University of Southern California Dissertations and Theses

Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults

(USC Thesis Other)

Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1

Reacting to the negative consequences of your decisions: An ATSS comparison of cognitive and
affective reactions in older versus younger adults

Michelle C. Feng, MA
University of Southern California
Clinical Psychology
Dissertation
April, 2014

Committee Members:
Gerald Davison, Chair
Steven Lopez
Mara Mather
David Walsh
Florence Clark

2

Acknowledgments
Thank you to Steve Lopez and Florence Clark for taking the time to support me and serve
on my committee in addition to their many departmental responsibilities. I am grateful to David
Walsh for sharing his intelligence and decision-making expertise with me, and for his
willingness to engage in intellectually stimulating theoretical, philosophical, and overall
completely enjoyable conversations over the years. Thank you to Mara Mather for always taking
the time to provide me with invaluable advice on conducting studies with older adults and for
suggesting ways to continually improve as a researcher. She has been such a strong source of
support for me ever since my first year proposal, and I am so appreciative for all her help and
confidence in my abilities. I would also like to thank Richard John for consulting with me on all
things statistically-related. His role has been a consistent and important one in the development
of my dissertation. Thank you to my family, my parents and sister, my fiancé, and my friends,
who have showered me with unwavering and unconditional emotional support over the years.
Thank you to my lab mates, Kean Hsu, Kalina Babeva, and Justin Hummer for serving as my
trusted consultants and comrades, who can truly empathize with the joys and frustrations of
implementing lab-based studies. I am grateful to have had such brilliant and dedicated research
assistants, Brittany Frederick, Kimberly Morton, Artemis Zavaliangos-Petropulu, Erin Lee,
Michael Sherwood, and Leslie Beightler. As I have told them several times, they are all
wonderful and appreciated. Lastly, my sincerest gratitude to my advisor, Jerry Davison a.k.a. Dr.
Jerry a.k.a. Doktorvater (du bist der beste), whose guidance and wisdom have helped me
navigate the waves throughout my graduate career. He has always known how and when to
support, to push, to foster my growth as a researcher, professional, person. I feel truly lucky to
have him as a mentor and I look forward to continuing to learn from him in the years ahead.
3

Table of Contents
Acknowledgments........................................................................................................................... 2
List of Tables .................................................................................................................................. 4
List of Figures ................................................................................................................................. 5
List of Appendices .......................................................................................................................... 6
Abstract ........................................................................................................................................... 7
Introduction ..................................................................................................................................... 8
The Positivity Effect.................................................................................................................... 8
Understanding the Positivity Effect .......................................................................................... 10
Coping Style .............................................................................................................................. 13
Decision Making ....................................................................................................................... 15
Regret ........................................................................................................................................ 18
The Present Study...................................................................................................................... 20
Method .......................................................................................................................................... 23
Procedure ................................................................................................................................... 24
Measures of Emotional Response ............................................................................................. 27
ATSS. .................................................................................................................................... 27
Self-reported regret and mood. .............................................................................................. 30
Results ........................................................................................................................................... 31
Perceptions Regarding the Airport Scenario ............................................................................. 32
Manipulation and Scenario Check ............................................................................................ 32
ATSS Coded Variables ............................................................................................................. 33
Self-reported Feelings of Regret ............................................................................................... 44
Future Time Perspective and Use of Coping Strategies ............................................................ 45
Results by Hypothesis ............................................................................................................... 46
Discussion ..................................................................................................................................... 49
Summary and Conclusions ........................................................................................................ 63
References ..................................................................................................................................... 65

4

List of Tables
Table 1: Table of conditions
Table 2: Participant characteristics at baseline
Table 3: Age differences in Perceptions of air travel and ATSS scenario
Table 4: Manipulation check and external validity of scenario conditions
Table 5: Means, SDs, estimated marginal means, and SEs for younger and older adults’
expressions of ATSS coded emotions, coping language, and Likert-rated regret by
condition
Table 6: Results for ATSS responses and self-reported regret
Table 7: Results for ATSS coping measure

