ArticlePDF Available

Abstract and Figures

Objectives: The mainstream position on regret in psychological literature is that its necessary conditions are agency and responsibility, that is, to choose freely but badly. Without free choice, other emotions, such as disappointment, are deemed to be elicited when the outcome is worse than expected. In two experiments, we tested the opposite hypothesis that being forced by external circumstances to choose an option inconsistent with one’s own intentions is an important source of regret and a core component of its phenomenology, regardless of the positivity/negativity of the post-decision outcome. Along with regret, four post-decision emotions – anger toward oneself, disappointment, anger toward circumstances, and satisfaction – were investigated to examine their analogies and differences to regret with regard to antecedents, appraisals, and phenomenological aspects. Methods: Through the scenario methodology, we manipulated three variables: choice (free/forced), outcome (positive/negative), and time (short/long time after decision-making). Moreover, we investigated whether responsibility, decision justifiability, and some phenomenological aspects (self-attribution, other attribution, and contentment) mediated the effect exerted by choice, singularly or in interaction with outcome and time, on the five emotions. Each study was conducted with 336 participants, aged 18–60. Results: The results of both studies were similar and supported our hypothesis. In particular, regret elicited by forced choice was always high, regardless of the valence of outcome, whereas free choice elicited regret was high only with a negative outcome. Moreover, regret was unaffected by responsibility and decision justifiability, whereas it was affected by the three phenomenological dimensions. Conclusion: Our results suggest that (1) the prevailing theory of regret is too binding, since it posits as necessary some requirements which are not; (2) the antecedents and phenomenology of regret are broader than it is generally believed; (3) decision-making produces a complex emotional constellation, where the different emotions, singularly and/or in combination, constitute the affective responses to the different aspects of decision-making.
Content may be subject to copyright.
Frontiers in Psychology | www.frontiersin.org 1 December 2021 | Volume 12 | Article 783248
ORIGINAL RESEARCH
published: 16 December 2021
doi: 10.3389/fpsyg.2021.783248
Edited by:
Petko Kusev,
University of Hudderseld,
UnitedKingdom
Reviewed by:
Rose Martin,
University of Surrey, UnitedKingdom
Franca Crippa,
University of Milano-Bicocca, Italy
*Correspondence:
Olimpia Matarazzo
olimpia.matarazzo@unicampania.it
Specialty section:
This article was submitted to
Cognition,
a section of the journal
Frontiers in Psychology
Received: 25 September 2021
Accepted: 02 November 2021
Published: 16 December 2021
Citation:
Matarazzo O, Abbamonte L,
Greco C, Pizzini B and Nigro G (2021)
Regret and Other Emotions Related
to Decision-Making: Antecedents,
Appraisals, and Phenomenological
Aspects.
Front. Psychol. 12:783248.
doi: 10.3389/fpsyg.2021.783248
Regret and Other Emotions Related
to Decision-Making: Antecedents,
Appraisals, and Phenomenological
Aspects
OlimpiaMatarazzo *, LuciaAbbamonte , ClaudiaGreco , BarbaraPizzini and
GiovannaNigro
Department of Psychology, University of Campania “Luigi Vanvitelli”, Caserta, Italy
Objectives: The mainstream position on regret in psychological literature is that its
necessary conditions are agency and responsibility, that is, to choose freely but badly.
Without free choice, other emotions, such as disappointment, are deemed to beelicited
when the outcome is worse than expected. In two experiments, wetested the opposite
hypothesis that being forced by external circumstances to choose an option inconsistent
with one’s own intentions is an important source of regret and a core component of its
phenomenology, regardless of the positivity/negativity of the post-decision outcome.
Along with regret, four post-decision emotions – anger toward oneself, disappointment,
anger toward circumstances, and satisfaction – were investigated to examine their
analogies and differences to regret with regard to antecedents, appraisals, and
phenomenological aspects.
Methods: Through the scenario methodology, wemanipulated three variables: choice
(free/forced), outcome (positive/negative), and time (short/long time after decision-making).
Moreover, we investigated whether responsibility, decision justifiability, and some
phenomenological aspects (self-attribution, other attribution, and contentment) mediated
the effect exerted by choice, singularly or in interaction with outcome and time, on the
ve emotions. Each study was conducted with 336 participants, aged 18–60.
Results: The results of both studies were similar and supported our hypothesis. In
particular, regret elicited by forced choice was always high, regardless of the valence of
outcome, whereas free choice elicited regret was high only with a negative outcome.
Moreover, regret was unaffected by responsibility and decision justiability, whereas it was
affected by the three phenomenological dimensions.
Conclusion: Our results suggest that (1) the prevailing theory of regret is too binding, since
it posits as necessary some requirements which are not; (2) the antecedents and
phenomenology of regret are broader than it is generally believed; (3) decision-making
produces a complex emotional constellation, where the different emotions, singularly and/
or in combination, constitute the affective responses to the different aspects of decision-making.
Keywords: regret, decision-making, post-decision emotions, antecedents, appraisals, phenomenology
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 2 December 2021 | Volume 12 | Article 783248
INTRODUCTION
e well-established psychological concept of regret based on
decision theories strongly links the genesis of this emotion to
the factual or counterfactual comparison between the actual
outcome resulting from the option chosen in decision-making
and the better outcome deriving from the foregone option (Bell,
1982; Loomes and Sugden, 1982; Tsiros, 1998; Roese and
Summerville, 2005; van Dijk and Zeelenberg, 2005; Zeelenberg
and Pieters, 2007; Beike et al., 2009). e corollary of this
conception is the assumption that decision-makers choose freely
and, thus, they have control over their actions and responsibility
for their choices. Self-blaming for a bad outcome deriving from
a wrong decision and the desire to undo this decision are
considered as specic characteristics of the phenomenology1 of
regret and as a discriminant criterion for distinguishing regret
from other decision-related emotions, such as disappointment
(Inman etal., 1997; Zeelenberg etal., 1998a,b; Gilbert etal., 2004).
It is worth noting that research on regret has come to a
standstill in the last years, maybe because the wide consensus
on the theory described above has led to the idea that the
knowledge of this emotion does not need to befurther developed.
However, in this paper, weaimed to point out that the prevailing
theory has not considered that an important source of regret
is represented by forced choices, that is, by choices due to
the constraint of external circumstances (e.g., physical accidents,
indigence, family pressures, and deaths) that compel people
to choose an option inconsistent with their intentions. A
corollary of this assumption is that forced choices can produce
regret even when their outcome is positive. Moreover, weaimed
to show that the phenomenology of the regret elicited by forced
choices is characterized by the awareness of having made an
unsatisfactory but unavoidable choice. Finally, we aimed to
reconsider analogies and dierences between regret and other
post-decision emotions.
In the following sections, the state of the art of the literature
on regret and other decision-related emotions is reviewed, with
particular reference to the aspects that have been investigated
in our studies.
Theoretical Background
A widely accepted denition posits that regret “is a comparison-
based emotion of self-blame, experienced when people realize
or imagine that their present situation would have been better
had they decided dierently in the past” (Zeelenberg and
Pieters, 2007, p.7). is conception originates from the regret
theory built by rational decision theorists in Economics (Bell,
1982; Loomes and Sugden, 1982). According to this theory,
decision-makers evaluate the outcome of the chosen alternative
by comparing it with the outcome of the nonchosen alternative(s):
Regret and rejoicing are the terms used to dene the result
of the unfavorable and favorable comparison, respectively. People
tend to avoid choices that could generate regret.
1
e term phenomenology designates the set of feeling, thinking, intentions,
and action tendencies that characterize the experience of emotion (see for
example, Frijda, 1987, 2009; Roseman et al., 1994).
Psychological research has expanded the economic
conception of regret by highlighting that the comparison
can occur not only at the factual but also at the counterfactual
level, by imagining the possible outcomes of the foregone
options (Kahneman and Miller, 1986; Kahneman, 1995;
Roese, 1999; van Dijk and Zeelenberg, 2005). Moreover,
such a research has also investigated structural features of
the regrettable decisions (e.g., deciding of acting vs. not
acting; time effect on action-based regret and inaction-based
regret, see below), appraisals (van Dijk and Zeelenberg,
2002), phenomenology (e.g., Zeelenberg etal., 1998a; Marcatto
and Ferrante, 2008; Summerville and Buchanan, 2014;
Buchanan et al., 2016), behavioral consequences (of both
anticipated and experienced regret; e.g., Mellers, 2000; Tsiros
and Mittal, 2000; Zeelenberg et al., 2001; Marcatto et al.,
2015; Davidai and Gilovich, 2018), and modes of regulation
of regret (e.g., Inman, 2007; Zeelenberg and Pieters, 2007;
Seta et al., 2008).
For example, many studies have been devoted to establish
whether actions are more intensely regretted than inactions
(see for a review, Feldman et al., 2020). e debate stemming
from the dierent positions has highlighted relevant phenomena,
such as the temporal pattern of regret (Gilovich and Medvec,
1994, 1995; Gilovich etal., 1998), according to which actions
are more regretted in the short term, whereas inactions are
more regretted in the long term. Another phenomenon is the
importance of the context in which decisions are made, which
makes sometimes actions and sometimes inactions more
regrettable (Seta et al., 2001; Zeelenberg et al., 2002; Beike
et al., 2009; Morrison and Roese, 2011; Seta and Seta, 2013;
Feldman, 2020).
e pivotal points of the conception of regret shared by
the above-mentioned studies are personal agency and
responsibility, which are considered as structural features and
central appraisals of regret. Consequently, the major components
of its phenomenology are held to be self-blame for a bad
decision and the desire to undo this decision. Similarly, since
regret is assumed to derive from controllable events, its behavioral
function is thought to be to modify the negative outcomes
of the decision, when possible, or the type of choice in the
future. ese features are considered as a discriminant criterion
for distinguishing regret from other negative emotions linked
to decision-making, such as disappointment.
Indeed, the disappointment theory formulated by decision
theorists (Bell, 1985; Loomes and Sugden, 1986) postulates
that disappointment originates from an outcome that is worse
than expected and that would have been better in a dierent
state of the world. Consequently, the point of reference of
regret is the nonchosen option; the point of reference of
disappointment is the expected but unrealized outcome within
the same chosen option.
A consistent body of research has tried to substantiate this
distinction psychologically by showing that regret and
disappointment have dierent patterns of appraisal, counterfactual
thoughts, phenomenology, and behavioral consequences (see
for reviews, Zeelenberg et al., 1998a, 2000a; van Dijk and
Zeelenberg, 2002; George and Dane, 2016).
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 3 December 2021 | Volume 12 | Article 783248
Although psychological research acknowledges that other
emotions, besides regret and disappointment, can originate
from decision-making (e.g., anger, sadness, satisfaction, and
happiness), regret is considered unique for its relation with
choice and responsibility (e.g., Roese and Summerville, 2005;
Zeelenberg and Pieters, 2007; Buchanan et al., 2016; Davidai
and Gilovich, 2018).
Compared to the position that ties regret to free choice
and responsibility, the position that does not consider these
requirements as essential, which we adopted in our research,
is denitely not dominant. In addition to philosophers, such
as Rorty (1980), Solomon (1983), and Taylor (1985), according
to which regret can also stem from events beyond personal
control or from choices for which no alternatives are available,
in psychology, this position has been advanced by Landman.
She states that it “is appropriate to speak of regret both with
reference to ones own free and voluntary acts (or omissions)
and also with reference to acts over which one had no personal
control” (Landman, 1987, p. 151). Subsequently, she denes
regret as a cognitive and emotional negative state, which can
be generated by a wide class of antecedents ranging from
misfortunes to mistakes and from voluntary actions and omissions
to uncontrollable and accidental events (Landman, 1993, p.27).
Connolly and colleagues have supported this position through
some scenario-based experiments (Connolly etal., 1997; Ordóñez
and Connolly, 2000) where they manipulated the agent of
decision-making (self vs. computer) and the outcome (better
than, worse than, similar as the status quo). On the whole,
their ndings show that responsibility can amplify regret but
is not a necessary condition for its genesis and its experience,
since regret can also be elicited by external decision-makers,
such as a computer. Regret, as well as disappointment, increases
especially when the outcome is worse than the status quo.
Moreover, Connolly and Reb (2005) and Connolly and Butler
(2006) criticized the economic conceptions of regret and
disappointment as being inadequate to account for genuinely
felt emotions resulting from decisions. ey state that experienced
emotions of regret and disappointment are similar to a greater
extent than rational decision theorists assumed and that emotional
reactions to choice outcomes are better understood in term
of negative (regret, disappointment, and sadness) vs. positive
(rejoicing, elation, and happiness) emotion clusters.
An attempt to conciliate the divergent positions about the
role of agency and responsibility in the genesis and experience
of regret has been the decision justication theory (DJT, Connolly
and Zeelenberg, 2002). According to DJT, regret stems from
two types of antecedents and related appraisals, which are not
necessarily co-present: a comparatively bad outcome and a
bad decision-making. Consequently, the theory posits that the
feeling of self-blame, which seems to be unique to the regret
phenomenology, compared to other decision-making-related
emotions, is associated with the awareness of having decided
badly (Pieters and Zeelenberg, 2005), that is, in a rash or
unjustied way. On the contrary, regret decreases or vanishes
in virtue of the awareness that the decision process has been
careful and accurate though its outcome has been negative,
as some studies have demonstrated (Inman and Zeelenberg,
2002; Kwong et al., 2013; Towers etal., 2016; Verbruggen and
De Vos, 2020).
