The affect heuristic in judgments of risks and benefits

TLDR: In this paper, the inverse relationship between perceived risk and perceived benefit was examined and it was shown that people rely on affect when judging the risk and benefit of specific hazards, such as nuclear power.

Abstract: This paper re-examines the commonly observed inverse relationship between perceived risk and perceived benefit. We propose that this relationship occurs because people rely on affect when judging the risk and benefit of specific hazards. Evidence supporting this proposal is obtained in two experimental studies. Study 1 investigated the inverse relationship between risk and benefit judgments under a time-pressure condition designed to limit the use of analytic thought and enhance the reliance on affect. As expected, the inverse relationship was strengthened when time pressure was introduced. Study 2 tested and confirmed the hypothesis that providing information designed to alter the favorability of one's overall affective evaluation of an item (say nuclear power) would systematically change the risk and benefit judgments for that item. Both studies suggest that people seem prone to using an ‘affect heuristic’ which improves judgmental efficiency by deriving both risk and benefit evaluations from a common source—affective reactions to the stimulus item. Copyright © 2000 John Wiley & Sons, Ltd.

Full text

The Aect Heuristic in Judgments of Risks
and Bene®ts
MELISSA L. FINUCANE,
1
* ALI ALHAKAMI,
2
PAUL SLOVIC
1
and STEPHEN M. JOHNSON
1
1
Decision Research, Eugene, OR, USA
2
Imam Muhammad Ibn Saud Islamic University Psychology Department,
Saudi Arabia
ABSTRACT
This paper re-examines the commonly observed inverse relationship between per-
ceived risk and perceived bene®t. We propose that this relationship occurs because
people rely on aect when judging the risk and bene®t of speci®c hazards.
Evidence supporting this proposal is obtained in two experimental studies. Study
1 investigated the inverse relationship between risk and bene®t judgments under a
time-pressure condition designed to limit the use of analytic thought and enhance
the reliance on aect. As expected, the inverse relationship was strengthened when
time pressure was introduced. Study 2 tested and con®rmed the hypothesis that
providing information designed to alter the favorability of one's overall aective
evaluation of an item (say nuclear power) would systematically change the risk
and bene®t judgments for that item. Both studies suggest that people seem prone
to using an `aect heuristic' which improves judgmental eciency by deriving
both risk and bene®t evaluations from a common source Ð aective reactions to
the stimulus item. Copyright
#
2000 John Wiley & Sons, Ltd.
KEY WORDS aect heuristic; judgment; perception of risk; perception of bene®t
Although aect has long played a key role in many behavioral theories, it has rarely been recognized as
an important component of human judgment and decision making. Perhaps be®tting its rationalistic
origins, the main focus of descriptive decision research has been cognitive, rather than aective. When
principles of utility maximization app eared to be descriptively inadequate, Simon (1956) oriented the
®eld toward problem solving and information-processing models based upon bounde d rationality and
concepts such as satis®cing (as opposed to maximizing). The work of Tversky and Kahneman (1974)
demonstrated how boundedly rational individuals employed heuristics such as availability, represent-
iveness, and anchoring and adjustment to make judgments and how they used simpli®ed strategies such
CCC 0894±3257/2000/010001±17$17.50
Copyright
#
2000 John Wiley & Sons, Ltd.
Journal of Behavioral Decision Making
J. Behav. Dec. Making, 13: 1±17 (2000)
* Correspondence to: Melissa L. Finucane, Decision Research, 1201 Oak Street, Eugene, OR 97401, USA.
E-mail: ®nucane@decisionresearch.org
Contract grant sponsor: US National Science Foundation.
Contract grant number: SBR 9422754.
Contract grant number: SBR 9709307.