5

List of Figures

Figure 1: ATSS-coded expressed regret for younger and older adults by condition
Figure 2: ATSS-coded expressed anger at others for younger and older adults by condition
Figure 3: ATSS-coded expressed anxiety for younger and older adults by condition
Figure 4: ATSS-coded expressed sadness for younger and older adults by condition
Figure 5: ATSS-coded P-E coping language for younger and older adults by condition
Figure 6: ATSS-coded emotion-focused coping language for younger and older adults by
condition
Figure 7: ATSS-coded problem-focused coping language for younger and older adults by
condition
Figure 8: Self-rated feelings of regret for younger and older adults by condition (Likert rated)

6

List of Appendices
Appendix A: Flow chart of study procedure
Appendix B: Event description
Appendix C: Instructions for the ATSS scenario
Appendix D: Scripts for choice and no choice (control) conditions
Appendix E: ATSS coding procedure

7

Abstract
Research on age differences in attention to and memory of emotional stimuli has shown
that compared to younger adults, older adults focus relatively more on positive and relatively less
on negative stimuli. We employed the Articulated Thoughts in Simulated Situations think-aloud
paradigm to determine whether this positivity effect in older adults extends to the way they
respond to the negative consequences of decisions. Younger and older adult participants were
asked to imagine themselves in a hypothetical airport scenario. One condition required them to
choose between their existing flight and an alternative flight heading to the same destination. An
unpleasant situation related to their choice unfolds resulting in a delayed flight. The control
condition engaged participants in the same scenario without the choice point. Younger adults in
our study were found to express more regret, guilt, anger at others, anger at themselves, anxiety,
and sadness compared to older adults when reacting to the same situations. Participants in the
choice condition were found to express more regret, guilt, and anger at the self compared to
those in the control condition. Furthermore, an age by condition interaction was found for
expressions of coping language (problem-focused and emotion-focused) which revealed older
adults’ flexibility in their use of coping strategies depending on the type of situation. These
findings support prior research and provide evidence for a positivity effect in older adults in
reaction to negative decision outcomes. The ATSS think-aloud procedure allowed us to examine
in ways not available with more conventional methods the immediate self-reported thoughts and
feelings of subjects to experimenter-controlled social situations including a control context.

8

Reacting to the negative consequences of your decisions: An ATSS comparison of cognitive
and affective reactions in older versus younger adults
Differences in the way older and younger adults attend to and recall emotional
information have been studied extensively, with research revealing a positivity bias (often
referred to as the “positivity effect”) in older adults compared with younger adults. These
differences have been shown to extend to age differences in decision making, with older adults
focusing more on positive than negative information when making a choice compared with
younger adults. However, few studies have tested age differences in the way individuals react to
the consequences of those decisions. The way people feel about the outcome of their decisions is
an important component of the decision-making process that is often ignored in decision-making
research. The present study seeks to provide new information regarding age-related differences
in how people react cognitively and emotionally to decisions that go awry.
The Positivity Effect
What is the positivity effect? Research on aging and emotion indicates a shift with
increasing age in the ratio of positive versus negative information attended to and remembered,
with older adults recalling and attending to more positive than negative information compared to
younger adults. This age by valence interaction is known as the “positivity effect” and has been
hypothesized to be the result of a greater focus among older adults on regulating emotion than
among younger adults (Mather & Carstensen, 2005; Reed & Carstensen, 2012; Scheibe &
Carstensen, 2010). For example, studies using attention probe and eye tracking methods reveal
that, when shown two pictures or faces at the same time, older adults spend relatively more time
looking at positive stimuli and relatively less of their time looking at negative stimuli compared
9