However, an experimental study of Matarazzo and Abbamonte
(2008) failed to nd that the justiability of the choice diminished
the intensity of regret. Instead, this study revealed that being
forced by external circumstances to choose an option dierent
from the one desired generates more intense regret than the
one produced by a free choice.
Interestingly, some studies on real life or career regrets have
questioned the relationship between the intensity of regret and
the controllability of the eliciting events. For example, Wilkinson
etal. (2015) found that while the level of regret for controllable
events varied as a function of individual self-esteem, the one
for uncontrollable events was always high. In the career domain,
Wrzesniewski et al. (2006), Berg et al. (2010), and Newton
et al. (2012) found more intense regrets among those who
had not chosen their current occupation, due to social or
family constraints, than among those who had chosen it.
In summary, although some studies have challenged the
idea that free choice and responsibility are a necessary condition
for the onset of regret, to the best of our knowledge, there
are no experimental studies that have systematically investigated
the hypothesis that forced choice is an important source of
regret and a core component of its phenomenology, regardless
of the valence of the post-decision outcome. e studies
presented here address this issue. As we will specify in the
following section, we use the oxymoron “forced choices” to
designate the choices for which no alternatives are available.
In these types of choices, the decision-maker is not an external
agent, as in the studies conducted by Connolly and colleagues
(Connolly et al., 1997; Ordóñez and Connolly, 2000), but the
individual herself. Nevertheless, decision-making is not free
because, due to external or internal impediments or constraints,
she cannot or feels she cannot choose the desired option.
Regret and Free vs. Forced Choice: The
Rationale of the Studies
e idea behind the present studies is that decision-making
does not imply as a necessary condition that individuals decide
freely. Indeed, people can be forced by external circumstances
to choose an option that is inconsistent with one’s own intentions.
In our opinion, this would elicit regret per se, independently
of the outcome and the justiability of decision-making, since
the forced choice implies giving up the initial and preferred
option. is eect should beparticularly robust in self-relevant
domains, due to the subjective importance of the preferred
but not-chosen option. e “chosen” option, as a substitute
for the former, would be inherently less attractive even if a
positive outcome were derived from it. Moreover, in accordance
with the “Zeigarnik eect” (1927–1935), the option that had
to be given up and was not carried out tends to remain in
the mind because of the psychological state of tension it
generated, which was not resolved by the behavior. As Savitsky
et al. (1997) pointed out, the persistence of unaccomplished
tasks is likely to elicit regret. Free choice, on the contrary,
should elicit regret only when decision-making results in a
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 4 December 2021 | Volume 12 | Article 783248
negative outcome. Consequently, regret stemming from forced
choice should bemore intense than the one derived from free
choice because in the latter case it is moderated by the outcome,
whereas in the former the outcome is deemed irrelevant.
More specically, the main hypothesis that was tested in
our studies is the following: A forced choice should elicit the
same high intensity of regret both in case of positive and
negative outcome, whereas a free choice should produce more
intense regret with a negative than a positive outcome.
A corollary of this hypothesis concerns the characteristics
of the phenomenology of regret: In the case of free choice,
it should bemarked by self-blame and the sense of personal
responsibility; whereas when the choice was forced, it should
becharacterized by the awareness of having made an unwanted
but inevitable choice. Consequently, responsibility should
be not considered as a necessary requisite for the onset and
the subjective experience of forced choice-related regret.
Moreover, even the justiability of the decision, which should
decrease the intensity of regret (Connolly and Zeelenberg,
2002), should not be relevant in the case of forced choice,
because here regret should focus on the option one was
forced to give up, not on the accuracy of decision process.
In our studies, we tested specically if these two appraisal
dimensions (responsibility and decision justiability) are
necessary for the onset of regret.
Our hypothesis and their corollaries, which question the
unicity of the genesis and experience of regret, raise the question
of whether and on what criteria regret can be dierentiated
from other post-decision emotions.
Many authors support the hypothesis that emotions arising
from decision-making have specic antecedents, patterns of
appraisal, and phenomenology (see “eoretical Background”).
However, other authors (Connolly and Butler, 2006) have pointed
out that it would bemore appropriate to classify these emotions,
at least when they are self-reported, into two clusters: one
formed by positive emotions (rejoicing, elation, happiness) and
the other by negative emotions (regret, disappointment, sadness,
and self-blame). Our position is somewhere in between these
two. We believe that it is generally arbitrary to establish strict
boundaries between emotions and that experiencing multiple
or mixed emotions is the norm rather than the exception
(Sabini and Silver, 2005; Heavey et al., 2017; Watson and
Stanton, 2017). However, we assume that there is, at least in
principle, a partial or fuzzy correspondence between emotional
lexicon and emotional experience (Sini et al., 2014; Scherer
and Fontaine, 2019) and between everyday life emotions, such
as those investigated in the present studies, and theoretical
emotions (Clore et al., 1987).
We adopt the perspective of appraisal theories (Frijda, 1986;
Lazarus, 1991; Roseman, 2011; Lerner et al., 2015), according
to which emotions are complex and targeted aective responses,
which are based on appraisal dimensions and include experiential
content and action tendencies. Hence, once emotions are
activated, they lead us to interpret environmental events in
line with the eliciting appraisals and to behave in order to
pursue emotion-based goals. us, investigating the
phenomenology of emotions is crucial to understand whether
and to what extent dierent emotional words designate dierent
emotional experiences, especially in situations from which a
wide range of emotions can derive and there are many available
labels to describe them.
We assume that an important decision made in a self-
relevant domain results in a broad emotional reaction, where
the composing emotions capture dierent aspects linked to
decision-making both in overlapping and in mutually independent
ways, so that each of them should have a specic conguration.
Along with regret, we investigated four emotions:
disappointment, anger toward oneself, anger toward
circumstances, and satisfaction. We focused our analysis on
antecedents, and some appraisal and phenomenological aspects.
In addition to disappointment, that is, the widely studied
emotion that is closest to regret, the other emotions were
chosen on the basis of the following criteria. Satisfaction, the
only positive emotion, has been selected because it should
increase with positive outcome. Moreover, since satisfaction
refers to the aective reaction to ones well-done actions (Mellers,
2000), it does imply ego-involvement and thus should increase
with free choice and decision justiability. Anger has been
included because it can be elicited by obstacles that prevent
the achievement of a goal, by frustration, and by wrong or
stupid actions (Izard, 1991); such aspects are involved in forced
choice, negative outcome, and bad (free) choice, respectively.
Two types of anger have been distinguished as a function of
the internal or external direction of this emotion, depending
on the nature of its antecedents, appraisals, and phenomenology.
Anger toward oneself arises from the self-attribution of a poor
result and involves self-blame for having taken the wrong
decision. Accordingly, it implies subjective responsibility and
was supposed to increase with free choice and bad outcome,
while decision justiability should decrease it. Anger toward
circumstances stems from the perception of an obstacle that
interferes with ones goals: us, it should increase with forced
choice and bad outcome and entail external attribution of ones
situation. Disappointment should augment with forced choice
and bad outcome for which one does not feel responsible. e
two latter emotions should benot aected by decision justiability
because they should befocused on outcome rather than on choice.
Based on this analysis, we predicted some overlap between
disappointment, regret, and anger toward circumstances, whereas
the unique features attributed to regret should rather characterize
anger toward oneself. In our view, the specicity of the nature
of regret is that it can be elicited by opposite antecedents
which, in turn, should generate contrasting subjective experiences.
Since we set out to systematically test our hypothesis about
the importance of forced choice on the most relevant phenomena
highlighted by research on regret, we also included in our
investigation two well-known phenomena: the eect of the
temporal dimension and of the type of behavior (action or
inaction) resulting from decision-making on the intensity of regret.
We examined the rst issue by considering the time elapsed
from decision-making (a few days vs. a year) as a moderator
of the eect of the type of choice. Instead, to investigate the
second issue, weconducted two similar studies with two dierent
samples: e only dierence between them was that in the
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 5 December 2021 | Volume 12 | Article 783248
rst study, the result of decision-making was an inaction,
whereas in the second was an action.
According to the temporal theory of regret (Gilovich and
Medvec, 1994, 1995), in the study on inaction, regret should
be more intense in the long than in the short term, whereas in
the study on action, the opposite pattern should befound. However,
wele open the hypothesis of whether this pattern also concerns
the forced choice-related regret, and wedid not formulate specic
hypotheses on the other emotions taken into consideration.
In the present studies, weadopted the scenario method, a
widely used tool in the experimental studies on regret (Connolly
et al., 1997; Zeelenberg et al., 1998b,c; Ordóñez and Connolly,
2000, studies 2 and 3; Zeelenberg et al., 2000b, 2002). is
method has the advantage of allowing easy experimental
manipulation without running into ethical issues arising from
putting participants in relevant real-world situations likely to
generate distress or other intense emotional states. As well as
other methods based on perspective taking, scenarios rely on
the assumption that participants put themselves in the shoes
of the protagonist and attribute to her/him what they would
likely feel or think in a similar situation.
In general, studies based on perspective taking (regardless
of the object of investigation and the type of technique used:
scenario, photo, videotape, etc.) have used two types of
instructions: imagine-self instructions and imagine-target
instructions (Davis etal., 2004). In the rst case, the instructions
asked participants to put themselves in the place of the protagonist
and imagine what they would feel in that situation. In the
second case, the instructions asked participants to imagine
what the protagonist was thinking or feeling.
Although some studies have found that imagine-self
instructions produced more emotional involvement than imagine-
target instructions (Batson etal., 1997; Davis etal., 2004, study
1), other studies (Davis et al., 2004, study 2; Avenanti et al.,
2006; Chambers and Davis, 2012) have found no dierence
between the two types of instructions. ey rather suggest
that perspective taking is the natural way to interact with
others, unless it is inhibited by specic instructions.
In regret studies, both instructions were used, particularly
the imagine-target instructions.
We used this type of instructions in our studies by asking
participants to imagine as vividly as possible what the protagonist
of the scenario was feeling and thinking.
The Studies
To test our hypothesis and its corollaries, two studies were
conducted, each of them based on the scenario method. In both
studies, the scenario protagonist had to make an important decision
for her career. In the rst study, the decision results in an inaction:
e protagonist does not change her job position. In the second,
the decision results in an action: She decide to change job.
In all scenarios, the outcome of the foregone option was
known to the scenario protagonist: us, a requirement of rational
decision theories was fullled. Actually, the protagonist could
make a factual comparison between the consequences of both
the chosen option and the nonchosen option, but he could also
make a counterfactual comparison between his current situation
(sometime aer the decision) and what it might have been if
he had made (or could have made) a dierent choice.
For each study, we used ve moderated mediation models
(depicted in Figure 1), one for each of the ve emotions
taken into account, to test our hypothesis. In each model,
choice (forced vs. free) was included as independent variable;
outcome (negative vs. positive) and time (long term vs. short
term aer decision-making) were included as moderators; some
intervening variables, that is, decision justiability, responsibility,
and phenomenological aspects (see below for their description),
were included as mediators; each of the ve emotions was
included as dependent variable. According to the tested
hypothesis, choice, singularly and/or by interacting with the
moderators, should aect emotion through the mediation of
the intervening variables. In other words, the eect of choice
on emotion could vary in function of moderators and should
be exerted through an indirect way, by means of the eects
on the intervening variables.
ese variables were assessed (see Materials and Procedure
section) by means of one direct question for decision justiability
(i.e., to what extent the scenarios protagonist judged the reasons
underlying his own decision as valid) and for responsibility
(i.e., to what extent she felt responsible for her choice),
respectively. Instead, the phenomenological aspects were assessed
by means of 14 questions, concerning feelings and factual and
counterfactual thoughts about the type of choice made, its
context, its degree of freedom, and its consequences. ese
questions were formulated for taking into account three
dimensions: self-attribution, external attribution, and
contentment. Note that we did not take into account all the
aspects of the emotion phenomenology, but only those wejudged
relevant for investigating the subjective repercussions deriving
from the type of choice. More specically, weconsidered neither
the intentions nor the action tendencies of the protagonist.
We expected that, for both studies, the independent variable
(choice) and the moderators (outcome and time) should
dierently aect both the ve emotions and the mediators
and that, in turn, the latter should aect the ve emotions
dierently. us, choice should exert conditional (in interaction
with moderators) direct and indirect (through mediators) eects
on the ve emotions. In the previous section, we delineated
some of the principal eects we predicted and the reasons
underlying these predictions.
We conducted both studies in the career domain, which
has been considered as one of the main domains where regret
can beproduced (Roese and Summerville, 2005; Sullivan etal.,
2007; Verbruggen and De Vos, 2020). e scenarios presented
two typical situations: deciding whether to accept a work
assignment abroad for a certain period (study 1) and deciding
which career to choose (study 2). ey were conducted with
employees of a wide age range (18–60 years): In this way,
we tried to foster the identication of the participants with
the protagonists of scenarios and to increase the ecological
validity of the studies. In addition, the wide age range allowed
us to investigate possible age-related dierences in emotional
reactions to the proposed decision situations.