as `elimination by aspects' to make choices (Tversky, 1972). Other investigators elaborated the cogni-
tive strategies underlying judgment and choice through models of constructed preferences (Slovic,
1995; Payne, Bettman, and Johnson, 1992), dominance structuring (Montgomery, 1983), and com-
parative advantages (Sha®r, Osherson, and Smith, 1989). In 1993, an entire issue of the journal
Cognition was dedicated to the topic of Reason-Based Choice, in which it was argued that `Decisions
. . . are often reached by focusing on reasons that justify the selection of one option over another'
(Sha®r, Simonson, and Tversky, 1993, p. 34). Similarly, a recent state-of-the-art review was titled
`Decision making from a cognitive perspective' (Busemeyer, Hastie, and Medin, 1995). In keeping with
its title, it contained almost no references to the in¯uence of aect on decisions.
Despite this cognitive emphasis, the importance of aect
1
is being acknowledged increasingly by
decision researchers. A limited role for aect was acknowledged by Sha®r et al. (1993) who conceded
that `People's choices may occasionally stem from aective judgments that preclude a thorough
evaluation of the options' (p. 32, emphasis added).
A strong early proponent of the importance of aect in decision making was Zajonc (1980), who
argued that aective reactions to stimuli are often the very ®rst reactions, occurring automatically and
subsequently guiding information processing and judgment. According to Zajonc, all perceptions
contain some aect. `We do not just see ``A house'': We see a handsome house, an ugly house, or a
pretentious house' (p. 154). He later adds, `We sometimes delude ourselves that we proceed in a rational
manner and weigh all the pros and cons of the various alternatives. But this is probably seldom the
actual case. Quite often ``I decided in favor of X'' is no more than ``I liked X'' . . . We buy the cars we
``like'', choose the jobs and houses we ®nd ``attractive'', and then justify these choices by various
reasons . . .' ( p. 155).
One of the most comprehensive and dramatic theoretical accounts of the role of aect in decision
making is presented in Damasio's (1994) somatic marker hypothesis. In seeking to determ ine `what in
the brain allows humans to behave rationally', Damasio argues that thought is made largely from
images, broadly construed to include perceptual and symbolic representations. A lifetime of learning
leads to these images to become `marked' by positive and negative feelings linked directly or indirectly
to somatic or bodily states. When a negative somatic marker is linked to an image of a future outcome,
it sounds an alarm. When a positive marker is associated with the outcome image, it becomes a beacon
of incentive. Damasio hypothesized that somatic markers increase the accuracy and eciency of the
decision process and the absence of such markers, observed in people with certain types of brain
damage, degrades decision performance.
Other theorists give aect a direct role in motivating behavior, asserting or implying that we
integrate positive and negative feelings according to some sort of automatic, rapid `aective algebra',
whose operations and rules remain to be discovered. Epstein's (1994) view on this is clear, though he
gives no clue as to how feelings are integrated:
The experiential system is assumed to be intimately associated with the experience of aect, . . . which
refer[s] to subtle feelings of which people are often unaware. When a person responds to an emo-
tionally signi®cant event . . . the experiential system automatically searches its memory banks for
related events, including their emotion al accompaniments . . . If the activated feelings are pleasant,
they motivate actions and thoughts anticipated to reproduce the feelings. If the feelings are un-
pleasant, they motivate actions and thoughts anticipated to avoid the feelings ( p. 716).
1
Aect may be viewed as a feeling state that people experience, such as happiness or sadness. It may also be viewed as a quality
(e.g. goodness or badness) associated with a stimulus. These two conceptions tend to be related. This paper will be concerned
with both of these aspects of aect.
Copyright
#
2000 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, Vol. 13, 1±17 (2000)
2 Journal of Behavioral Decision Making Vol. 13, Iss. No. 1