to younger adults (Isaacowitz, Toner, Goren & Wilson, 2006a, 2006b; Knight et al., 2007;
Mather & Carstensen, 2003; Rösler et al., 2005).
A positivity effect in memory among older adults is also commonly found, in that they
tend to recall more positive than negative stimuli in comparison to younger adults (Charles,
Mather, & Carstensen, 2003; Leigland, Schulz, & Janowsky, 2004; Mather & Carstensen, 2003;
Mather & Knight, 2005). For example, Charles and colleagues (2003) found that after showing
younger and older adults pictures varying in emotional valence, older adults later recalled more
positive compared to negative pictures. However, younger adults did not show a similar
positivity effect, and instead recalled similar amounts of positive and negative pictures. With
regard to emotional experiences, older adults have also been shown to distort their memory of
past events more positively than younger adults (Kennedy, Mather, & Carstensen, 2004).
In addition to attention and memory, studies have also found age differences in emotional
experience. Overall, older adults report experiencing more satisfaction and emotional well-being
compared to younger adults (for a review, see Scheibe & Carstensen, 2010). A longitudinal study
was conducted by Carstensen and colleagues (2011) measuring emotional experiences (e.g.,
anger, guilt, excitement, joy) in daily life multiple times a day for one week. This was done in
three waves approximately five years apart, over a span of thirteen years. Results revealed that
increasing age was associated with improvement in overall emotional well-being and that
emotional well-being was predictive of survival.
Age-related differences in experience of discrete emotions have also been examined. For
example, older adults have reported experiencing less anger in their daily lives and when
responding to interpersonal situations compared to younger adults (Birditt & Fingerman, 2003;
Birditt & Fingerman, 2005; Blanchard-Fields & Coats, 2008; Charles & Carstensen, 2008;
10

Chipperfield, Perry, & Weiner, 2003). Older adults have also reported lower rates of anxiety,
worry, and major depression than younger adults (see Basevitz, Pushkar, Chaikelson, Conway, &
Dalton, 2008; Blazer, 2003; Piazza & Charles, 2006). Assessing discrete emotions has aided our
understanding of which emotions are more or less susceptible to age-related positivity effects.
Fear, for example, has not been shown to follow the same pattern (Mather & Knight, 2006),
perhaps because alertness of threatening stimuli is a more automatic response and therefore less
under our control compared to other emotions. Sadness also appears to be an emotion that is less
susceptible to age-related biases, with inconsistent findings of the positivity effect from different
researchers (Birditt & Fingerman, 2003; Blanchard-Fields & Coats, 2008; Charles & Carstensen,
2008; Charles, Carstensen, & McFall, 2001).
While there is a rapidly growing literature assessing age-related differences in the
expression and memory of negative emotions, little exists specifically in the realm of emotion-
related choice outcomes. Do older and younger adults react differently when experiencing the
same unfavorable outcome for which they feel responsible? And if so, which emotions reveal an
age-related positivity effect? Examining these questions would help to bridge the literature on
age-related changes in emotional reactions and literature on decision-making and the emotional
consequences of making a decision.
Understanding the Positivity Effect
The positivity effect in older adults has been hypothesized to be due to a motivational
shift and increased focus on emotionally meaningful goals, which typically results in more
complex emotional experiences as well as better emotion regulation (Carstensen, Fung, &
Charles, 2003). Regulating emotions can be accomplished by focusing on maximizing
experienced positive affect (e.g. seeking company of loved ones) and/or minimizing negative
11