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 6 December 2021 | Volume 12 | Article 783248
Finally, it is worth specifying that in Italian, there are
two terms corresponding to English regret, namely, rimpianto
and rammarico, which are often used as synonyms, though
their meaning varies slightly depending on the dictionaries
(e.g., the Treccani Dictionary associates rammarico mainly
with inaction, while the Garzanti Dictionary mainly associates
rimpianto with inaction). We decided to select the most
frequently occurring term on the web, which is rimpianto.
For disappointment, the corresponding Italian term
is delusione.
STUDY 1
e study 1 investigated regret and other emotional reactions
stemming from inaction.
Method
Design and Participants
A 2 (choice: free vs. forced) × 2 (outcome: negative vs. positive) × 2
(time: short term vs. long term) between-subject design was
carried out. Gender and age were considered as covariates.
ree hundred thirty-six volunteer (unpaid) participants were
recruited from dierent companies, sports centers, and
universities in Campania. ey were equally distributed by
gender (168 males and 168 females), were aged between 18
and 60 (Mean = 33.45; S.D. = 10.82), and were randomly assigned
to one of the 8 experimental conditions (n = 42 for each
condition). All participants were employed. Among the students,
only student workers were selected. Most of the participants
(63.1%) had a high school diploma, 28.9% had a college degree,
and 8% had completed compulsory education.
To determine the sample size, an a priori power analysis
for ANCOVA on the ve emotions taken into account was
performed by using G*Power (Faul et al., 2007). To detect a
medium-small eect size (f = 0.20) and achieve a power of
(1-β) of 0.95 with an error probability of 0.05, a minimum
sample size of 327 participants was required. We decided to
increase the number of participants to 336in case we had to
exclude someone for incomplete responses. Actually, no missing
data were found. All participants executed the experimental
task individually, at the place where they were recruited. ey
gave their informed consent before starting the experiment.
Materials and Procedure
In conformity with the experimental conditions, eight scenarios
were built with the same structure and three sources of variation:
type of choice (free vs. forced), outcome (negative vs. positive),
and time (consequences of decision-making evaluated in long
vs. short term). In all scenarios, the protagonist is an executive
of a company who has to decide whether to accept the oer
to manage for 2 years a branch that the company is planning
to open abroad. e job is very well paid, and the sector in
which the branch will operate is very interesting; moreover,
the protagonist is oered the possibility of a signicant career
advancement once (s)he returns. In all scenarios, decision-
making results in inaction: e protagonist decides to stay in
the same workplace, and the assignment is given to a colleague.
In the free choice conditions, the protagonist does not have
any constraints: (s)he is aware that the proposal to manage
FIGURE1 | Model 10 of PROCESS macro.
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 7 December 2021 | Volume 12 | Article 783248
the branch abroad is a great opportunity for her career but
prefers to remain in her workplace.
In the forced choice conditions, a physical accident prevents
the protagonist from choosing the desired option (going to
manage the company branch abroad) and (s)he is therefore
forced to choose to stay in the same workplace. In the positive
outcome conditions, the protagonist gets a career advancement
and a small pay raise. In the negative outcome conditions, the
company decides to downsize some areas of production, including
the one where the protagonist works, because the costs of opening
the foreign branch have implied a restructuring of activities.
Consequently, any possibility of career advancement disappears
and the protagonists position in the company becomes much
more marginal. In both types of outcomes, the colleague who
has gone abroad is very happy with his/her work. e outcome
is evaluated a few days (short term condition) or 1 year (long
term condition) aer the choice. For each experimental condition,
in half of the scenarios the protagonist is a woman; in the
other half, it is a man. However, the scenarios were assigned
randomly to the participants, without doing any matching between
the gender of the protagonist and the one of the participant.
In all experimental conditions, participants received a two-page
booklet, in the rst of which one of the eight short stories
was reported. Aer reading the scenario, they were asked to
imagine as vividly as possible the thoughts and feelings of
the protagonist, and then to evaluate, on a nine-point scale
(1 = not at all; 9 = very much):
o how important she felt the object of the decision was for her
life (“how important does the protagonist feel the decision
whether to accept or refuse to manage a company branch
abroad was for her life?);
o whether she perceived her decision as a free choice (“To what
extent does the protagonist feel the choice of not accepting
to manage a company branch abroad was a free choice?);
o whether she felt responsible for her own choice (“To what
extent does the protagonist feel responsible for not accepting
to manage a company branch abroad?”);
o whether she judged her decision as justiable (“To what extent
does the protagonist judge the reasons for not accepting to
manage a company branch abroad as valid?”);
o whether she judged her present working condition as positive
(“To what extent does the protagonist judge her present
working condition as positive?”).
In this way, the eectiveness of the experimental manipulation
was checked and two of the putative mediation variables between
type of choice and regret, that is, responsibility and decision
justiability, were assessed. Henceforth, the ve variables will
be labelled as Intermediate Variables (InVs).
Successively, participants evaluated on a nine-point scale
(1 = not at all; 9 = very much) how intensely the protagonist
was supposed to feel ve emotions: anger toward circumstances,
anger toward oneself, disappointment, regret, and satisfaction.
Finally, they estimated on a nine-point scale (1 = not likely
at all; 9 = extremely likely) the probability with which the
protagonist was supposed to agree with the 14 items of a
questionnaire describing thoughts (including counterfactual
thoughts) and feelings related to the emotional experience.
e questionnaire (reported in Figure 2) was built in order
to capture the protagonist’s possible reactions at the time of
evaluating the outcome of his decision in relation to
three dimensions:
o self-attribution of the outcome due to free choice (e.g., e
situation Iamcurrently in is due to the decision Ifreely made;
If Ihad chosen dierently, my situation would bebetter now;
e responsibility for the situation Iamcurrently in is mine);
o external attribution of the outcome due to forced choice (e.g.,
e situation Iamcurrently in is due to the decision Iwas
forced to make; At the time of the decision Iknew that the
outcome of my choice would beworse than Iwanted; Iknow
Icould not have chosen dierently);
o aective evaluation of the current state (e.g., Iamhappy with
the consequences of my choice; Iamnot happy with the way
things turned out).
Four dierent random sequences including the ve InVs,
the ve emotions, and the 14 items of the questionnaire were
built, and the participants were pseudo-randomly assigned to
the sequences.
It is worthy to note that the we used two questions on
responsibility. e rst question asked how responsible the
protagonist felt for her own choice when it was made and
aimed to evaluate the perceived appraisal of the type of choice;
the other question was in the phenomenology questionnaire
and asked how responsible she felt for the situation in which
she was in at the moment of the evaluation of the
decision consequences.
Aer completing the questionnaire, participants gave back
the booklet. ey were asked whether they had any questions
and then were thanked for their participation.
Before conducting the experiment, a pilot study was carried
out with 20 unpaid volunteer employees in order to test the
believability of the scenarios, the clarity of the instructions
(with particular reference to manipulation check questions),
and the perceived importance of the target of the decision-
making. No changes were required aer this study.
Results
All data analyses were carried out using SPSS 18.0 IBM soware.
Manipulation Check
In Ta b le  1 , the means and standard deviations of the participants
responses to the ve InVs (importance, responsibility, perception
of the freedom of the choice, decision justiability, and
evaluation of present working condition) and the ve dependent
variables (DVs, regret, disappointment, satisfaction, anger
toward oneself, and anger toward circumstances) are shown
as a function of the independent variables (IVs, choice,
outcome, and time).
e rst analysis, a 2 x 2 x 2 MANCOVA, was performed
to check whether the manipulated experimental conditions
aected the ve InVs in the predicted directions. Gender
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 8 December 2021 | Volume 12 | Article 783248
(M = 1; F = 0) and age of participants were included as
covariates. e multivariate tests were signicant for all the
three IVs. e univariate tests revealed three main eects
due to choice, outcome, and time, and two interaction eects:
outcome x choice and time x choice. e main eects were
examined through pairwise comparisons with Bonferroni
correction; the interaction eects were examined through
simple eects analysis with Bonferroni adjustment. Choice
aected all the InVs, except importance: responsibility,
perception of the freedom of the choice, and positive evaluation
FIGURE2 | Items of the phenomenology questionnaire.
TABLE1 | Study 1 – means and standard deviations (S.D.) of intermediate variables and emotions as a function of choice, outcome, and time.
Free choice Forced choice
Positive outcome Negative outcome Positive outcome Negative outcome
Short term Long term Short term Long term Short term Long term Short term Long term
Intermediate variables
Decision justiability 6.67 (2.12) 6.26 (2.00) 6.45 (1.90) 6.26 (1.86) 6.64 (2.11) 6.86 (1.92) 6.67 (2.02) 7.21 (1.84)
Evaluation of present work
condition 5.31 (1.68) 6.57 (1.58) 2.29 (1.04) 4.33 (1.98) 3.40 (1.29) 5.26 (1.75) 1.52 (0.59) 3.31 (1.72)
Importance 6.21 (2.18) 7.26 (1.53) 7.14 (2.07) 7.71 (1.18) 6.93 (1.76) 7.00 (1.68) 7.62 (1.34) 7.69 (1.42)
Perception of the freedom of
the choice 6.71 (2.51) 7.19 (1.71) 6.98 (2.23) 7.48 (1.57) 3.33 (2.28) 3.26 (2.31) 3.45 (2.58) 3.60 (2.60)
Responsibility 6.86 (2.67) 8.14 (1.46) 7.19 (2.05) 7.98 (1.42) 2.95 (2.43) 3.55 (2.41) 3.40 (2.64) 3.45 (2.80)
Emotions
Anger toward circumstances 4.90 (2.98) 5.19 (2.06) 7.38 (2.11) 6.33 (2.43) 7.45 (1.76) 7.31 (2.09) 7.55 (2.03) 8.57 (0.80)
Anger toward oneself 3.90 (2.54) 5.12 (2.31) 6.95 (2.20) 7.31 (2.01) 4.31 (2.57) 4.07 (2.10) 4.14 (2.39) 5.05 (2.43)
Disappointment 4.69 (2.91) 4.88 (2.24) 7.36 (2.03) 7.33 (1.98) 6.48 (2.19) 6.26 (2.13) 7.05 (2.30) 7.95 (1.64)
Regret 4.79 (2.66) 5.90 (2.23) 7.43 (1.89) 7.64 (2.09) 7.17 (1.78) 6.50 (2.22) 7.00 (2.05) 7.88 (1.67)
Satisfaction 4.81 (2.43) 5.60 (1.80) 2.57 (1.85) 3.00 (2.22) 3.02 (2.33) 4.10 (2.05) 2.12 (1.81) 2.50 (1.69)
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 9 December 2021 | Volume 12 | Article 783248
of present working condition increased with free choice,
while decision justiability increased with forced choice.
Outcome aected only the positive evaluation of present
working condition, which increased with positive outcome,
and importance, which increased with negative outcome.
Time aected the positive evaluation of present working
condition, responsibility, and importance, all of which increased
in long term. e outcome × choice interaction aected the
positive evaluation of present working condition, which
increased with positive outcome and free choice compared
to positive outcome and forced choice. e time × choice
interaction concerned importance, which increased with
forced choice compared to free choice in short term, while
in long term, there was no dierence as a function of the
type of choice. No eects were due to gender or age. In
Supplementary Table 1, the results of the MANCOVA
are reported.
Effects of Choice, Outcome, and Time on
Emotion Intensity
Subsequently, ve 2 x 2 x 2 ANCOVAs were conducted on the
ve emotions to test the eects of the three IVs on their
intensity. Gender and age were included as covariates.
e results (reported in Supplementary Table 2) showed
that all emotions were aected by choice and outcome, separately
and in interaction; moreover, anger toward oneself and
satisfaction were aected also by time, while regret, anger
toward oneself, and anger toward circumstances were also
inuenced by the three-way interaction time x outcome x
choice. e main and the interaction eects were examined
with the same procedure described for the MANCOVA. In
detail, regret, disappointment, and anger toward circumstances
increased with forced choice, while anger toward oneself and
satisfaction increased with free choice; all emotions increased
with negative outcome, except satisfaction, which increased
with positive outcome. Anger toward oneself and satisfaction
increased in long term. Regret and disappointment increased
with forced choice when the outcome was positive, whereas
no dierence in intensity was found as a function of the
type of choice when the outcome was negative. However, by
examining these interactions as a function of the type of
choice, the results slightly diered for the two emotions: With
free choice, the intensity of regret increased when the outcome
was negative, while with forced choice, such an intensity was
always high irrespective of the outcome. On the contrary,
the intensity of disappointment always increased with negative
outcome, even if this eect was higher with free than forced
choice. In addition, for regret, the two-way interaction was
further qualied in light of the three-way interaction: With
free choice, its intensity was much higher when the outcome
was negative rather than positive, both in short and in long
term; instead, with forced choice, there was no intensity
dierence depending on outcome in short term, whereas in
long term, the regret intensity slightly increased when the
outcome was negative.
Anger toward oneself increased with free choice compared
to forced choice when the outcome was negative, whereas with
positive outcome, no dierence in intensity was found as a
function of the type of choice: e average scores were always
quite low. In light of the three-way interaction, this emotion
always increased with free choice compared to forced choice,
except for short term and positive outcome, where no signicant
dierence emerged as a function of choice: Its intensity was
low both in free and forced choice. Anger toward circumstances
increased with forced choice in both types of outcome, but
this eect was higher with positive outcome. When considering
the three-way interaction, this emotion increased with forced
choice in all conditions, except for short term and negative
outcome, where the intensity did not vary as a function of
the type of choice.