Also emphasizing the motivational role of aect, Mowrer (1960a, b) conceptualizes conditioned
emotional responses to images as prospective gains and losses that directly `guide and control
performance in a generally sensible adaptive manner' (1960a, p. 30). He criticizes theorists who post-
ulate purely cognitive variables such as expectancies as intervening between stimulus and response,
reiterating the concern of Guthrie (1952) that we must be careful not to leave the organism at the choice
point `lost in thought'. Mowrer's solution is to view expectancies more dynamically (as conditioned
emotions such as hopes and fears) which serve as motivating states leading to action.
Despite the increasing popularity of aect in research programs and recent attempts to acknowledge
the importance of the interplay between aect and cognition, little progress has been made in develop-
ing a theory about the role of aect in judgment and decision making . Drawing on ideas about aect
marking images (e.g. Damasio, 1994), which in turn motivates behavior (e.g. Epstein, 1994; Mowrer,
1960a, b), we propose that aect is an essential component in many forms of judgment and decision
making. The ideas articulated below are intended as a ®rst step toward encouraging the development of
theory about, and methods for exposing, the role of aect in judgment.
The basic tenet in this paper is that images, marked by positive and negative aective feelings, guide
judgment and decision making.
2
Speci®cally, we propose that people use an aect heuristic to make
judgments. That is, representations of objects and events in people's minds are tagged to varying
degrees with aect. People consult or refer to an `aective pool' (containing all the positive and
negative tags associated with the representations consciously or unconsciously) in the process of
making judgments. Just as imaginability, memorability, and similarity serve as cues for probability
judgments (e.g. the availability and representativeness heuristics), aect may serve as a cue for many
important judgments. Using an overall, readily available aective impression can be far easier Ð more
ecient Ð than weighing the pros and cons or retrieving from memory many relevant examples,
especially when the required judgment or decision is complex or mental resources are limited. This
characterization of a mental short-cut leads us to label the use of aect an `heuristic'.
To illustrate the role of aect in judgment, and show how we can ascertain people's use of the aect
heuristic, we focus on a speci®c problem, namely, trying to explain the often observed inverse rela-
tionship between judgments of risk and bene®t.
USING THE AFFECT HEURISTIC TO EXPLAIN THE INVERSE RELATIONSHIP
BETWEEN PERCEIVED RISK AND PERCEIVED BENEFIT
Within an analytic view of judgment and decision making, risk and bene®t are distinct concepts. The
nature of the gains attained from pursuit of a hazardous activity or technology is qualitatively dierent
from the nature of the risks. For instance, the bene®t gained from using roller blades (e.g. entert aining
pastime) is dierent from the risk (e.g. injury from a car collision). Driving to work, eating beef, and
using a cellular phone are other examples of activities with distinct bene®ts and risks. Though distinct,
risks and bene®ts generally tend to be positively correlated. Whereas activities that bring great bene®ts
may be high or low in risk, activities that are low in bene®t are unlikely to be high in risk (if they were,
they would be proscribed), suggesting the positive correlation in Exhibit 1.
Although risk and bene®t may be positively correlated in the environment, numerous studies have
shown them to be negatively related in people's minds. For example, Fischho et al. (1978), Slovic et al.
(1991), and McDaniels et al. (1997) reported that for many hazards the greater the perceived bene®t,
2
This is not meant to imply that only aect in¯uences judgment and decision making. Clearly, many other cognitive operations
have been shown to be important (Payne, Bettman, and Johnson, 1993) and need to be integrated with our emerging
understanding of the role of aect in judgments.
Copyright
#
2000 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, Vol. 13, 1±17 (2000)
M. L. Finucane et al. The Aect Heuristic 3

the lower the perceived risk, and vice versa. Smoking, alcoholic beverages, and food additives tend to
be seen as very high in risk and relatively low in bene®t, while vaccines, antibiotics, and X-rays tend to
be seen as very high in bene®t and relatively low in risk.
A study by Alhakami and Slovic (1994) suggested that risk and bene®t may be inversely related in
people's minds because an aective feeling is referred to when the risk or bene®t of speci®c hazards is
judged. Speci®cally, Alhakami and Slovic observed that the relationship between perceived risk and
perceived bene®t was linked to an individual's general aective evaluation of a hazard. If an activity
was `liked', people tended to judge its risks as low and its bene®ts as high. If the activity was `disliked',
the judgments were opposite Ð high risk and low bene®t. The model implied by this behavior is
similar to the model Zajonc proposed in 1980. Our model assumes that aect comes prior to, and
directs, judgments of risk and be ne®t. See Exhibit 2.
Despite the evidence in support of the role of aect in judgment and decision making, a co gnitive
interpretation of Alha kami and Slovic's (1994) results cannot be excluded completely. Their experi-
mental design cannot rule out the possibility that risk and bene®t judgments are correlated negatively
because individuals approach the judgment tasks analytically, producing a `net riskiness' or `net
bene®t' judgment rather than independent judgments of risk and bene®t. That is, individuals may be
making judgments (regardless of whether the rating scale focuses only on risk or on bene®t) by
deliberating on what the net dierence between risk and bene®t is for any particular item.
Exhibit 1. Hypothesized relationship between risk and bene®t in the environment. Risk and bene®t are positively
correlated across activities
Exhibit 2. A model of the aect heuristic explaining the risk/bene®t confounding observed by Alhakami and
Slovic (1994). Judgments of risk and bene®t are assumed to be derived by reference to an overall aective
evaluation of the stimulus item
Copyright
#
2000 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, Vol. 13, 1±17 (2000)
4 Journal of Behavioral Decision Making Vol. 13, Iss. No. 1