affect (avoidance of conflict). While modest increases in positive affect have been shown
(Mroczek & Kolarz, 1998), studies have more consistently found a decrease in negative affect
with increasing age (e.g., Charles, Reynolds, & Gatz, 2001; Carstensen, Pasupathi, Mayr, &
Nesselroade, 2000). Perhaps for this reason, the overwhelming majority of studies have focused
on age differences in negative emotions or general feelings related to well-being.
Socioemotional Selectivity theory (SST) provides one explanation for this positivity
effect, positing that this bias is due primarily to a shifting of focus from future-oriented goals to
more present-oriented goals as individuals perceive a decreasing amount of time left in their lives
(Carstensen, Isaacowitz, & Charles, 1999; Carstensen, Mikels, & Mather, 2006). According to
SST, due to their more apparent bounded future, older adults compared to younger adults tend to
increase their focus on goals that can be achieved in the present, such as regulating emotions in a
way that maximizes positive feelings and/or minimizes negative feelings. On the other hand,
young adults are more likely to seek goals that provide future realized benefit, such as attend
networking functions to acquire contacts. Examining the relationship between perceived time left
in life and emotion regulation strategies remains a fruitful area of research. Measuring time
perspective in addition to testing for the positivity effect in older adults would be beneficial in
increasing our understanding of why the positivity effect occurs.
Alternative theories suggest that the emotional-motivational changes occurring with
increasing age are the result of normal age-related changes in brain systems (Baltes & Baltes,
1990; Cacioppo, Berntson, Klein, & Poehlmann, 1998; Labouvie-Vief, 2003; Leigland et al.,
2004). However, some experimental studies have reported evidence inconsistent with structural
degradation as the sole cause for the positivity effect in older adults (Fung, Carstensen, & Lutz,
1999; Mather & Knight, 2005; for a review, see Samanez-Larkin & Carstensen, 2011). For
12

example, while age is highly correlated with future time perspective, younger individuals have
been shown to prioritize emotional goals when a constraint on their time left in life is apparent
(Carstensen & Fredrickson, 1998; Fung & Carstensen, 2006). In addition, when older adults
were asked to imagine a more expansive time perspective, they have been shown to look more
like typical younger adults in the context of goal prioritization (Fung et al., 1999). These findings
support the influence of time perspective in the shift in goal prioritization (for a review, see
Carstensen, 2006).
Having said this, the positivity effect is shown most often when cognitive control
mechanisms needed to implement emotion regulation goals are available (Knight et al., 2007;
Mather & Carstensen, 2005; Mather & Knight, 2005; Petrican, Moscovitch & Schimmack, 2008;
for a review, see Kryla-Lighthall & Mather 2009). Löckenhoff and Carstensen (2007) studied the
influence of goal prioritization on the positivity effect and found that the effect was eliminated
when older adults were asked to focus on accurately retaining information. These findings
indicate that older adults’ positivity effect is the result of goal -directed processes that help direct
attention and shape memory.
The Strength and Vulnerability Integration model (SAVI: Charles, 2010; Charles &
Luong, 2013) provides a framework to test when and why emotion regulation abilities improve
with increasing age. It combines the importance of future time perspective (life left to live) noted
from SST with the added value of experience garnered from accrued knowledge (life lived). This
theoretical model highlights strategies used, such as avoidance of emotionally distressing
interactions, to help maintain older adults’ emotional well -being. An explanation of when these
strategies no longer work and why is also provided within this framework. Specifically, this
model suggests that emotional vulnerabilities in later life occur when older adults are unable to
13

avoid and instead are subjected to high levels of sustained emotional arousal. Chronic and
uncontrollable stressors such as living with chronic pain and severe caregiver burden are
examples of this. Under such circumstances, the positivity effect is often attenuated and in some
cases, nullified or reversed.
Within the context of the present study, SST and SAVI would suggest that a positivity
effect in older adults’ reaction to an unfavorable choice outcome may be due to their more salient
goal of regulating emotions compared to younger adults. While we did not experimentally test
future time perspective in the present study, we were interested in how our participants’
perceived time left in life was associated with the way they expressed use of coping strategies in
reaction to the study scenario. Therefore, as a complement to the main analyses, we measured
participants’ future time perspective and examined its relation ship to the amount of problem-
focused and emotion-focused coping strategies verbalized.
Coping Style
Differences in coping style between younger and older adults is a well-researched topic
(Birditt & Fingerman, 2005; Blanchard-Fields, Stein, & Watson, 2004; Coats & Blanchard-
Fields, 2008). Coping has been defined as the thoughts and actions that manage stressful internal
or external demands (Lazarus & Folkman, 1984). According to the lifespan theory of control,
optimal development occurs with the ability to know when to focus on changing the environment
(primary control) and when to concentrate on modifying one’s internal state (secondary control)
when confronted with problematic situations (Heckhausen & Schulz, 1995). These definitions,
while broader and more focused on control, are similar to what Lazarus and his colleagues define
as problem-focused and emotion-focused coping (Folkman, Lazarus, Pimley, & Novacek, 1987).
The lifespan theory of control predicts that as adults grow older, they are likely to encounter
14