Satisfaction increased with free choice but only in positive
outcome condition; indeed, no intensity dierence was found
depending on choice when the outcome was negative.
All emotions except satisfaction increased in females compared
to males. No eects were due to age.
Phenomenological Dimensions and Moderated
Mediation Analyses
To reduce the number of items presented in the questionnaire
on the emotional experience, a factor analysis with the
principal component extraction method was conducted.
Before the analysis, one negative item concerning contentment
dimension (“I’m not happy with the way things turned out”)
was reverse-coded. Varimax rotation was used after controlling,
through Oblimin rotation, that there was no correlation
higher than 0.30 among components. Three components
with eigenvalue >1, explaining 65.11% of the total variance,
were extracted. The three components and the items loading
on each of them corresponded to those that had been
hypothesized in the construction of the questionnaire. They
were labelled self-attribution, external attribution, and
contentment. The values of Cronbachs alpha for each dimension
were 0.861, 0.773, and 0.796, respectively. The results are
reported in Tab l e  2 . The factor scores were saved for the
subsequent analyses.
To test the central hypothesis of the study, ve moderated
mediation analyses were performed. is analysis (Muller etal.,
2005; Hayes, 2018a,b; Hayes and Rockwood, 2020) examines
whether the putative eect of the independent variable (IV)
on the dependent variable (DV) is exerted through intermediate
variables – Mediators – and whether the mediated eect diers
as a function of the values of other variables included in the
design – Moderators – which are supposed to aect the
relationship between VI, Mediators, and DV. All moderated
mediation analyses were conducted using the PROCESS macro
for SPSS 3.1 (Hayes, 2018a), which allows 92 dierent models
of moderation, mediation and moderated mediation analyses
to be tested. e macro employs a bootstrapping method for
estimating indirect eects, that is, the eects of IV on DV
through mediating variables at dierent values of moderators:
95% bias-corrected condence intervals were calculated through
10,000 bootstrap samples.
In each analysis, wetested a model (model 10 of PROCESS
macro, see Figure 1), according to which the type of choice
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 10 December 2021 | Volume 12 | Article 783248
(IV) should aect, singularly or in interaction with outcome
and time (Moderators), the corresponding emotion (DV) through
the mediation of 5 variables: responsibility, decision justiability,
and the three dimensions extracted from the factor analysis,
that is, self-attribution, external attribution, and contentment.
For each analysis, choice (1 = forced choice; 0 = free choice),
outcome (1 = negative outcome; 0 = positive outcome), and time
(1 = long term; 0 = short term) were included in the model as
dummy variables. Gender (1 = male; 0 = female) and age were
included as covariates.
All results were examined with reference to the intercept
(i.e., free choice, positive outcome, and short term). Concerning
the relationship between the IV and/or the moderators and
the mediators (which was the same for each analysis), the
results were similar to those of the MANCOVA and were
reported in Supplementary Material (Supplementary Table3
and subsequent description).
When mediators were included in the model, the results
on the five emotions (reported in Supplementary Table 4)
were the following. The probability of regret increased with
forced choice, negative outcome, self-attribution, and external
attribution, while decreased with contentment. The conditional
direct effect of choice on the probability of regret was
moderated by outcome and time, being significant only for
positive outcome and short time, as the ANCOVA already
showed. The conditional indirect effects of choice through
self-attribution and external attribution were not moderated
by outcome and time, since they were significant for both
types of outcome and both in short and long term. More
precisely, self-attribution, which decreased with forced choice,
increased regret: So, the indirect effect was negative. Instead,
external attribution, which increased with forced choice,
increased regret: So, the indirect effect was positive. The
conditional indirect effect of choice through contentment
was moderated by outcome, being significant only with
positive outcome, in both values of time. In detail, the
probability of contentment, which decreased with forced
choice and positive outcome, decreased regret: Thus, the
indirect effect was positive.
The probability of anger toward oneself increased with
negative outcome, responsibility, and self-attribution and
decreased with decision justifiability, contentment, and with
the interaction choice x outcome. The negative interaction
replicated the results already highlighted by ANCOVA, that
is, the effect of negative outcome diminished with forced
choice. Choice exerted indirect effects through responsibility
and self-attribution and conditional indirect effects through
decision justifiability and contentment. In detail, irrespective
of the values of outcome or time, forced choice decreased
responsibility and self-attribution, which in turn increased
anger toward oneself: Thus, the indirect effects were negative
and were not moderated by outcome or time. Instead, these
variables moderated the indirect effect of forced choice
through decision justifiability, which was significant only
with negative outcome and long term, and through
contentment, which was significant only with positive outcome,
both in short and in long term: In these conditions, forced
choice decreased contentment, which in turn decreased the
probability of anger toward oneself, making the indirect
effect positive.
e probability of disappointment increased with negative
outcome and external attribution, while decreased with
contentment. Choice exerted indirect eects through external
attribution and conditional indirect eects through contentment.
Specically, independently on the values of outcome and time,
forced choice increased external attribution, which in turn
increased the probability of disappointment. Instead, the indirect
eect of choice was moderated by outcome, being signicant
only with positive outcome, in short and long term: In these
conditions, forced choice decreased contentment, which in turn
decreased the probability of disappointment, making the indirect
eect positive.
TABLE2 | Study 1 – rotated component matrix of phenomenology questionnaire.
Item
Component
Self-attribution External attribution Contentment
1 – Ihave to blame myself for the choice Imade. 0.851 0.036 0.202
2 – Ihave to blame myself for the situation Iamcurrently in. 0.767 0.051 0.018
3 – The responsibility for the situation Iamcurrently in is mine. 0.764 0.219 0.262
4 – Iamaware that Icould have chosen differently. 0.735 0.253 0.098
5 – The situation Iamcurrently in is due to the decision Ifreely made. 0.706 0.357 0.275
6 – If Ihad chosen differently. my situation would bebetter now. 0.690 0.219 0.209
7 – The situation Iamcurrently in is due to the decision Iwas forced to make. 0.282 0.729 0.181
8 – If Icould have chosen differently. my current situation would bebetter. 0.169 0.726 0.288
9 – At the time of the decision Iknew that the outcome of my choice would beworse
than Iwanted. 0.066 0.705 0.025
10 – Ihave to blame the external circumstances for the choice Iwas forced to make. 0.499 0.609 0.180
11 – Iamaware Icould not have chosen differently. 0.493 0.608 0.055
12 – Iamhappy with the consequences of my choice. 0.111 0.070 0.890
13 – Things have gone well. 0.002 0.047 0.887
14 – Iamnot happy with the way things turned out. (reversed item) 0.061 0.186 0.699
The bold values indicate the questionnaire items loading on each component.
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 11 December 2021 | Volume 12 | Article 783248
e probability of anger toward circumstances increased
with negative outcome, external attribution, and the choice x
time interaction, whereas decreased with self-attribution,
contentment, and the choice x outcome interaction. Choice
exerted a conditional direct eect through positive outcome
and long term: e eect of negative outcome decreased with
forced choice, where the probability of the emotion was high
in both outcomes (negative interaction choice x outcome); with
forced choice, the probability of the emotion increased in the
long term (positive interaction choice x time). Choice exerted
indirect eects through self-attribution and external attribution
and conditional indirect eect through contentment. More
precisely, irrespective of the values of outcome and time, forced
choice decreased the eect of self-attribution and increased
the eect of external attribution. e former dimension decreased
anger toward circumstances, whereas the latter increased it.
us, the indirect eects became positive in both cases. Instead,
the indirect eect through contentment was signicant only
for positive outcome, in both values of time: In these conditions,
forced choice diminished contentment, which in turn diminished
this emotion, making the conditional indirect eect positive.
e probability of satisfaction decreased with negative outcome
and increased with decision justiability and contentment.
Choice exerted only a conditional indirect eect through
contentment. Indeed, with positive outcome, in both short
and long term, forced choice diminished contentment, which
in turn increased satisfaction. us, such an eect was negative.
e probability of all emotions, except satisfaction, decreased
in males compared to females. No signicant eects were
due to age.
STUDY 2
Study 2 was conducted to investigate regret and other emotional
reactions that originate from decision-making resulting in action.
It had the same structure as study 1, except for the scenarios.
Method
Participants
Once again participants were 336 unpaid volunteers, aged
between 18 and 60 (Mean = 32.25; S.D. = 11.72). ey were
recruited in the same way as study 1, and all were workers.
us, also in this study, participants were equally distributed
by gender. More than half of them (55.7%) had a high school
diploma, 31.8% had a college degree, and 12.5% had completed
compulsory education.
Materials and Procedure
e eight scenarios (choice x outcome x time) were built
around the following plot: e protagonist chooses to leave
a promising basketball career to work in a company. In the
free choice conditions, s/he does not feel like facing the risks
and uncertainties of a sports career and prefers to opt for a
safer job. S/he therefore leaves basketball and nds a job with
a solid company. In the forced choice conditions, a physical
accident prevents the protagonist from continuing his sports
career and heis therefore forced to seek a job that is compatible
with his changed physical conditions. Finally, he nds
employment with a solid company. In the positive outcome
conditions, the protagonist realizes that the new job, although
a bit monotonous, leaves him free time to cultivate his interests
and is also well paid. In the negative outcome conditions,
the protagonist realizes that her job does not highlight her
skills, is monotonous, and does not provide for career
advancement. In all conditions, the protagonist learns that
his place on the basketball team has been taken by another
player who has also turned out to bevery good. e outcome
is evaluated a few days (short-term conditions) or 1 year (long-
term conditions) aer the choice. For each experimental
condition, half of times the protagonist is a woman; the other
half it is a man.
e procedure and the other materials were the same as
study 1. Only the 5 manipulation check questions were slightly
modied to adapt them to the dierent scenarios. For example,
the question on the responsibility was the following: “To what
extent does the protagonist feel responsible to take on another
job instead of a sports career?” e question on the decision
justiability was the following “To what extent does the
protagonist consider the reasons for taking another job instead
of a sports career to be valid?”
Also before this experiment, a pilot study was carried out
with 20 volunteer employees in order to test the plausibility
of the scenarios and the clarity of the instructions. No changes
were needed aer this study.
Results
e same statistical analyses of study 1 were performed also
for study 2.
Manipulation Check
In Table3, the means and standard deviations of the participants
responses to the ve intermediate variables and the ve emotions
are shown as a function of the independent variables.
e results of the 2x2x2 MANCOVA conducted on the
InVs (importance, responsibility, perception of the freedom
of the choice, decision justiability, positive evaluation of
present working condition), with age and gender as covariates,
showed that multivariate tests were signicant for age, for
the three IVs, for choice x time and outcome x time
interactions. e univariate tests revealed that choice aected
all InVs; outcome aected decision justiability and evaluation
of present working condition; time aected the perception
of the freedom of the choice; choice x time aected
responsibility and perception of the freedom of the choice;
outcome x time aected the evaluation of present working
condition. Pairwise comparisons and simple eects analysis,
both of them with Bonferroni adjustment, were used to
investigate main and interaction eects. All InVs increased
with free choice; decision justiability and positive evaluation
of present working condition increased with positive outcome;
in short term, the perception of the freedom of the choice
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 12 December 2021 | Volume 12 | Article 783248
augmented. e interaction between choice and time revealed
that with free choice, the responsibility did not vary as a
function of time, whereas with forced choice, responsibility
was higher in long than in short term. On the contrary,
the perception of the freedom of the choice was independent
of time in the forced choice condition, while it was higher
in short than in long term with free choice. e outcome
x time interaction showed that, with positive outcome, the
evaluation of present working condition did not vary depending
on time, whereas, with negative outcome, this evaluation
was more positive in long than in short time. e evaluation
of the protagonists present working condition was more
positive as the age of the participants increased. No signicant
eects were due to gender. In Supplementary Table 5, the
results of the MANCOVA are reported.
Effects of Choice, Outcome, and Time on
Emotion Intensity
e ve 2x2x2 ANCOVAs conducted on the ve emotions (with
gender and age as covariates) to test the eects of the three
IVs on their intensity showed that all emotions were aected
by choice and by outcome; moreover, regret, disappointment,
and anger toward oneself were also aected by choice x outcome
interaction. More specically, regret, disappointment, and anger
toward circumstances increased with forced choice, whereas
anger toward oneself and satisfaction increased with free choice;
all emotions increased with negative outcome, except for
satisfaction, which increased with positive outcome. e intensity
of regret and disappointment increased with forced choice
compared to free choice, when outcome was positive, while
no signicant dierence depending on choice was found when
the outcome was negative. Anger toward oneself always increased
with free choice compared to forced choice but this eect was
higher with negative than with positive outcome. Regret
diminished when the participants’ age increased, whereas the
opposite eect was found on satisfaction. No signicant eects
were due to gender. In Supplementary Table 6, the results of
the ANCOVAs are reported.
