### abstract ###
people tend to overestimate emotional responses to future events
this study examined whether such affective forecasting errors occur for feelings of regret  as measured by self-report and subsequent decision-making
some participants played a pricing game and lost by a narrow or wide margin  while others were asked to imagine losing by such margins
participants who experienced a narrow loss reported more regret than those who imagined a narrow loss
furthermore  those experiencing a narrow loss behaved more cautiously in a subsequent gambling task
thus  the study provides self-report and behavioral evidence for a reversal of the affective forecasting phenomenon for feelings of regret
### introduction ###
people want to be happy  when confronted with a decision  they'd rather choose the path to joy than the path to misery
if only it were that simple
one problem is that the paths are rarely labeled  instead people have to imagine the feelings that the outcome of their decisions would trigger
and  although people are relatively skilled at predicting how a certain outcome will make them feel  they are less skilled at predicting the intensity and duration of those emotions
you can guess that getting tenure will make you happy  but the elation will dissipate faster than you would imagine
this asymmetry between anticipating and experiencing emotions has been studied in the laboratory  as well as in natural settings  under the banner of  affective forecasting   CITATION
in a typical laboratory study of affective forecasting  participants first report their general happiness and next engage in some experimental procedure with uncertain outcome
finally  some participants are asked to predict how happy they would feel after receiving a particular outcome  while others are told the outcome first and then are asked to report the happiness they experience
this allows researchers to measure the extent to which anticipated and experienced feelings depart from the baseline level of happiness reported at the beginning of the session
the typical result is that participants overestimate the duration of their forecasted emotions  which leads them to overestimate how intense their feelings will be some time after the outcome
for example  when asked to imagine how they would feel  NUMBER  minutes after losing money in a gambling task  people predict more unhappiness than the unhappiness they experience  NUMBER  minutes after actually losing  CITATION
people also overestimate the intensity of their initial reaction  CITATION
considerable evidence has accumulated over the last  NUMBER  years in support of the basic finding that people overestimate the impact that future events will have on their emotional response  CITATION
most of these studies have equated emotional response with changes in valence
for example  some studies have asked participants to report their happiness on a scale ranging from not happy to very happy
other studies have asked participants to report the intensity of several different emotions and those reports were later converted into composites scores of positive and negative valence  CITATION
this emphasis on the pleasant unpleasant dimension has proved useful for the study of affective forecasting
however this approach also begs the question of whether the forecast bias extends beyond the pleasant unpleasant dimension to include biases when predicting the intensity of specific emotions
we know that people are relatively skilled at predicting which specific emotions e g   anger  fear  disgust  etc
they will experience in different situations  CITATION
we know less about how accurate people are at predicting the intensity and duration of those specific emotions
a few studies have started to examine this question particularly for the feeling of regret
the current study is another step in that direction
regret can be defined as  a negative  cognitively based emotion that we experience when realizing or imagining that our present situation would have been better  had we decided differently   CITATION
thus  regret is unique among emotions in its reliance on counterfactual thinking   if i had chosen differently  then things would have turned out better
  when counterfactuals are readily available  regret seems to grow more intense
for example  people's intuition is that poor choices are most regretful when they are unusual choices  in those cases  the conventional approach is readily available as a counterfactual
for a similar reason  people believe that they will experience more regret when the desired outcome is narrowly missed than when it is missed by a wide margin
presumably  a loss by a narrow margin highlights the unfulfilled success  CITATION
as a consequence  people believe that missing a flight by  NUMBER  minutes should cause more regret than missing it by an hour
psychologists have shared that same intuition for over  NUMBER  years  CITATION
the intuition that narrow margins of loss are highly effective at triggering regret has been challenged by a recent study on affective forecasting  CITATION
in a naturalistic setting  people who missed the subway by a narrow margin reported no more regret than those who missed it by a wide margin
this result was in stark contrast to the predictions people made when interviewed before the train arrived
the finding was replicated in an experimental setting
for this  a new set of participants was recruited to play a game inspired by the tv show the price is right
before starting the game  participants were asked to report their emotional state at the time  including a report on their level of regret and of disappointment
next  people were asked to rank several household products by price
participants had to create two arrangements that exemplified their two best guesses  and then proceeded to choose the best one for an opportunity to win a prize
some participants were asked to predict how much regret they would feel if they were to find out that the correct arrangement was the one they did not choose a narrow margin of loss
this condition aimed to experimentally recreate the situation in which a participant is asked to imagine having narrowly missed a desired outcome e g   catching the subway
participants' predictions in this condition were compared to the actual feelings of regret  as reported by participants who actually experienced a narrow margin of loss
other participants were asked to predict how much regret they would feel if they found out that neither of the two arrangements was correct a wide margin of loss
as hypothesized by the authors of the study  participants predicted they would feel more regret after losing by a narrow margin  but in reality they did not
in other words  there was an asymmetry between the amount of regret anticipated and the amount of regret experienced
these results were consistent with a theory of affective forecasting that has received substantial empirical support over the years  CITATION
at the same time  some aspects of the experimental design cast doubt on the generalizability of the findings
in particular  participants in gilbert et al 's  CITATION  study were offered a consolation prize after being told they had lost
this small prize may have helped participants to re-conceptualize the loss as a minor win  thus reducing the amount of regret experienced
in fact  previous research shows that people recover most quickly from negative events when it is easier for them to rationalize the outcome to their favor  CITATION
thus  the absence of a consolation prize might lead to a very different pattern of results  one in which the experience of regret is stronger than anticipated
our study tested this hypothesis
another limitation of the  NUMBER  study by gilbert et al is its reliance on a self-report measure
while this approach has been validated for basic emotions such as happiness  its use for assessing more complex emotions such as regret is a matter of debate
asking people directly to report how much regret they are feeling assumes that people can easily associate their feelings with the verbal label
although some researchers claim this is indeed possible  CITATION   others argue that such direct questioning fails to discriminate regret from other negative emotions such as disappointment  CITATION
furthermore  self-report measures are subject to possible carry-over effects when the same emotion is probed twice  as is often the case in studies of affective forecasting
one way to overcome the limitations of self-report measures is to explore the influence that regret has on future behavior
some researchers have argued that emotions in general  and regret in particular  exist for the sake of guiding behaviors  CITATION
many of the choices people make aim to minimize the experience of regret  and anticipation of regret serves to deter risk taking  CITATION
merely imagining the regretful consequences of our potential actions is sometimes enough to make us decide differently  CITATION
thus  by measuring risky behavior after a narrow loss it should be possible for us to assess whether experiencing or imagining the narrow loss is more effective in triggering the emotion
we explored this possibility with a variant of a gambling task first introduced by slovic  CITATION   which our participants completed after participating in the price is right task
participants viewed ten cards face down on a computer screen with the instructions that nine of the cards were good i e   selecting any of them would earn the participant a dollar per card  but one of the cards was a disaster card i e   choosing it would cause the loss of every dollar collected up to that point and end the game
following these instructions  participant reported how many cards they would like to turn
after turning the initial number of cards  participants made a second and more important decision
they had to decide whether to stop and walk away with the money or continue and pick one more card for the chance to win another dollar
thus  the opportunity of winning a relatively small prize was pitted against the risk of losing a larger amount i e   a regret prone situation
in sum  using self-report and behavioral measures of regret  we aimed to show that in the absence of a consolation prize  the experience of regret is larger than anticipated