more situations that are out of their control (e.g. health problems), and therefore secondary
control processes (i.e. emotional coping) become increasingly important in maintaining relatively
stable perceptions of primary control (feelings of independence and ability to act on the
environment) (Heckhausen & Schulz, 1995). While theoretically different, predictions of
behavior from this theory are similar to those of SST, namely, that since the action of regulating
emotion is considered a secondary control process, emotion regulation skills should be used
proportionally more often than primary control processes (i.e. problem solving) in older adults
compared with younger adults when coping with an unpleasant situation.
In fact, there is strong evidence that older adults may be more likely to actively regulate
their emotions when situations become emotionally difficult compared to younger adults. For
example, older adults report more deliberate withdrawal from interpersonal conflict compared
with younger adults (Birditt, Fingerman, & Almeida, 2005). When already in a conflict situation,
unlike younger adults, who appear more inclined to use problem-focused strategies in both
emotional and non-emotional situations, older adults appear more flexible in the strategies they
use to deal with unpleasant situations (Blanchard-Fields, Jahnke, & Camp, 1995; Blanchard-
Fields, Mienaltowski, & Baldi, 2007). Specifically, older adults have been shown to use
relatively more problem-focused strategies when reacting to instrumental situations but more
emotion-focused strategies when reacting to interpersonal situations. Researchers have suggested
that this signifies older adults’ adaptability in their use of coping strategies to appropriately deal
with the situation at hand. However, in general older adults have reported to prefer using fewer
problem-focused strategies and more emotion-regulation strategies such as avoiding
confrontation and emotion suppression when possible, compared to proactive emotion regulation
strategies such as reflecting on and processing negative emotions (Blanchard-Fields et al., 2004).
15

These age differences have been hypothesized to be due to a shift in goal prioritization (Coats &
Blanchard-Fields, 2008).
Overall, there is a sizable literature supporting older adults’ flexibility in the use of
coping strategies as well as an increased use of emotion-focused strategies compared to younger
adults when dealing with highly emotional and interpersonal problems (Blanchard-Fields, 2007).
However, little research has been conducted to determine whether the adaptive ability that older
adults display in these situations extends to unpleasant situations in which they may feel
personally responsible. Given the literature showing a positivity effect in aging and its effect on
coping strategies, it is possible that the added feeling of responsibility that comes with having
made a bad decision will affect younger and older adults differently. Understanding age
differences in how individuals react and cope with the negative consequences of their decisions
would therefore be an important area of study to complement both the literature on the positivity
effect in aging as well as the decision-making literature.
Decision Making
Research suggests that when making a decision, older adults search for information to a
lesser extent than younger adults (Mata & Nunes, 2010; Yoon, Cole, & Lee, 2009), exhibit more
comprehension errors (Finucane et al., 2002), and are more likely to avoid making a choice
compared to younger adults (for a review, see Mather, 2006). Older adults have also shown a
preference for making fewer choices (Reed, Mikels, & Simon, 2008) and appear to value choice
less (Mikels, Reed, & Simon, 2009) than younger adults. Older adults also tend to value certainty
more heavily than younger adults when making choices based on gain/loss probabilities (Mather
et al., 2012). However, quality of choice (Mata & Nunes, 2010; Tanius et al., 2009) and the
ability to make decisions under uncertainty (Hosseini et al., 2010) do not appear to deteriorate
16

