods (e.g., adaptive function. It allows new stimuli to be
willingness to pay, ratings of moral satisfaction categorized efficiently, by comparing their fea-
for contributing) yielded essentially inter- tures to those of category prototypes.3 For
changeable rankings of a set of policy issues.
Here again, a basic affective response appeared
to be the common factor.
3 Stored information about individual exemplars also
Kahneman et al. (1997) suggested that peo- contributes to categorization.
Patient A Patient B
8 8-
A
7 7-
6 6
4D 5 t5. I
a-
4 - c4
3 - l 3
.
IL
2 - 2-
1 - 1
0 0
0 10 20 0 10 20
ended. Instead, the physician waited for about aneman, 1994). When a choice is to be made,
minute, leaving the instrument stationary. The the option that is associated with the higher
experience during the extra period was uncom- remembered utility (more liked) is chosen.
fortable, but the procedure guaranteed that theThis mode of choice is likely to yield choices
colonoscopy never ended in severe pain. Pa- that do not maximize the utility that will
tients reported significantly more favorable actually be experienced (Kahneman et al.,
global evaluations in this experimental condi- 1997).
tion than in the control condition (Redelmeier et
al., 2003). C. Other Prototype Heuristics
Violations of dominance have also been
confirmed in choices. Kahneman et al. (1993) The pattern of results observed in diverse
exposed participants to two cold-pressor ex- studies of prototype heuristics suggests the need
periences, one with each hand: a "short" ep- for a unified interpretation, and raises a signif-
isode (immersion of one hand in 14?C water icant challenge to treatments that deal only with
for 60 seconds), and a "long" episode (the one domain. A number of authors have offered
short episode, plus an additional 30 seconds competing interpretations of base-rate neglect
during which the water was gradually warmed (Leda Cosmides and John Tooby, 1996;
to 15?C). When they were later asked which Jonathan Jay Koehler, 1996), insensitivity to
of the two experiences they preferred to re- scope in WTP (Raymond Kopp, 1992), and
peat, a substantial majority chose the long duration neglect (Ariely and Loewenstein,
trial. This pattern of choices is predicted from 2000). In general however, these interpretations
the Peak/End rule of evaluation that was de- are specific to a particular task, and would not
scribed earlier. Similar violations of domi- carry over to demonstrations of extension ne-
nance were observed with unpleasant sounds glect in the other tasks that have been dis-
of variable loudness and duration (Charles A. cussed. In contrast, the account offered here
Schreiber and Kahneman, 2000). These vio- (and developed in greater detail by Kahneman
lations of dominance suggest that choices be-and Frederick, 2002) is equally applicable to
tween familiar experiences are made in an diverse tasks that require an assessment of an
intuitive process of "choosing by liking." Ex- extensional target attribute.
tended episodes are represented in memory by The cases that have been discussed are only
a typical moment-and the desirability or illustrations, not a comprehensive list of proto-
aversiveness of the episode is dominated by type heuristics. For example, the same form of
the remembered utility of that moment (Kah- nonextensional thinking explains why the me-
dian estimate of the annual number of murders mind? The answer, as usual in psychology, is a
in Detroit is twice as high as the estimate of the list of relevant factors.
number of murders in Michigan (Kahneman Research has established that the ability to
and Frederick, 2002). It also explains why avoid errors of intuitive judgment is impaired
professional forecasters assigned a higher by time pressure (Finucane et al., 2000), by
probability to "an earthquake in California concurrent involvement in a different cognitive
causing a flood in which more than 1,000 task (Gilbert, 1989, 1991, 2002), by performing
people will drown" than to "a flood some- the task in the evening for "morning people"
where in the United States in which more than and in the morning for "evening people" (Galen
1,000 people will drown" (Tversky and Kah- V. Bodenhausen, 1990), and, surprisingly, by
neman, 1983). being in a good mood (Alice M. Isen et al.,
As these examples illustrate, there is no guar- 1988; Herbert Bless et al., 1996). Conversely,
anteed defense against violations of monotonic- the facility of System 2 is positively correlated
ity. How could a forecaster who assigns a with intelligence (Stanovich and West, 2002),
probability to a lethal flood ensure (in finite with the trait that psychologists have labeled
time) that there is no subset of that event which "need for cognition" (which is roughly whether
would have appeared even more probable? people find thinking fun) (Eldar Shafir and
More generally, the results reviewed in this Robyn A. LeBoeuf, 2002), and with exposure to
section suggest a profound incompatibility be- statistical thinking (Richard E. Nisbett et al.,
tween the capabilities and operational rules of 1983; Franca Agnoli and David H. Krantz,
intuitive judgment and choice and the norma- 1989; Agnoli, 1991).
