378 • CHAPTER 13 Reasoning and Decision Making
are emotions that are associated with the act of making a decision. For example, a
Deal or No Deal contestant who is trying to decide whether to accept or turn down
the bank’s offer may feel extremely anxious. This anxiety is the integral emotion asso-
ciated with making the decision, and it is probable that this emotion could affect the
decision.
Incidental immediate emotions are emotions that are unrelated to the decision.
Incidental emotions can be caused by a person’s general disposition (the person is natu-
rally happy, for example), or something that happened earlier in the day, or reacting to
the general environment such as background music being played in a game show or the
yells of the game show audience.
Each of these types of emotions can potentially have an effect on decisions, but
only expected emotion, which involves some element of rational thought, can be han-
dled within the expected utility framework. However, in the next section we will see
that expected emotions may not accurately predict the actual emotion that would result
from the outcome of a decision.PEOPLE INACCURATELY PREDICT THEIR EMOTIONS
A basic characteristic of research on decisions is the phenomenon of risk aversion—the
tendency to avoid taking risks. For example, a Deal or No Deal contestant who decides
to accept the banker’s offer rather than take a chance on winning big or losing it all may
be motivated by risk aversion.
Expected emotions are one of the determinants of risk aversion, because one of the
things that increase the chance of risk aversion is the tendency to believe that a particu-
lar loss will have a greater impact than a gain of the same size (Tversky & Kahneman,
1991). For example, if people believe it would be very disturbing to lose $100 but only
slightly pleasant to win $100, then this would cause them to decline a bet for which the
odds are 50-50, such as fl ipping a coin (win $100 for heads; lose $100 for tails). In fact,
because of this effect, some people are reluctant to take a 50-50 bet in which
winning pays $200 and losing pays $100, even though in accordance with util-
ity theory, this would be a good bet (Kermer et al., 2006).
Deborah Kermer and coworkers (2006) studied this effect by doing an
experiment that compared people’s expected emotions with their actual emo-
tions. They gave participants $5 and told them that based on a coin fl ip they
would either win an additional $5 or lose $3. Participants rated their happiness
before the experiment started and then predicted how their happiness would
change if they won the coin toss (gain $5, so they have $10) or lost it (lose $3, so
they have $2). The results of these ratings are indicated by the left pair of bars in
● Figure 13.10. Notice that the participants predicted that the negative effect of
losing $3 would be greater than the positive effect of winning $5.
After the coin toss, in which some participants won and some lost, they car-
ried out a fi ller task for 10 minutes and then rated their happiness. The bars on
the right show that the actual effect of losing was substantially less than predicted,
but the positive effect of winning was only a little less than predicted. As a result,
the positive effect of winning and negative effect of losing were about equal.
Why do people overestimate what their negative feelings will be? One
reason is that when making their prediction they don’t take into account the
various coping mechanisms they may use to deal with adversity. For example,
a person who doesn’t get a job he wanted might rationalize the failure by say-
ing “The salary wasn’t what I really wanted” or “I’ll fi nd something better.” In
Kremer’s experiment, participants predicting how they would feel if they lost
focused on losing $5; after the outcome was determined, participants who actu-
ally lost focused on the fact that they still had $2 left.
The results of Kremer’s experiment, plus others, show that the inability to
correctly predict the emotional outcome of a decision can lead to ineffi cient deci-
sion making (Peters et al., 2006; Wilson & Gilbert, 2003). We will now see how
emotions that aren’t even related to making the decision can affect the decision.Win $5Lose $3Before coin
flip (expected)–40–1123–2Change from baseline happiness–3After coin
flip (actual)●FIGURE 13.10 The results of Kermer et
al.’s (2006) experiments showing that people
overestimate the expected negative eff ect
of losing (left red bar), compared to the
actual eff ect of losing (right red bar). (Source:
Based on D. A. Kermer, E. Driver-Linn, T. D. Wilson, &
D. T. Gilbert, “Loss Aversion Is an Aff ective Forecasting
Error,” Psychological Science, 17, 649–653, 2006.)Copyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.