with age. In fact, some evidence even suggests that older adults are more skilled decision makers
than younger adults (Kim & Hasher, 2005).
Studies examining emotion-laden choices have found a positivity effect in older adults
when evaluating information seeking and decision making behaviors. For example, Löckenhoff
and Carstensen (2004) found that affective valence of material determined age differences in
how much information younger and older adults sought, in that older adults stated they were
more likely to ask for additional information if framed positively (gain frame) compared to
negatively (risk frame) whereas younger adults showed the opposite pattern. When asked to
make a choice between health plans, older adults were found to review a greater proportion of
positive to negative attributes compared to younger adults (Löckenhoff & Carstensen, 2007).
Older adults have also been shown to be more choice-supportive than younger adults when asked
to recall decisions they have made (Mather & Johnson, 2000). That is, when asked to attribute
features to provided options (including the chosen option) after some time delay, older adults
were more likely to attribute positive over negative features to their chosen option than were
younger adults. This pattern of results suggests that older adults more actively focus their
cognitive resources on positive emotional goals compared with younger adults (Mather &
Knight, 2005).
While there is a growing literature on age differences in decision making, the majority of
research to this point has focused on the process of making the decision and the quality of the
decision made. Few studies within the aging literature have examined how the decision maker
feels after the outcome is revealed. In other words, little is known about how younger and older
adults cognitively appraise and emotionally react to a situation after the decision is made and the
consequences are revealed. This component of decision making is important to study because the
17

way individuals feel about the outcome of a decision has been shown to influence future decision
making behavior (Ratner & Herbst, 2005). Furthermore, playing a significant role in the
occurrence or severity of an unfavorable situation may increase personal responsibility for the
negative situation, which can increase negative affect (Burger, 1989) and more specific emotions
such as regret and self-blame. Given the literature on a positivity effect in older adults, it is
possible that making a decision that results in a negative outcome will affect younger and older
adults differently. Research on age and emotional reactions to the consequences of their
decisions would therefore be important in understanding these distinctions.
One study measuring age differences in satisfaction asked older and younger adults to
choose among four gift options (pen, mug, keychain flashlight, whiteboard with marker) in a lab
setting (Kim, Healey, Goldstein, Hasher, & Wiprzycka, 2008). In one condition, participants
were asked to write down what they liked and disliked about each choice (to engage cognitive
control processes) before choosing their preferred gift. In the control condition, participants
made a choice without writing down their evaluations. All participants were asked to rate how
much they liked their chosen product at the moment, and then were asked again approximately
two weeks later. Results for both initial and delayed ratings of satisfaction indicated that older
adults were more satisfied with their choice than younger adults were, but only in the condition
where they were asked to evaluate their choices on paper. Breaking down how younger and older
adults evaluated their choices showed that older adults listed more positive and fewer negative
attributes than younger adults did. These results are consistent with previous studies showing a
positivity effect in older adults for attention and memory (Mather & Carstensen, 2005) and
highlight the importance of cognitive engagement in the regulation of emotions. The present
18