Phenomenological Dimensions and Moderated
Mediation Analyses
e factor analysis performed to reduce the number of items
presented in the questionnaire on the emotional experience
yielded similar results to study 1.
ree components with eigenvalue >1, explaining 62.54%
of the total variance, were extracted following the same procedure
of study 1. e items loading on each component were the
same of study 1. Once again, the components were labelled
self-attribution, external attribution, and contentment. e values
of Cronbachs alpha for each dimension were 0.850, 0.704,
and 0.831, respectively. e results are reported in Tab l e 4.
e factor scores were saved for the subsequent analyses.
Also in study 2, ve moderated mediation analyses were
performed to investigate whether the eect of choice on the
ve emotions was moderated by outcome or time and was
mediated by responsibility, decision justiability, and the three
dimensions extracted from the factor analysis, that is, self-
attribution, external attribution, and contentment. e model
10 of the Process macro 3.1 for SPSS (Hayes, 2018a) was
tested again (see Figure 1) following the same procedure of
study 1.
All results were interpreted with reference to the intercept
(i.e., free choice, positive outcome, and short term).
e results concerning the relationship between choice and/
or the moderators and the mediators (which was the same
for each analysis) were similar to those of the MANCOVA
and were reported in Supplementary Material (see
Supplementary Table 7 and subsequent description).
TABLE3 | Study 2 – means and standard deviations (S.D.) of intermediate variables and emotions as a function of choice, outcome, and time.
Free choice Forced choice
Positive outcome Negative outcome Positive outcome Negative outcome
Short term Long term Short term Long term Short term Long term Short term Long term
Intermediate variables
Decision justiability 6.86 (1.80) 7.02 (1.20) 6.29 (1.86) 6.81 (1.38) 6.71 (2.12) 6.62 (2.16) 5.86 (2.25) 6.00 (1.96)
Evaluation of present work
condition 6.98 (1.24) 6.31 (1.46) 3.21 (1.80) 4.19 (1.89) 6.48 (1.38) 5.71 (1.67) 2.36 (1.34) 2.86 (1.42)
Importance 7.71 (1.53) 7.05 (1.58) 7.69 (1.30) 7.52 (1.53) 7.17 (1.50) 6.86 (2.02) 7.29 (1.94) 7.31 (1.79)
Perception of the freedom of
the choice 6.69 (1.54) 5.24 (2.03) 6.83 (1.48) 5.86 (1.95) 2.12 (1.55) 2.31 (1.52) 2.12 (1.31) 2.17 (1.48)
Responsibility 7.52 (1.27) 6.81 (1.94) 7.17 (1.75) 7.36 (1.69) 2.07 (1.47) 2.76 (2.31) 2.14 (1.49) 3.29 (1.85)
Emotions
Anger toward circumstances 4.43 (2.66) 4.40 (2.30) 5.90 (2.35) 6.33 (2.04) 6.81 (2.06) 7.05 (1.90) 8.19 (1.06) 7.62 (1.75)
Anger toward oneself 3.98 (2.32) 4.02 (2.05) 6.86 (2.08) 6.10 (2.27) 2.88 (1.91) 3.45 (1.81) 4.21 (1.97) 4.43 (2.61)
Disappointment 4.05 (2.11) 4.21 (1.68) 7.24 (2.16) 6.62 (2.12) 5.60 (2.71) 6.31 (2.30) 7.40 (1.91) 7.36 (1.99)
Regret 5.02 (2.23) 4.33 (1.86) 6.98 (2.31) 6.67 (1.97) 6.45 (2.04) 6.71 (1.94) 7.17 (1.83) 7.33 (1.82)
Satisfaction 6.24 (1.19) 5.83 (1.51) 3.24 (1.91) 3.71 (1.86) 5.31 (1.85) 4.86 (2.17) 2.52 (1.53) 2.33 (1.28)
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 13 December 2021 | Volume 12 | Article 783248
Aer including the mediators in the model, the results on
the ve emotions were the following.
e probability of regret increased with forced choice, negative
outcome, self-attribution, and external attribution, while
decreased with contentment, choice × outcome interaction, and
the participants’ age. e conditional direct eect of choice
on the probability of regret was moderated by outcome, being
signicant only for positive outcome (as the ANCOVA showed).
Indeed, the negative interaction between choice and outcome
indicated that with forced choice the eect of negative outcome
decreased because in these conditions the probability of regret
was high in both outcomes (a result already highlighted by
the ANCOVA). e conditional indirect eects of choice through
self-attribution and external attribution were not moderated
by outcome and time, since they were signicant for both
types of outcome and both in short and long term. e signs
of these eects were the same as in study 1. e conditional
indirect eect of choice through contentment was moderated
by outcome, being signicant only with positive outcome, in
both values of time. e sign of this eect was positive, as
well as in study 1.
The probability of anger toward oneself increased with
negative outcome, responsibility, and self-attribution and
decreased with decision justifiability, contentment, and with
the choice x outcome interaction. The negative interaction
replicated the results already showed by ANCOVA, that is,
the probability of this emotion increased with negative
outcome, when the choice was free, but this effect decreased
when the choice was forced because the intensity of anger
toward oneself was low in both outcomes. Choice exerted
an indirect effect through self-attribution and a conditional
indirect effect through contentment. More specifically,
regardless of the values of outcome or time, forced choice
decreased self-attribution, which in turn increased anger
toward oneself: Thus, the indirect effect was negative and
was not moderated by outcome or time. Instead, the indirect
effect of choice was moderated by outcome, being significant
only with positive outcome, both in short and in long term:
Here, forced choice decreased contentment, which in turn
decreased the probability of anger toward oneself, making
the indirect effect positive.
e probability of disappointment increased with negative
outcome and external attribution, while decreased with
contentment and with choice × outcome interaction. e latter
indicated that with negative outcome, the probability of
disappointment increased with free choice but that with forced
choice this eect decreased because the intensity of
disappointment was quite similar in both outcomes. Choice
exerted a positive indirect eect through external attribution
and a positive conditional indirect eect through contentment,
as well as in study 1.
e probability of anger toward circumstances increased
with forced choice, negative outcome, external attribution, and
decreased with responsibility and contentment. e eect of
choice was moderated by outcome, being signicant only with
positive outcome in both values of time: With positive outcome,
the probability of this emotion was higher in forced than in
free choice, while with negative outcome, it was similar in
both types of choice.
Choice exerted an indirect eect through external attribution
and a conditional indirect eect through contentment. More
precisely, irrespective of the values of outcome and time,
forced choice increased the eect of external attribution,
which in turn increased the probability of this emotion: us,
the indirect eect was positive. Instead, the indirect eect
through contentment was signicant only for positive outcome
in both values of time: here forced choice diminished
contentment, which in turn diminished the probability of
anger toward circumstances, making the conditional indirect
eect positive.
The probability of satisfaction decreased with negative
outcome and increased with contentment and the participants
TABLE4 | Study 2 – rotated component matrix of phenomenology questionnaire.
Item
Component
Self-attribution External attribution Contentment
1 – Ihave to blame myself for the choice Imade. 0.852 0.010 0.105
2 – Ihave to blame myself for the situation Iamcurrently in. 0.841 0.074 0.114
3 – The responsibility for the situation Iamcurrently in is mine. 0.740 0.337 0.155
4 – If Ihad chosen differently, my situation would bebetter now. 0.674 0.205 0.260
5 – Iamaware Icould have chosen differently. 0.642 0.338 0.096
6 – The situation Iamcurrently in is due to the decision Ifreely made. 0.637 0.429 0.273
7 – Ihave to blame the external circumstances for the choice Iwas forced to make. 0.243 0.701 0.072
8 – Iknow Icould not have chosen differently. 0.327 0.667 0.000
9 – At the time of the decision Iknew that the outcome of my choice would beworse than Iwanted. 0.097 0.652 0.076
10 – The situation Iamcurrently in is due to the decision Iwas forced to make. 0.359 0.637 0.119
11 – If Icould have chosen differently, my current situation would bebetter. 0.213 0.556 0.358
12 – Things have gone well 0.051 0.038 0.908
13 – Iamhappy with the consequences of my choice. 0.048 0.059 0.901
14 – Iamnot happy with the way things turned out. (reversed item) 0.154 0.253 0.728
The bold values indicate the questionnaire items loading on each component.
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 14 December 2021 | Volume 12 | Article 783248
age. Choice exerted only a conditional indirect effect through
contentment. Indeed, with positive outcome, in both short
and long term, forced choice diminished contentment, which
in turn increased satisfaction. Thus, such an effect
was negative.
e eects of age were found only on regret and satisfaction,
whereas no signicant eects were due to gender. ese results
are reported in Supplementary Table 8.
DISCUSSION
e results of these studies largely support our hypothesis.
We will rst discuss the results that are common to both of
them and then the dierences between them. In particular,
we will discuss the eect of the temporal dimension on the
ve emotions aerward, since it was somewhat dierent in
the two studies.
Both studies showed that a forced choice is a relevant
source of regret. Regardless of whether the decision-making
resulted in action or inaction, regret was more intense when
elicited by a forced than a free choice. Indeed, while regret
was always high when the outcome was negative, irrespective
of the type of choice, forced choice-related regret did not
diminish with a positive outcome, contrarily to regret elicited
by a free choice. Actually, in both studies, this pattern was
similar also for disappointment, a finding already found by
some authors (Connolly and Butler, 2006; Matarazzo and
Abbamonte, 2008), suggesting that for lay people the two
emotions are more similar than both decision theorists and
many psychologists have argued. Moreover, neither regret
nor disappointment was affected by the two appraisal
dimensions we took into account, responsibility for the
choice and decision justifiability. Once again, this finding
disconfirms the prevailing theoretical perspective on regret
which considers responsibility as both a necessary requisite
for regret and a discriminant criterion for differentiating it
from disappointment (see Theoretical background section).
Besides, decision justifiability has been considered as a
powerful condition capable of reducing the regret intensity
(Connolly and Zeelenberg, 2002), at least as to concerns
the amount of regret deriving from deciding badly (Pieters
and Zeelenberg, 2005). However, in our studies, responsibility
and/or decision justifiability did not affect regret but the
two types of anger and satisfaction, as wewill discuss later.
Note that in both studies, the manipulation of responsibility,
and more generally of all intermediate variables, was successful:
Responsibility increased with free compared to forced choice,
suggesting that participants understood the meaning of the term
correctly. As concerns decision justiability, this variable increased
with forced vs. free choice in the rst study, whereas the opposite
result emerged from the second study: Anyway, in both studies,
the average scores were always high or medium-high, implying
that participants felt that the reasons underlying the decision
of the scenario’s protagonist were well-founded. us, the nding
that two appraisal dimensions were irrelevant not only for
disappointment but also for regret indicates that responsibility
is not a necessary requisite for regret and that the DJT (Connolly
and Zeelenberg, 2002) does not take into account all the sources
of regret. Our results reveal that the importance of the forced
choice has been disregarded. However, our studies suggest that
the distinction between regret and disappointment, as well
as the other emotions taken into account, can be identied
in the phenomenological aspects weinvestigated. More precisely,
the moderated mediation analyses showed that, in conformity
with our assumptions, each of the ve emotions considered,
while having some overlapping characteristics, had a specic
prole resulting from the way in which for each of them the
relationship between antecedents, appraisal, and phenomenological
aspects took shape. As mentioned above, to build the
phenomenology questionnaire, wehave used some items describing
the thoughts and feeling linked to a free choice resulting in a
bad outcome, that is, self-blame, upward counterfactuals, awareness
of the freedom of the choice, and self-accountability. In addition,
we have thought of a version of the same items, which was
compatible with a forced choice. Finally, we have used some
items describing the aective (positive and negative) correlates
of the decision consequences. In both studies, the three
phenomenological dimensions (self-attribution, external
attribution, and contentment) were aected by the type of choice
in the expected direction: Self-attribution increased with free
choice and with negative outcome, whereas external attribution
increased with forced choice but was independent of the outcome.
Contentment increased with free choice and positive outcome.
In both studies, the moderated mediation analyses showed
that regret was the only emotion that increased with both
self-attribution and external attribution, whereas decreased with
contentment as well as all other negative emotions. In addition,
it was the only emotion on which choice exerted both a direct
(in interaction with outcome and/or time) and an indirect
(through phenomenology) eect in both studies. In our opinion,
this is the most relevant nding of these studies since it supports
the hypothesis that regret derives from two types of choice
and that the forced choice is at least as important as the free
choice. In addition, it suggests that the emotional experience
of regret is more complex than it is generally thought in literature.
Disappointment increased with external attribution but was
not aected by self-attribution. Consequently, although this emotion
has the same relationship as regret to antecedents and the two
appraisals, its phenomenology is primarily focused on awareness
of forced choice. is result questions the dominant conception
of disappointment because it reveals how central the relationship
with the choice, not just with the outcome, is in its phenomenology.
Anger toward circumstances was the other emotion with
similar characteristics to regret. In both studies, it increased
with forced choice, negative outcome, and external attribution.
Nevertheless, in the rst study, this emotion decreased with
self-attribution, instead of increasing like regret; in the second
study, it decreased when responsibility increased, whereas
responsibility never aects regret.
e negative emotion that most diered from regret was
anger toward oneself. In both studies, it was the only emotion
that increased with free choice in the presence of a negative
outcome. With a positive outcome, in fact, the emotion decreased
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 15 December 2021 | Volume 12 | Article 783248
with both types of choice. Furthermore, it increased with
responsibility, self-attribution, and decreased with decision
justiability, while it was not aected by external attribution.
ese results suggest that, at least for the participants in our
studies, anger toward oneself is the emotion that encompasses
the features that the mainstream of the economic and psychological
literature attribute to regret. We will discuss this issue later.