tive standards for beliefs and preferences. The The question of the precise conditions under
logic of belief and choice requires accurate which errors of intuition are most likely to be
evaluation of extensional variables. In contrast, prevented is of methodological interest to psy-
intuitive thinking operates with exemplars or chologists, because some controversies in the
prototypes that have the dimensionality of indi- literature on cognitive illusions are resolved
vidual instances and lack the dimension of when this factor is considered (see Kahneman
extension. and Frederick, 2002; Kahneman, 2003b). One
of these methodological issues is also of con-
VII. The Boundaries of Intuitive Thinking siderable substantive interest: this is the distinc-
tion between separate evaluation and joint
The judgments that people express, the ac- evaluation (Hsee, 1996).
tions they take, and the mistakes they commit In the separate evaluation condition of List's
depend on the monitoring and corrective func- study of dominance violations, for example,
tions of System 2, as well as on the impressions different groups of traders bid on two sets of
and tendencies generated by System 1. This baseball cards; in joint evaluation each trader
section reviews a selection of findings and ideas evaluated both sets at the same time. The results
about the functioning of System 2. A more were drastically different. Violations of mono-
detailed treatment is given in Kahneman and tonicity, which were very pronounced in the
Frederick (2002) and Kahneman (2003b). between-groups comparison, were eliminated in
Judgments and choices are normally intui- the joint evaluation condition. The participants
tive, skilled, unproblematic, and reasonably in the latter condition evidently realized that one
successful (Klein, 1998). The prevalence of of the sets of goods included the other, and was
framing effects, and other indications of super- therefore worth more. Once they had detected
ficial processing such as the bat-and-ball prob- the dominance relation, the participants con-
lem, suggest that people mostly do not think strained their bids to follow the rule. These
very hard and that System 2 monitors judg- decisions are mediated by System 2. Thus, there
ments quite lightly. On some occasions, how- appear to be two distinct modes of choice:
ever, the monitoring of System 2 will detect a "choosing by liking" selects the most attractive
potential error, and an effort will be made to option; "choosing by rule" conforms to an ex-
correct it. The question for this section can be plicit constraint.
formulated in terms of accessibility: when do Prospect theory introduced the same distinc-
doubts about one's intuitive judgments come to tion between modes of choice (Kahneman and
Tversky, 1979). The normal process corre- depends on the factors of attention and accessi-
sponds to choice by liking: the decision maker bility. The fact that System 2 "knows" the dom-
evaluates each gamble in the choice set, then inance rule and "wants" to obey it only
selects the gamble of highest value. In prospect guarantees that the rule will be followed if a
theory, this mode of choice can lead to the potential violation is explicitly detected.
selection of a dominated option.5 However, the System 2 has the capability of correcting
theory also introduced the possibility of choice other errors, besides violations of dominance. In
by rule: if one option transparently dominates particular, the substitution of one attribute for
the other, the decision maker will select the another in judgment inevitably leads to errors
dominant option without further evaluation. To in the weights assigned to different sources
test this model, Tversky and Kahneman (1986) of information, and these could-at least in
constructed a pair of gambles that satisfied three principle-be detected and corrected. For ex-
criteria: (i) gamble A dominated gamble B; (ii) ample, a participant in the Tom W. study (see
the prospect-theory value of B was higher than Figure 8a) could have reasoned as follows:
the value of A; (iii) the gambles were complex, "Tom W. looks very much like a library science
and the dominance relation only became appar- student, but there are very few of those. I should
ent after grouping outcomes. As expected from therefore adjust my impression of probability
other framing results, most participants in the downward." Although this level of reasoning
experiment evaluated the gambles as originally should not have been beyond the reach of the
formulated, failed to detect the relation between graduate students who answered the Tom W.