study expands on this area of research by examining the thoughts and emotions individuals have
in response to an unfavorable decision outcome.
Regret
One of the most intensively studied “post -decision” emotions is the feeling and
expression of regret. For years, philosophers, economists, and psychologists have examined the
key components of regret and how regret, as well as anticipation of regret, can affect future
behavior. This is an important area of study since level of regret has been shown to associate
negatively with subjective well-being and self-reported physical health in both younger and older
adults (Dijkstra & Barelds, 2008; Jokisaari, 2003). Because anticipated regret has also been
shown to play a role in the decision-making process (Zeelenberg, 1999), understanding the
cognitive appraisals that occur when regret is felt is significant in gaining a more comprehensive
perspective on the decision making process.
In general, regret is defined as a complex negative emotion, one that requires both an
unpleasant cognitive and emotional state (Landman, 1993). Comparing a chosen outcome with
alternatives that are perceived to have a better outcome has been shown to lead to regret
(Connolly & Zeelenberg, 2002). These mental images of possible choice alternatives and their
consequences are known as counterfactuals (Roese, 1997). In general, degree of regret has been
associated with factors such as controllability of the antecedent, outcome closeness, ability to
undo the decision, and the normality of the situation itself (Roese, 1997). Overall, the literature
in this field indicates that the perception of a decision’s outcome is a key factor in eliciting
negative emotions such as regret, which may have an impact on overall life satisfaction.
The experience of regret and counterfactual thinking in older adults has been examined to
a much lesser extent than in younger adults. In general, older adults have self-reported
19

experiencing fewer regrets than younger adults (Timmer, Westerhoff & Dittmann-Kohli, 2005);
however it is unclear whether this age difference is due primarily to older adults’ avoidance of
potentially negative consequences (antecedent-focused coping) or to differences in reaction to an
unfavorable outcome (response-focused coping). Furthermore, research on regret with older
adults has focused primarily on severe and lifelong regrets (Wrosch, Bauer & Scheier, 2005;
Wrosch & Heckhausen, 2002), with less emphasis on everyday regret. Some research reveals a
significant negative association between frequency of reported everyday regret and age
(Bjalkebring, Vastfjall, & Johansson, 2013). Everyday regret was self-reported to decrease in
both intensity and duration as age increased, while life regret was associated with the opposite
pattern (V stf ll, eters, lkebring, ). owever, given that older adults have been
shown to avoid negative situations more than younger adults (Mather, 2006), it is likely that the
situations younger and older adults in that study encountered were very different. Therefore,
research on how younger and older adults react to the same situation is of importance in order to
control for variations in environment. The present study sought to clarify whether age differences
are associated with differences in response-focused coping strategies by limiting people’s ability
to use antecedent-coping strategies to control their environment. The use of a particular think-
aloud cognitive assessment paradigm enabled us to do this.
Overall, compared to the literature on the processes of making a decision, individuals’
thoughts and feelings towards the outcome of their everyday decisions has been largely
understudied, particularly with an older adult population. The current study tested age
differences in participants’ level of regret of a choice that led them to an unfavorable outcome.
20

The Present Study
The present study examined age differences in reaction to the consequences of a bad
decision outcome, and how those reactions differ when compared to the same situation without
the choice component. The Articulated Thoughts in Simulated Situations (ATSS) paradigm
(Davison, Robins, & Johnson, 1983) was used to capture expressions of thoughts and emotions
as well as coping in experimenter-controlled situations and in a virtually online fashion.
Participants in one condition (choice condition) engaged in a hypothetical scenario that took
place in an airport, and were asked to choose between two different flights (their existing flight
and an alternative flight leaving earlier) that were going to the same destination. Participants
were asked to imagine that it was important that they get to their destination in a timely manner.
Regardless of their choice, but unbeknownst to the participant, the scenario results in a negative
experience, including their flight becoming delayed while the flight they did not choose leaves
on schedule. Participants in the other condition (no choice control condition) were taken through
the same scenario; however these participants were not given a choice to switch to an alternative
flight at the beginning (see Table 1 for breakdown of groups). Having younger and older adults
experience the same scenario allowed us to more accurately compare and assess age differences
in negative reactions and coping strategies. In other words, we sought to limit the variability in
environmental stimuli provided to younger and older adults. In addition, while questionnaires
often require retrospective reporting and do not control for varying environments between
subjects, the present study relies much less on memory and instead, asks participants to verbalize
whatever they are thinking and feeling at that moment. This is the central characteristic of the
ATSS paradigm.
21