As to concerns satisfaction – the only positive emotion
we took into account because of manipulation of the outcome
– the results are congruent with the predictions derived by
appraisal theories: In both studies, it increased with positive
outcome, free choice, and contentment; in the rst study, it
also increased with decision justiability. As expected, neither
of the other two dimensions of phenomenology had any eect on it.
Although the results of the two studies were very similar,
some dierences regarding the temporal dimension and the
two covariates emerged: Now, we will briey discuss them.
In study 2, time did not aect the intensity of emotions,
but did aect responsibility, self-attribution, and external
attribution in interaction with choice. Summing up, when the
choice was forced, responsibility and self-attribution tended to
be lower in the short than in the long term, whereas with
free choice, they were always very high. e probability of
external attribution increased in the long term with free choice,
whereas it was always high in both time measures when choice
was forced.
In study 1, the eect of time was more relevant, since it
concerned responsibility, the same two phenomenological
dimensions, and almost all emotions except disappointment.
Responsibility, self-attribution, and external attribution increased
in the long term, even if the eect on self-attribution was
produced in interaction with choice. e eect of time on regret,
anger toward oneself, and anger toward circumstances was
moderated by the three-way interaction with choice and outcome.
us, in neither study did wend suciently clear and robust
results to make a congruent explanatory hypothesis. Although
the temporal dimension was more signicant when decision-
making resulted in inaction rather than in action, our ndings
do not support the temporal theory of regret (Gilovich and
Medvec, 1994, 1995), according to which regret related to an
omission should bemore intense in the long than in the short
term. In our rst study, both regret and the two types of anger
showed such a tendency only in specic interactions with choice
and outcome. Moreover, as to concerns regret, the dierence
in intensity due to the passage of time occurred between two
experimental conditions where the average scores were particularly
low, that is, the conditions with free choice and positive outcome,
compared in short and long term. Consequently, it seems more
appropriate to infer from our results that, regardless whether
regret derives from an omission or a commission, its intensity
is independent of temporal dimension.
e other dierences between our two studies concern the
eect of gender and age. In the rst study, all emotions except
satisfaction increased in females compared to males, whereas
no signicant eects were due to age. In study 2, gender did
not aect results, but regret diminished with age, whereas
satisfaction and contentment increased with it.
To the best of our knowledge, only few studies found a
gender dierence between action and inaction regrets, in the
opposite direction to our studies. In the context of romantic
(Roese etal., 2006) or sexual (Kennair etal., 2016) relationships,
men tend to feel more regrets for initiatives or opportunities
which were not taken (inaction); instead, women tend to regret
choosing an inappropriate partner or having casual sex (action).
In other domains, such as education, achievement, or social
relationships, no gender dierence seems to have been discovered
(e.g., Beike et al., 2009; Davidai and Gilovich, 2018), although
Morrison and Roese (2011) found that men reported more
action-focused life regrets than women. Moreover, some studies
in the career domain have found that women regret not having
followed their calling (Stewart and Vandewater, 1999;
Wrzesniewski et al., 2006; Newton et al., 2012). us,
we speculated that in study 1, women identied more than
men with the protagonist of the omission scenarios and attributed
a greater intensity of regret to him/her than men did.
As regards the eect of age found in study 2, which shows
that as a function of age regret decreased whereas satisfaction
and contentment increased, to our knowledge, no other studies
have found this eect. e studies that focused on the relationship
between regret and subjective well-being in the life cycle
(Jokisaari, 2003; Wrosch etal., 2005) or on collecting life regret
in a US representative sample (Morrison and Roese, 2011)
did not nd any age dierence in the action eect. e suggestion
stemming from the research on emotion regulation, according
to which old adults are oen more capable to regulate negative
emotions than the young (Carstensen et al., 2003; Blanchard-
Fields et al., 2004), is hardly plausible, since this eect was
limited to action-based scenarios and did not extend to omission-
based scenarios. Although this nding does not concern the
main goals of our research nor was expected, we think that
it deserves to be further investigated in future research.
To conclude this discussion, we would like to focus on
two points.
e rst concerns the question of whether the dierence
between our results and the majority of those reported in the
literature is merely a matter of terminology. As we wrote above,
the studies dealing with terms from everyday life, as in our
case, always have to deal with the problem that there is no
one-to-one correspondence between the terms as they are used
by lay people and as they are conceived in the specialist literature
(Nisbett and Wilson, 1977). is is particularly true when referring
to emotions (Sabini and Silver, 2005) and when comparing studies
conducted in dierent countries and with dierent languages
(e.g., Giorgetta etal., 2012). However, in response to this possible
objection, it should be noted that the assumption underlying
the translatability of one language into another is that there is
at least partial overlap between the meaning of the terms in
dierent languages, in the sense that they designate similar referents.
Perhaps the terms rimpianto (or rammarico) and regret are
not completely superimposable, nor are the English and Italian
terms of the other emotions investigated, therefore in our
studies, wehave considered the appraisal and phenomenological
aspects of emotions to overcome this possible issue. e success
of the manipulation check suggests that participants found the
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 16 December 2021 | Volume 12 | Article 783248
proposed stories believable and that they interpreted them
correctly. us, the emotional reaction attributed to the
protagonists of the stories seems to beplausible, and this seems
to confer ecological validity to the studies.
e use of the scenario method is the nal point on which
we would like to focus.
As we wrote in The rationale of the studies section, this
method has the advantage of allowing an accurate manipulation
of the experimental variables, especially if the object of the
studies concerns decisions in a self-relevant domain and
not trivial everyday decisions. The latter can easily be the
object of behavioral experiments, because the manipulation
of the variables related to them does not raise the ethical
issues that would arise in manipulating self-relevant variables.
However, the fact that emotional reactions are attributed
to a third person and not felt in the first person could
cast doubts on the correspondence between the participants’
real subjective experience and the one attributed to the
scenario protagonist. We think there are several ways to
respond to such an objection. The first is that perspective
taking has been considered as a natural aptitude of living
being, rooted in the brain. Notably, simulation theory (Gordon,
1986; Goldman, 1989, 1992; Gallese and Goldman, 1998)
assumes that the natural way of attributing mental states
to others rests on the automatic ability of humans to imagine
what they would feel in the situation the other person is
in and to extend their own thoughts and feelings to the
other person. Second, several studies on perspective taking
reported in The rationale of the studies section showed
that people tend to put themselves in the other person’s
shoes, regardless of whether the instructions request to do
it. Third, the fact that in the scenario method, people do
not feel directly investigated can allow them to express more
freely what they would really think or feel in a similar situation.
Finally, even if one admits that there is not a total correspondence
between the experience in rst and in third person, the scope
of our studies was to compare the emotional reactions elicited,
or supposed to be elicited, by free vs. forced choice. Since these
reactions were investigated through the same method, the
comparison between them remains valid regardless of the possible
dierence between emotions experienced in the rst or third person.
CONCLUSION
Overall, we believe that our results can contribute to a deeper
understanding of regret and other post-decision emotions along
three directions: First, they suggest that the conception of
regret derived from decision theorists is too binding, since it
posits as necessary some requirements which are not. Second,
they reveal that the antecedents and phenomenology of regret
are broader than it is generally believed. ird, they reveal
that decision-making (specially when its object is important
for our life) does not elicit a reduced number of sharply
separated emotions; rather, it produces a complex emotional
constellation, where the dierent emotions, singularly and/or
in combination with the others, constitute the aective responses
to the dierent aspects of the decision-making process. e
emotions wetook into account capture some of these aspects,
although they are far from covering them all.
LIMITATIONS AND FUTURE RESEARCH
In our opinion, the major limitation of our studies is that
we did not consider the possible behavioral consequences of
emotional experiences. In this way, wehave omitted to investigate
the motivational function of emotions (Frijda, 1986; Lazarus,
1991; Roseman, 2011) and, consequently, to examine further
dierences and similarities between the ve emotions considered.
If we had asked the participants what the protagonist of
the scenario was thinking of doing to deal with the situation
she was experiencing, we would have had a more complete
picture of the emotional experiences investigated, since the
behavioral consequences would also have been included. Future
research should investigate these aspects.
DATA AVAILABILITY STATEMENT
e datasets presented in this study can be found in online
repositories. e names of the repository/repositories and
accession number(s) can be found at: https://doi.org/10.6084/
m9.gshare.14444417.v1.
ETHICS STATEMENT
e studies involving human participants were reviewed and
approved by Department of Psychology- University of Campania
Luigi Vanvitelli. e participants provided their written informed
consent to participate in this study.
AUTHOR CONTRIBUTIONS
OM conceptualized the studies, analyzed the data, and wrote
the manuscript. CG and BP conducted literature searches,
collected the data, and prepared the gures and tables. LA
and GN revised the manuscript. All authors contributed to
the article and approved the submitted version.
FUNDING
e department of Psychology of the University of Campania
Luigi Vanvitelli paid the publication fee.
SUPPLEMENTARY MATERIAL
e Supplementary Material for this article can be found
online at: https://www.frontiersin.org/articles/10.3389/fpsyg.2021.
783248/full#supplementary-material
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 17 December 2021 | Volume 12 | Article 783248
REFERENCES
Avenanti, A., Minio-Paluello, I., Bufalari, I., and Aglioti, S. (2006). Stimulus-
driven modulation of motor-evoked potentials during observation of others’
pain. NeuroImage 32, 316–324. doi: 10.1016/j.neuroimage.2006.03.010
Batson, C. D., Early, S., and Salvarani, G. (1997). Perspective taking: imagining
how another feels versus imaging how youwould feel. Personal. Soc. Psychol.
Bull. 23, 751–758. doi: 10.1177/0146167297237008
Beike, D. R., Markman, K. D., and Karadogan, F. (2009). What we regret most
are lost opportunities: a theory of regret intensity. Personal. Soc. Psychol.
Bull. 35, 385–397. doi: 10.1177/0146167208328329
Bell, D. E. (1982). Regret in decision making under uncertainty. Oper. Res.
30, 961–981. doi: 10.1287/opre.30.5.961
Bell, D. E. (1985). Disappointment in decision making under uncertainty. Oper.
Res. 33, 1–27. doi: 10.1287/opre.33.1.1
Berg, J. M., Grant, A. M., and Johnson, V. (2010). When callings are calling:
craing work and leisure in pursuit of unanswered occupational callings.
Organ. Sci. 21, 973–994. doi: 10.1287/orsc.1090.0497
Blanchard-Fields, F., Stein, R., and Watson, T. L. (2004). Age dierences in
emotion-regulation strategies in handling everyday problems. J. Gerontol. B
Psychol. Sci. Soc. Sci. 59, 261–269. doi: 10.1093/geronb/59.6.p261
Buchanan, J., Summerville, A., Lehmann, J., and Reb, J. (2016). e regret
elements scale: distinguishing the aective and cognitive components of
regret. Judgm. Decis. Mak. 11, 275–286.
Carstensen, L. L., Fung, H. H., and Charles, S. T. (2003). Socioemotional
selectivity theory and the regulation of emotion in the second half of life.
Motiv. Emot. 27, 103–123. doi: 10.1023/A:1024569803230
Chambers, J. R., and Davis, M. H. (2012). The role of the self in perspective-
taking and empathy: ease of self-simulation as a heuristic for inferring
empathic feelings. Soc. Cogn. 30, 153–180. doi: 10.1521/soco.2012.30.2.153
Clore, G. L., Ortony, A., and Foss, M. A. (1987). e psychological foundations
of the aective lexicon. J. Pers. Soc. Psychol. 53, 751–766. doi:
10.1037/0022-3514.53.4.751
Connolly, T., and Butler, D. (2006). Regret in economic and psychological
theories of choice. J. Behav. Decis. Mak. 19, 139–154. doi: 10.1002/
bdm.510
Connolly, T., Ordóñez, L. D., and Coughlan, R. (1997). Regret and responsibility
in the evaluation of decision outcomes. Organ. Behav. Hum. Decis. Process.
70, 73–85. doi: 10.1006/obhd.1997.2695
Connolly, T., and Reb, J. (2005). Regret in cancer-related decisions. Health
Psychol. 24, S29–S34. doi: 10.1037/0278-6133.24.4.S29
Connolly, T., and Zeelenberg, M. (2002). Regret in decision making. Cur r. D ir.
Psychol. Sci. 11, 212–216. doi: 10.1111/1467-8721.00203
Davidai, S., and Gilovich, T. (2018). e ideal road not taken: the self-discrepancies
involved in people’s most enduring regrets. Emotion 18, 439–452. doi: 10.1037/
emo0000326
Davis, M. H., Soderlund, T., Cole, J., Gadol, E., Kute, M., Myers, M., et al.