them, chose the option they liked most, and question, the evidence shown in Figure 8 shows
exhibited the predicted violation of dominance. that few, if any, of these respondents had the
The cold-pressor experiment that was de- idea of adjusting their predictions to allow for
scribed earlier (Kahneman et al., 1993) is the different base rates of the alternative out-
closely analogous to the study of nontransparent comes. The explanation of this result in terms of
dominance that Tversky and Kahneman (1986)accessibility is straightforward: the experiment
reported. A substantial majority of participants provided no explicit cues to the relevance of
violated dominance in a direct and seeminglybase rates.
transparent choice between cold-pressor experi- Base-rate information was not completely ig-
ences. However, postexperimental debriefingsnored in experiments that provided stronger
indicated that the dominance was not in fact cues, though the effects of this variable were
transparent. The participants in the experiment consistently too small relative to the effect of
did not realize that the long episode included the the case-specific information (Jonathan St. B. T.
short one, although they did notice that the Evans et al., 2002). The evidence of numerous
episodes differed in duration. Because they studies supports the following conclusions: (i)
failed to detect that one option dominated the the likelihood that the subject will detect a mis-
other, the majority of participants chose as peo- weighting of some aspect of the information
ple commonly do when they select an experi- depends on the salience of cues to the relevance
ence to be repeated: they "chose by liking," of that factor; (ii) if the misweighting is de-
selected the option that had the higher remem- tected, there will be an effort to correct it; (iii)
bered utility, and thereby agreed to expose the correction is likely to be insufficient, and the
themselves to a period of unnecessary pain final judgments are therefore likely to remain
(Kahneman, 1994; Kahneman et al., 1997). anchored on the initial intuitive impression
The complex pattern of results in the studies (Gretchen B. Chapman and Johnson, 2002).
of dominance in the joint-evaluation design Economists may be struck by the emphasis
suggests three general conclusions: (i) choices on salient cues and by the absence of financial
that are governed by rational rules do exist, but incentives from the list of major factors that
(ii) these choices are restricted to unusual cir- influence the quality of decisions and judg-
cumstances, and (iii) the activation of the rules ments. However, the claim that high stakes
eliminate departures from rationality is not sup-
ported by a careful review of the experimental
5 Cumulative prospect theory (Tversky and Kahneman,
evidence (Camerer and Robin M. Hogarth,
1992) does not have this feature. 1999). A growing literature of field research and
field experiments documents large and system- may be more difficult to translate into the the-
atic mistakes in some of the most consequential oretical language of economics. The core ideas
financial decisions that people make, including of the present treatment are the two-system
choices of investments (Brad M. Barber and structure, the large role of System 1 and the
Terrance Odean, 2000; Benartzi and Thaler, extreme context-dependence that is implied by
2001), and actions in the real estate market the concept of accessibility. The central charac-
(David Genesove and Christopher J. Mayer, teristic of agents is not that they reason poorly
2001). The daily paper provides further evi- but that they often act intuitively. And the be-
dence of poor decisions with large outcomes. havior of these agents is not guided by what
The present analysis helps explain why the they are able to compute, but by what they
effects of incentives are neither large nor robust. happen to see at a given moment.