STUDY 1: RISK AND BENEFIT JUDGMENTS UNDER TIME PRESSURE
Study 1 attempted to rule out the possibility of a cognitive explanation of the inverse risk/bene®t
relationship by using a `time-pressure' methodology (see Benson and Beach, 1996; Ordo
Â
n
Ä
ez and
Benson, 1997). Inducing time pressure when individuals are making judgments has two main con-
sequences compatible with the goal of exposing the importance of aect in judgment. As outlined by
Maule and Svenson (1993), perception that time is limited may in¯uence judgment by:
(1) Inducing aective changes via a generally increased arousal level (making `hot' aective processes
more salient than analytical cognitive processes to individuals); and
(2) Reducing cognitive resources available for analytic deliberation during risk and bene®t judgments
(because the awareness of time pressure demands that resources are allocated to monitoring the
time available).
Most time-pressure research has examined how cognitive processes and outcomes change as a result
of the minimization of cognitive eort (see Beach and Mitchell, 1978; Edland and Svenson, 1993;
Kerstholt, 1994; Payne, Bettman, and Johnson, 1988; Svenson, Edland, and Slovic, 1990). However,
from our perspective the methodol ogy is relevant because of its potential for manipulating individuals'
heuristic reliance on aect when making judgments.
So far, few researchers have exploited the potential of time-pressure studies to highlight the use of
aectively-based heuristics. One exception was an examination of the in¯uence of attitudes and stereo-
types on judgments of people belonging to particular social categories by Dijker and Koomen (1996).
They found greater dierences between `ingroup' an d `outgroup' targets when judgments were made
under time pressure, suggesting that when processing is dicult people tend to use an `acceptability
heuristic' (i.e. the tendency to produce socially acceptable judgments; see Tetlock, Skitka, and
Boettger, 1989). Dijker and Koomen reported that under time pressure sub jects seemed to base their
judgments on both positive and ne gative aspects of their attitudes; they concluded that emotional
responses toward social groups are important components in judgments. Similarly, we anticipated that
reliance on aect in the domain of judgments about hazards could be exposed by indu cing time
pressure.
According to the aect heuristic, people may judge the risks and bene®ts of hazards by accessing a
pool of positive and ne gative feelings they associate with the hazards. The aect heuristic is more
ecient than analytic processing. Thus, compared with individuals under no time pressure, we expec-
ted those under time pressure to rely more heavily on aect because eciency is important. Stronger
negative correlations between risk and bene®t judgments were expected for participants in the time-
pressure than in the no-time-pressure condition.
Method
Participants
Fifty-four ®rst-year Psychology students from the University of Western Australia (mean age 19 years)
participated in the study for course credit. Females constituted 78% of the sample.
Design
Participants were randomly assigned to one of two conditions (time pressure or no time pressure) and
to one of two counterbalancing orders (risk judgments followed by bene®t judgme nts, or vice versa).
Copyright
#
2000 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, Vol. 13, 1±17 (2000)
M. L. Finucane et al. The Aect Heuristic 5

Frequently asked questions

Q1. What have the authors contributed in "The a€ect heuristic in judgments of risks and bene®ts" ?

This paper re-examines the commonly observed inverse relationship between perceived risk and perceived bene®t. The authors propose that this relationship occurs because people rely on a€ect when judging the risk and bene®t of speci®c hazards. As expected, the inverse relationship was strengthened when time pressure was introduced. Both studies suggest that people seem prone to using an ` a€ect heuristic ' which improves judgmental eciency by deriving both risk and bene®t evaluations from a common source Ð a€ective reactions to the stimulus item. 

Q2. What have the authors stated for future works in "The a€ect heuristic in judgments of risks and bene®ts" ?

Future researchers need to explore more deeply the underlying a ective mechanisms by which judgments are made, as well as the interplay between a ect and cognition in reasoning. One suggestion is that researchers examine the relationship between the a ect and availability heuristics. 

Q3. What is the effect of the manipulation on the perception of an attitude object?

Previous researchers have found that the favorability of the overall impression of an attitude object is a good predictor of how strongly positive or negative qualities are ascribed to the object (Klauer and Stern, 1992). 

Q4. How many attempted manipulations of a ect and response?

Recall that each of the 219 subjects received three vignettes, making a total of 219 3 657 attempted manipulations of a ect and response. 

Q5. How long did the computer wait for the scale to be rated?

in the time-pressure condition the clock was set to run out of time after 5.2 seconds for each item, at which point a yellow sign ¯ashed above the scale saying `You MUST click on the scale NOW', and the computer made a beeping sound until a rating was made. 

Q6. How long did it take to do the risk judgments?

As expected, participants in the time-pressure condition took signi®cantly less time to do the risk judgments than did participants in the no-time-pressure condition (M 101.24 seconds, SD 15.94 versus M 127.27 seconds, SD 29.12); t(52) 4.11, p5 0.001.