(2004). Cognitions associated with attempts to empathize: how do weimagine
the perspective of another? Personal. Soc. Psychol. Bull. 30, 1625–1635. doi:
10.1177/0146167204271183
Faul, F., Erdfelder, E., Lang, A.-G., and Buchner, A. (2007). G*power 3: a
exible statistical power analysis program for the social, behavioral, and
biomedical sciences. Behav. Res. Methods 39, 175–191. doi: 10.3758/BF03193146
Feldman, G. (2020). What is normal? Dimensions of action-inaction normality
and their impact on regret in the action-eect. Cognit. Emot. 34, 728–742.
doi: 10.1080/02699931.2019.1675598
Feldman, G., Kutscher, L., and Yay, T. (2020). Omission and commission in
judgment and decision making: understanding and linking action-inaction
eects using the concept of normality. Soc. Personal. Psychol. Compass
14:e12557. doi: 10.1111/spc3.12557
Frijda, N.H. (1986). e Emotions. Cambridge, UK: Cambridge University Press.
Frijda, N. H. (1987). Emotion, cognitive structure, and action tendency. Cognit.
Emot. 1, 115–143. doi: 10.1080/02699938708408043
Frijda, N. H. (2009). Emotion experience and its varieties. Emot. R ev. 1, 264–271.
doi: 10.1177/1754073909103595
Gallese, V., and Goldman, A. I. (1998). Mirror neurons and the simulation
theory of mind-reading. Trends Cogn. Sci. 2, 493–501. doi: 10.1016/
S1364-6613(98)01262-5
George, J. M., and Dane, E. (2016). Aect, emotion, and decision making.
Organ. Behav. Hum. Decis. Process. 136, 47–55. doi: 10.1016/j.obhdp.2016.06.004
Gilbert, D. T., Morewedge, C. K., Risen, J. L., and Wilson, T. D. (2004). Looking
forward to looking backward: the misprediction of regret. Psychol. Sci. 15,
346–350. doi: 10.1111/j.0956-7976.2004.00681.x
Gilovich, T., and Medvec, V. H. (1994). The temporal pattern to the
experience of regret. J. Pers. Soc. Psychol. 67, 357–365. doi:
10.1037/0022-3514.67.3.357
Gilovich, T., and Medvec, V. H. (1995). The experience of regret: what,
when, and why. Psychol. Rev. 102, 379–395. doi: 10.1037/0033-
295X.102.2.379
Gilovich, T., Medvec, V. H., and Kahneman, D. (1998). Varieties of regret: a
debate and partial resolution. Psychol. Rev. 105, 602–605. doi: 10.1037/0033-
295X.105.3.602
Giorgetta, C., Zeelenberg, M., Ferlazzo, F., and D'Olimpio, F. (2012). Cultural
variation in the role of responsibility in regret and disappointment: e
Italian case. J. Econ. Psychol. 33, 726–737. doi: 10.1016/j.joep.2012.02.003
Goldman, A. I. (1989). Interpretation psychologized. Mind Lang. 4, 161–185.
doi: 10.1111/j.1468-0017.1989.tb00249.x
Goldman, A. I. (1992). In defense of the simulation theory. Mind Lang. 7,
104–119. doi: 10.1111/j.1468-0017.1992.tb00200.x
Gordon, R. M. (1986). Folk psychology as simulation. Mind Lang. 1, 158–171.
doi: 10.1111/j.1468-0017.1986.tb00324.x
Hayes, A. F. (2018a). Introduction to Mediation, Moderation, and Conditional
Process Analysis: A Regression-Based Approach (2nd Edn.). New York, NY:
Guilford Press.
Hayes, A. F. (2018b). Partial, conditional, and moderated moderated mediation:
quantication, inference, and interpretation. Commun. Monogr. 85, 4–40.
doi: 10.1080/03637751.2017.1352100
Hayes, A. F., and Rockwood, N. J. (2020). Conditional process analysis: concepts,
computation, and advances in the modeling of the contingencies of mechanisms.
Am. Behav. Sci. 64, 19–54. doi: 10.1177/0002764219859633
Heavey, C. L., Leorge, N. L., Lapping-Carr, L., and Hurlburt, R. T. (2017).
Mixed emotions: toward a phenomenology of blended and multiple feelings.
Emot. Re v. 9, 105–110. doi: 10.1177/1754073916639661
Inman, J. J. (2007). Regret regulation: disentangling self-reproach from learning.
J. Consum. Psychol. 17, 19–24. doi: 10.1207/s15327663jcp1701_4
Inman, J. J., Dyer, J. S., and Jia, J. (1997). A generalized utility model of
disappointment and regret eects on post-choice valuation. Mark. Sci. 16,
97–111. doi: 10.1287/mksc.16.2.97
Inman, J. J., and Zeelenberg, M. (2002). Regret in repeat purchase versus
switching decisions: the attenuating role of decision justiability. J. Consum.
Res. 29, 116–128. doi: 10.1086/339925
Izard, C.E. (1991). e Psychology of Emotions. New York: Plenum.
Jokisaari, M. (2003). Regret appraisals, age, and subjective well-being. J. Res.
Pers. 37, 487–503. doi: 10.1016/S0092-6566(03)00033-3
Kahneman, D. (1995). “Varieties of counterfactual thinking,” in What Might
Have Been: e Social Psychology of Counterfactual inking. eds. N. J. Roese
and J. M. Olson (Mahwah: Lawrence Erlbaum Associates, Inc.), 375–396.
Kahneman, D., and Miller, D. T. (1986). Norm theory: comparing reality to
its alternatives. Psychol. Rev. 93, 136–153. doi: 10.1037/0033-295X.93.2.136
Kennair, L. E. O., Bendixen, M., and Buss, D. M. (2016). Sexual regret: tests
of competing explanations of sex dierences. Evol. Psychol. 14, 1–9. doi:
10.1177/1474704916682903
Kwong, J. Y., Wong, K. F. E., and Tang, S. K. (2013). Comparing predicted
and actual aective responses to process versus outcome: an emotion-as-
feedback perspective. Cognition 129, 42–50. doi: 10.1016/j.cognition.2013.05.012
Landman, J. (1987). Regret: a theoretical and conceptual analysis. J. eory
Soc. Behav. 17, 135–160. doi: 10.1111/j.1468-5914.1987.tb00092.x
Landman, J. (1993). Regret: e Persistence of the Possible. New York/Oxford:
Oxford University Press.
Lazarus, R. S. (1991). Progress on a cognitive-motivational-relational theory
of emotion. Am. Psychol. 46, 819–834. doi: 10.1037/0003-066X.46.8.819
Lerner, J. S., Li, Y., Valdesolo, P., and Kassam, K. S. (2015). Emotion and
decision making. Annu. Rev. Psychol. 66, 799–823. doi: 10.1146/annurev-
psych-010213-115043
Loomes, G., and Sugden, R. (1982). Regret theory: an alternative theory of
rational choice under uncertainty. Econ. J. 92, 805–824. doi: 10.2307/2232669
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 18 December 2021 | Volume 12 | Article 783248
Loomes, G., and Sugden, R. (1986). Disappointment and dynamic consistency
in choice under uncertainty. Rev. Econ. Stud. 53, 271–282. doi: 10.2307/2297651
Marcatto, F., Cosulich, A., and Ferrante, D. (2015). Once bitten, twice shy:
experienced regret and non-adaptive choice switching. PeerJ 3:e1035. doi:
10.7717/peerj.1035
Marcatto, F., and Ferrante, D. (2008). e regret and disappointment scale: an
instrument for assessing regret and disappointment in decision making.
Judgm. Decis. Mak. 3, 87–99.
Matarazzo, O., and Abbamonte, L. (2008). Regret, choice, and outcome. Intl.
J. Phys. Beh. Sci. 2, 225–233. doi: 10.5281/zenodo.1074950
Mellers, B. A. (2000). Choice and the relative pleasure of consequences. Psychol.
Bull. 126, 910–924. doi: 10.1037/0033-2909.126.6.910
Morrison, M., and Roese, N. J. (2011). Regrets of the typical American: ndings
from a nationally representative sample. Soc. Psychol. Personal. Sci. 2, 576–583.
doi: 10.1177/1948550611401756
Muller, D., Judd, C. M., and Yzerbyt, V. Y. (2005). When moderation is mediated
and mediation is moderated. J. Pers. Soc. Psychol. 89, 852–863. doi:
10.1037/0022-3514.89.6.852
Newton, N., Torges, C., and Stewart, A. (2012). Women’s regrets about their
lives: cohort dierences in correlates and contents. Sex Roles 66, 530–543.
doi: 10.1007/s11199-012-0126-6
Nisbett, R. E., and Wilson, T. D. (1977). Telling more than we can know:
verbal reports on mental processes. Psychol. Rev. 84, 231–259. doi:
10.1037/0033-295X.84.3.231
Ordóñez, L. D., and Connolly, T. (2000). Regret and responsibility: a reply to
Zeelenberg et al. (1998). Organ. Behav. Hum. Decis. Process. 81, 132–142.
doi: 10.1006/obhd.1999.2834
Pieters, R., and Zeelenberg, M. (2005). On bad decisions and deciding badly:
when intention–behavior inconsistency is regrettable. Organ. Behav. Hum.
Decis. Process. 97, 18–30. doi: 10.1016/j.obhdp.2005.01.003
Roese, N. J. (1999). Counterfactual thinking and decision making. Psychon.
Bull. Rev. 6, 570–578. doi: 10.3758/BF03212965
Roese, N. J., Pennington, G. L., Coleman, J., Janicki, M., Li, N. P., and
Kenrick, D. T. (2006). Sex differences in regret: all for love or some
for lust? Personal. Soc. Psychol. Bull. 32, 770–780. doi: 10.1177/0146167
206286709
Roese, N. J., and Summerville, A. (2005). What we regret most…and
wh y. Personal. Soc. Psychol. Bull. 31, 1273–1285. doi: 10.1177/0146167
205274693
Rorty, A. O. (1980). “Agent regret,” in Explaining Emotions. ed. A. O. Rorty
(Los Angeles: University of California Press), 489–506.
Roseman, I. J. (2011). Emotional behaviors, emotivational goals, emotion strategies:
multiple levels of organization integrate variable and consistent responses.
Emot. Re v. 3, 434–443. doi: 10.1177/1754073911410744
Roseman, I. J., Wiest, C., and Swartz, T. S. (1994). Phenomenology, behaviors,
and goals dierentiate discrete emotions. J. Pers. Soc. Psychol. 67, 206–211.
doi: 10.1037/0022-3514.67.2.206
Sabini, J., and Silver, M. (2005). Why emotion names and experiences don't
neatly pair. Psychol. Inq. 16, 1–10. doi: 10.1207/s15327965pli1601_01
Savitsky, K., Medvec, V. H., and Gilovich, T. (1997). Remembering and regretting:
the Zeigarnik eect and the cognitive availability of regrettable actions and
inactions. Personal. Soc. Psychol. Bull. 23, 248–257. doi: 10.1177/01461
67297233004
Scherer, K. R., and Fontaine, J. R. J. (2019). e semantic structure of emotion
words across languages is consistent with componential appraisal models
of emotion. Cognit. Emot. 33, 673–682. doi: 10.1080/02699931.2018.1481369
Seta, J. J., McElroy, T., and Seta, C. E. (2001). To do or not to do: desirability
and consistency mediate judgments of regret. J. Pers. Soc. Psychol. 80, 861–870.
doi: 10.1037/0022-3514.80.6.861
Seta, C. E., and Seta, J. J. (2013). Regret in pursuit of change and maintenance
goals. Motiv. Emot. 37, 177–184. doi: 10.1007/s11031-012-9283-8
Seta, C. E., Seta, J. J., McElroy, G. T., and Hatz, J. (2008). Regret: the roles
of consistency-t and counterfactual salience. Soc. Cogn. 26, 700–719. doi:
10.1521/soco.2008.26.6.700
Sini, B., Tinti, C., and Galati, D. (2014). “Semantic structure of emotional
lexicon” in Encyclopedia of Quality of Life and Well-Being Research. ed. A.
C. Michalos (Dordrecht: Springer)
Solomon, R. (1983). e Passions: e Myth and Nature of Human Emotion.
Notre Dame: Notre Dame University Press.
Stewart, A. J., and Vandewater, E. A. (1999). “If I had it to do over again…”:
midlife review, midcourse corrections, and women’s well-being in midlife.
J. Pers. Soc. Psychol. 76, 270–283. doi: 10.1037/0022-3514.76.2.270
Sullivan, S. E., Forret, M. L., and Mainiero, L. A. (2007). No regrets? An
investigation of the relationship between being laid o and experiencing
career regrets. J. Manag. Psychol. 22, 787–804. doi: 10.1108/02683940710837723
Summerville, A., and Buchanan, J. (2014). Functions of personal experience
and of expression of regret. Personal. Soc. Psychol. Bull. 40, 463–475. doi:
10.1177/0146167213515026
Taylor, G. (1985). Pride, Shame, and Guilt: Emotions of Self-Assessment. 1st
Edn Oxford University Press.
Towers, A., Williams, M. N., Hill, S. R., Philipp, M. C., and Flett, R. (2016).