High stakes surely increase the amount of at- These propositions suggest heuristic ques-
tention and effort that people invest in their tions that may guide attempts to predict or ex-
decisions. But attention and effort by them- plain behavior in a given setting: "What would
selves do not purchase rationality or guarantee an impulsive agent be tempted to do?" "What
good decisions. In particular, cognitive effort course of action seems most natural in this
expended in bolstering a decision already made situation?" The answers to these questions will
will not improve its quality, and the evidence often identify the judgment or course of action
suggests that the share of time and effort de- to which most people will be attracted. For
voted to such bolstering may increase when the example, it is more natural to join a group of
stakes are high (Jennifer S. Lemer and Philip E. strangers running in a particular direction than
Tetlock, 1999). Effort and concentration are to adopt a contrarian destination. However, the
likely to bring to mind a more complete set of two-system view also suggests that other ques-
considerations, but the expansion may yield an tions should be raised: "Is the intuitively attrac-
inferior decision unless the weighting of the tive judgment or course of action in conflict
secondary considerations is appropriately low. with a rule that the agent would endorse?" If the
In some instances-including tasks that require answer to that question is positive, then "How
predictions of one's future tastes-too much likely is it in the situation at hand that the
cognitive effort actually lowers the quality of relevant rule will come to mind in time to over-
performance (Wilson and Jonathan W. ride intuition?" Of course, this mode of analysis
Schooler, 1991). Klein (2003, Ch. 4) has argued also allows for differences between individuals,
that there are other situations in which skilled and between groups. What is natural and intui-
decision makers do better when they trust their tive in a given situation is not the same for
intuitions than when they engage in detailed everyone: different cultural experiences favor
analysis. different intuitions about the meaning of situa-
tions, and new behaviors become intuitive as
VIII. Concluding Remarks skills are acquired. Even when these complex-
ities are taken into account, the approach to the
The rational agent of economic theory would understanding and prediction of behavior that
be described, in the language of the present has been sketched here is simple and easy to
treatment, as endowed with a single cognitive apply, and likely to yield hypotheses that are
system that has the logical ability of a flawless generally plausible and often surprising. The
System 2 and the low computing costs of Sys- origins of this approach are in an important
tem 1. Theories in behavioral economics have intellectual tradition in psychology, which has
generally retained the basic architecture of the emphasized "the power of the situation" (Lee
rational model, adding assumptions about cog- Ross and Nisbett, 1991).
nitive limitations designed to account for spe- The present treatment has developed several
cific anomalies. For example, the agent may be themes: that intuition and reasoning are alterna-
rational except for discounting hyperbolically, tive ways to solve problems, that intuition re-
evaluating outcomes as changes, or a tendency sembles perception, that people sometimes
to jump to conclusions. answer a difficult question by answering an
The model of the agent that has been pre- easier one instead, that the processing of infor-
sented here has a different architecture, which mation is often superficial, that categories are
Behavior and Organization, March 1980, . "Rational Choice and the Framing of
1(1), pp. 36-90. Decisions." Journal of Business, October
. "Mental Accounting and Consumer 1986, 59(4), pp. S251-78.
Choice." Marketing Science, Summer 1985, . "Loss Aversion in Riskless Choice: A
4(3), pp. 199-214. Reference-Dependent Model." Quarterly
. Quasi rational economics. New York: Journal of Economics, November 1991,
Russell Sage Foundation, 1991. 106(4), pp. 1039-61.
. The winner's curse: Paradoxes and . "Advances in Prospect Theory: Cumu-
anomalies of economic life. New York: Freelative Representation of Uncertainty." Jour-
Press, 1992. nal of Risk and Uncertainty, October 1992,
. "Mental Accounting Matters." Journal 5(4), pp. 297-323.
of Behavioral Decision Making, July 1999, Tversky, Amos and Redelmeier, Donald A. "On
12(3), pp. 183-206. the Framing of Multiple Prospects." Psycho-
. "Toward a Positive Theory of Con-
logical Science, May 1992, 3(3), pp. 191-93.
sumer Choice," in Daniel Kahneman and Wilson, Timothy D. Strangers to ourselves:
Amos Tversky, eds., Choices, values, and Discovering the adaptive unconscious.
frames. New York: Cambridge University Cambridge, MA: Harvard University Press,
Press, 2000, pp. 268-87. 2002.
Tversky, Amos and Kahneman, Daniel. "Judg- Wilson, Timothy D. and Schooler, Jonathan W.
ment under Uncertainty: Heuristics and "Thinking Too Much: Introspection Can Re-
Biases." Science, September 1974, duce the Quality of Preferences and Deci-
185(4157), pp. 1124-31. sions." Journal of Personality and Social
. "The Framing of Decisions and the Psychology, February 1991, 60(2), pp. 181-
Psychology of Choice." Science, January 92.
1981, 211(4481), pp. 453-58. Zajonc, Robert B. "Emotions," in Daniel T. Gil-
. "Extensional Versus Intuitive Reason- bert, Susan T. Fiske, and Gardner Lindzey,
ing: The Conjunction Fallacy in Probability eds., Handbook of social psychology, 4th Ed.,
Judgment." Psychological Review, October Vol. 1. New York: Oxford University Press,
1983, 90(4), pp. 293-315. 1998, pp. 591-632.