What makes for the most intense regrets? Comparing the eects of several
theoretical predictors of regret intensity. Front. Psychol. 7:1941. doi: 10.3389/
fpsyg.2016.01941
Tsiros, M. (1998). Eect of regret on post-choice valuation: the case of more
than two alternatives. Organ. Behav. Hum. Decis. Process. 76, 48–69. doi:
10.1006/obhd.1998.2793
Tsiros, M., and Mittal, V. (2000). Regret: a model of its antecedents and consequences
in consumer decision making. J. Consum. Res. 26, 401–417. doi: 10.1086/209571
van Dijk, W. W., and Zeelenberg, M. (2002). Investigating the appraisal patterns
of regret and disappointment. Motiv. Emot. 26, 321–331. doi: 10.1023/A:
1022823221146
van Dijk, E., and Zeelenberg, M. (2005). On the psychology of ‘if only’: regret
and the comparison between factual and counterfactual outcomes. Organ.
Behav. Hum. Decis. Process. 97, 152–160. doi: 10.1016/j.obhdp.2005.04.001
Verbruggen, M., and De Vos, A. (2020). When people don’t realize their career
desires: toward a theory of career inaction. Acad. Manag. 45, 376–394. doi:
10.5465/amr.2017.0196
Watson, D., and Stanton, K. (2017). Emotion blends and mixed emotions in
the hierarchical structure of aect. Emot. Rev. 9, 99–104. doi: 10.1177/
1754073916639659
Wilkinson, M. R., Ball, L. J., and Alford, D. (2015). Counterfactual reasoning for
regretted situations involving controllable versus uncontrollable events: the
modulating role of contingent self-esteem. ACP 11, 22–30. doi: 10.5709/acp-0168-4
Wrosch, C., Bauer, I., and Scheier, M. F. (2005). Regret and quality of life
across the adult life span: the inuence of disengagement and available
future goals. Psychol. Aging 20, 657–670. doi: 10.1037/0882-7974.20.4.657
Wrzesniewski, A., Tosti, J., and Landman, J. (2006). If I could turn back time:
occupational regret and its consequences for work and life. Working paper.
Yale School of Management, CT./May Meaning Meeting 5/7/05. Available
at: https://www.slideserve.com/lonato/if-i-could-turn-back-time-occupational-
regret-and-its-consequences-for-work-and-life. (Accessed July 2021).
Zeelenberg, M., Inman, J. J., and Pieters, R. G. (2001). “What we do when
decisions go awry: behavioral consequences of experienced regret,” in Conict
and Tradeos in Decision Making. eds. E. U. Weber, J. Baron and G. Loomes
(Cambridge, United Kingdom: Cambridge University Press), 136–155.
Zeelenberg, M., and Pieters, R. (2007). A theory of regret regulation 1.0. J.
Consum. Psychol. 17, 3–18. doi: 10.1207/s15327663jcp1701_3
Zeelenberg, M., van den Bos, K., van Dijk, E., and Pieters, R. (2002). e
inaction eect in the psychology of regret. J. Pers. Soc. Psychol. 82, 314–327.
doi: 10.1037/0022-3514.82.3.314
Zeelenberg, M., Van der Pligt, J., and De Vries, N. K. (2000b). Attributions
of responsibility and aective reactions to decision outcomes. Acta Psychol.
104, 303–315. doi: 10.1016/S0001-6918(00)00034-2
Zeelenberg, M., van Dijk, W. W., and Manstead, A. S. R. (1998b). Reconsidering
the relation between regret and responsibility. Organ. Behav. Hum. Decis.
Process. 74, 254–272. doi: 10.1006/obhd.1998.2780
Zeelenberg, M., van Dijk, W. W., Manstead, A. S., and van der Pligt, J. (2000a).
On bad decisions and disconrmed expectancies: the psychology of regret
and disappointment. Cognit. Emot. 14, 521–541. doi: 10.1080/026999300402781
Zeelenberg, M., van Dijk, W. W., Manstead, A. S. R., and van der Pligt, J.
(1998a). e experience of regret and disappointment. Cognit. Emot. 12,
221–230. doi: 10.1080/026999398379727
Zeelenberg, M., van Dijk, W. W., van der Pligt, J., Manstead, A. S., van
Empelen, P., and Reinderman, D. (1998c). Emotional reactions to the outcomes
of decisions: the role of counterfactual thought in the experience of regret
and disappointment. Organ. Behav. Hum. Decis. Process. 75, 117–141. doi:
10.1006/obhd.1998.2784
Matarazzo et al. Regret and Post-decision Emotions
Frontiers in Psychology | www.frontiersin.org 19 December 2021 | Volume 12 | Article 783248
Zeigarnik, B. (1935). “On nished and unnished tasks,” in A Dynamic eory
of Personality. ed. K. Lewin (New York, NY: McGraw-Hill), 300–314. Original
work published in 1927.
Conict of Interest: e authors declare that the research was conducted in
the absence of any commercial or nancial relationships that could be construed
as a potential conict of interest.
Publisher’s Note: All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their aliated organizations,
or those of the publisher, the editors and the reviewers. Any product that may
be evaluated in this article, or claim that may be made by its manufacturer, is
not guaranteed or endorsed by the publisher.
Copyright © 2021 Matarazzo, Abbamonte, Greco, Pizzini and Nigro. is is an open-
access article distributed under the terms of the Creative Commons Attribution License
(CC BY). e use, distribution or reproduction in other forums is permitted, provided
the original author(s) and the copyright owner(s) are credited and that the original
publication in this journal is cited, in accordance with accepted academic practice.
No use, distribution or reproduction is permitted which does not comply with these terms.
Article
Background: Ninety percent of parents of pediatric oncology patients report distressing, emotionally burdensome healthcare interactions. Assuring supportive, informative treatment discussions may limit parental distress. Here, we interview parents of pediatric surgical oncology patients to better understand parental preferences for surgical counseling. Methods: We interviewed 10 parents of children who underwent solid tumor resection at a university-based, tertiary children's hospital regarding their preferences for surgical discussions. Thematic content analysis of interview transcripts was performed using deductive and inductive methods. Results: Three main themes were identified: (1) the emotional burden of a pediatric cancer diagnosis; (2) complexities of treatment discussions; (3) collaborative engagement between parents and surgeons. Within the collaborative engagement theme, there were four sub-themes: (1) variable informational needs; (2) parents as advocates; (3) parents as gatekeepers of information delivery to their children, family, friends, and community; (4) parental receptivity to structured guidance to support treatment discussions. Two cross-cutting themes were identified: (1) perception that no treatment decision needed to be made regarding surgery and (2) reliance on diverse support resources. Conclusions: Parents feel discussions with surgeons promote informed involvement in their child's care, but they recognize that there may be few decisions to make regarding surgery. Even when parents perceive that there are there are no decisions to make, they prioritize asking questions to advocate for their children. The emotional burden of a cancer diagnosis often prevents parents from knowing what questions to ask. Merging this data with our prior pediatric surgeon interviews will facilitate development of a novel decision support tool that can empower parents to ask meaningful questions. Level of evidence: III.
Article
Full-text available
Research on action and inaction in judgment and decision making now spans over 35 years, ‎with ever-growing interest. Accumulating evidence suggests that action and inaction are ‎perceived and evaluated differently, affecting a wide array of psychological factors from ‎emotions to morality. These asymmetries have been shown to have real impact on choice ‎behavior in both personal and interpersonal contexts, with implications for individuals and ‎society. We review impactful action-inaction related phenomena, with a summary and ‎comparison of key findings and insights, reinterpreting these effects and mapping links ‎between effects using norm theory's (Kahneman & Miller, 1986) concept of normality. ‎Together, these aim to contribute towards an integrated understanding of the human psyche ‎regarding action and inaction.‎
Article
Full-text available
The widely-replicated action-effect (Kahneman & Tversky, 1982a) describes a phenomenon in ‎which negative outcomes are associated with higher regret when they are a result of action ‎compared to inaction. The highly influential norm-theory (Kahneman & Miller, 1986) ‎theorized that the effect could be explained using the concept of normality, arguing that ‎inaction is more ‘normal’. I aimed to clarify the concept of normality and examine the impact ‎on regret in the action-effect by contrasting three identified categories: past-behavior ‎normality, expectations normality, and social-norms normality. In three exploratory ‎experiments (N1 = 213, N2 = 300, N3 = 303) and one concluding pre-registered combined ‎experiment (N = 403) I found that the three normality categories had distinct effects with ‎consistent medium to strong impact on the regret action-effect (d = .51 to d =.85) and no ‎interactions. The action-effect was significantly weakened into an inaction-effect in the joint ‎effects of any two types of the three normality categories (d = 1.56 to 1.61) and with all three ‎combined (d = 2.75). In total, I concluded three replications for effects of each of the normality ‎dimensions, overall nine successful replications of previous findings.‎ All materials, data, and code are available on https://osf.io/wmkpe/. http://mgto.org/wp-content/uploads/2019/12/Feldman-2020-Cogemo-What-is-normal-normality-action-effect-combined.pdf
Article
Behavioral scientists use mediation analysis to understand the mechanism(s) by which an effect operates and moderation analysis to understand the contingencies or boundary conditions of effects. Yet how effects operate (i.e., the mechanism at work) and their boundary conditions (when they occur) are not necessarily independent, though they are often treated as such. Conditional process analysis is an analytical strategy that integrates mediation and moderation analysis with the goal of examining and testing hypotheses about how mechanisms vary as a function of context or individual differences. In this article, we provide a conceptual primer on conditional process analysis for those not familiar with the integration of moderation and mediation analysis, while also describing some recent advances and innovations for the more experienced conditional process analyst. After overviewing fundamental modeling principles using ordinary least squares regression, we discuss the extension of these fundamentals to models with more than one mediator and more than one moderator. We describe a differential dominance conditional process model and overview the concepts of partial, conditional, and moderated moderated mediation. We also discuss multilevel conditional process analysis and comment on implementation of conditional process analysis in statistical computing software.
Article
Appraisal theories of emotion, and particularly the Component Process Model, claim that the different components of the emotion process (action tendencies, physiological reactions, expressions, and feeling experiences) are essentially driven by the results of cognitive appraisals and that the feeling component constitutes a central integration and representation of these processes. Given the complexity of the proposed architecture, comprehensive experimental tests of these predictions are difficult to perform and to date are lacking. Encouraged by the “lexical sedimentation” hypothesis, here we propose an indirect examination of the compatibility of the theoretical assumptions with the semantic structure of a set of major emotion words as measured in a cross-language and cross-cultural study. Specifically, we performed a secondary analysis of the large-scale data set with ratings of affective features covering all components of the emotion process for 24 emotion words in 27 countries, constituting profiles of emotion-specific appraisals, action tendencies, physiological reactions, expressions, and feeling experiences. The results of a series of hierarchical regression analyses to examine the prediction of the theoretical model are highly consistent with the claim that appraisal patterns determine the structure of the response components, which in turn predict central dimensions of the feeling component.
Article
Mediation of X’s effect on Y through a mediator M is moderated if the indirect effect of X depends on a fourth variable. Hayes [(2015). An index and test of linear moderated mediation. Multivariate Behavioral Research, 50, 1–22. doi:10.1080/00273171.2014.962683] introduced an approach to testing a moderated mediation hypothesis based on an index of moderated mediation. Here, I extend this approach to models with more than one moderator. I describe how to test if X’s indirect effect on Y is moderated by one variable when a second moderator is held constant (partial moderated mediation), conditioned on (conditional moderated mediation), or dependent on a second moderator (moderated moderated mediation). Examples are provided, as is a discussion of the visualization of indirect effects and an illustration of implementation in the PROCESS macro for SPSS and SAS.
Article
Research on the structural features of people’s most enduring regrets has focused on whether they result from having acted or having failed to act. Here we focus on a different structural feature, their connection to a person’s self-concept. In 6 studies, we predict and find that people’s most enduring regrets stem more often from discrepancies between their actual and ideal selves than their actual and ought selves. We also provide evidence that this asymmetry is at least partly due to differences in how people cope with regret. People are quicker to take steps to cope with failures to live up to their duties and responsibilities (ought-related regrets) than their failures to live up to their goals and aspirations (ideal-related regrets). As a consequence, ideal-related regrets are more likely to remain unresolved, leaving people more likely to regret not being all they could have been more than all they should have been.
Article
After using descriptive experience sampling to study randomly selected moments of inner experience, we make observations about feelings, including blended and multiple feelings. We observe that inner experience usually does not contain feelings. Sometimes, however, feelings are directly present. When feelings are present, most commonly they are unitary. Sometimes people experience separate emotions as a single experience, which we call a blended feeling. Occasionally people have multiple distinct feelings present simultaneously. These distinct multiple feelings can be of opposite valence, with one pleasant and the other unpleasant. We provide examples that inform theories of emotions and discuss the important role observational methodology plays in the effort to understand inner experience including feelings.
Article
We explore the implications of a hierarchical structure, consisting of (a) the higher order dimensions of nonspecific Positive Activation and Negative Activation and (b) multiple specific negative affects (e.g., fear, sadness, and anger) and positive affects (e.g., joviality, self-assurance, and attentiveness) at the lower level. Emotional blends of the same valence (e.g., simultaneously experiencing both fear and sadness) are an essential part of this structure and form the basis of the higher order Negative and Positive Activation dimensions. Mixed cross-valence emotions (e.g., feeling both nervous and alert) are not central to this hierarchical scheme but are compatible with it. We examine the frequency of pure emotional states, same-valence emotional blends, and cross-valence mixed emotions in a large momentary mood sample.