Our first aim was to test for age differences in cognitive and emotional reactions to
unfavorable situations. Based on previous studies showing an age-related positivity effect in
attention and memory of older adults, we hypothesized that (1) younger adults would express
more regret, guilt, anger (at others and at themselves), and more anxiety during the scenarios
than our sample of older adults in both the choice and no choice conditions. No prediction was
made regarding age differences in expression of sadness, as previous studies reveal mixed
findings (Charles & Carstensen, 2008; Charles, Carstensen, & McFall, 2001).
Our second aim was to test for differences in cognitive and emotional responses to a
condition in which participants experience an unpleasant situation to which they have not
contributed (no choice control condition) compared to one in which participants make a choice
that leads to the same unpleasant outcome (choice condition). Expanding on the decision making
and regret literature, we hypothesized that (2) participants in the choice condition (condition
where they were asked to choose between their existing flight and an alternative flight) would
express more regret, guilt, and anger at themselves than participants in the no choice condition.
While data were collected and analyzed for exploratory purposes, no specific hypotheses were
made for condition differences in expressions of anger at others, anxiety, or sadness as they are
emotions that are less theoretically tied to choice and decision-making. Given that participants in
the choice condition were explicitly asked to make a choice that led them to an unfavorable
outcome, we expected participants in this condition to feel more responsible for the outcome in
comparison to participants in the no choice condition. This expectation was tested by analyzing
differences in self-reported feelings of responsibility by condition (based on a question asked
after the scenario) and used as a manipulation check measure.
22

Our third aim was to test whether younger and older adults were differentially affected by
the choice condition when compared to the no choice condition. Specifically, we were interested
in whether age differences would be more prominent in a situation where choice was a salient
factor in the scenario. ased on evidence indicating older adults’ greater focus on emotion
regulation and their positivity effect compared to younger adults, we hypothesized an (3)
interaction between age and choice condition for expressions of regret, guilt, and anger at the self
(the most prominent post-decision emotions). Specifically, we anticipated that compared to older
adults, younger adults would be more negatively affected by the salient choice component. Age
differences in the choice condition were hypothesized to be even larger than age differences
found in the control condition. No interaction hypotheses were made for expressions of anger at
others, anxiety, or sadness, as we believed these emotions to be less affected by choice saliency.
Our fourth aim was to test for age differences in coping style when reacting to
unfavorable situations. We hypothesized that (4) older adults would use more emotion-focused
than problem-focused coping strategies compared to younger adults, who may be inclined to use
more problem-focused coping strategies in general. No specific prediction was made regarding
the interaction between age and condition for coping style; however exploratory analyses were
conducted to examine the potential relationship.
In addition to these hypotheses, we were interested in examining whether future-time
perspective was related to the amount of emotion regulation strategies expressed within our
study. According to SST, a more limited time-perspective influences goal priorities, shifting
them from more future-oriented to more present-oriented, which included maintaining emotional
well-being in the moment. Exploratory analyses were conducted to examine whether future-time
23

perspective was associated with participants’ usage of problem -focused and emotion-focused
coping language.
The present study expands upon the current literature by testing whether the positivity
effect extends to post-decision making thoughts and emotional reactions. Using a think-aloud
assessment method to test these differences also extends the current literature in this field by
allowing for near-immediate verbal expression of emotional statements while in a controlled
setting. The controlled environment was an important feature within this study, as younger and
older adults’ everyday contextual environments can vastly differ, which can in itself affect the
amount and intensity of negative emotions experienced. We were particularly interested in how
older and younger adults react to the same negative scenarios, and the ATSS method permitted
this examination.
Method
Sample and Participant Selection
One hundred and seven younger and 81 older adults were recruited to take part in this
study. Participants had normal or corrected to normal vision and hearing, no reported diagnosis
of a cognitive disorder, and were able to come to the USC campus. Younger adults ages 18-25
years old and older adults ages 65 and over were recruited. Younger adults were undergraduate
students attending the University of Southern California and were offered experimental credit
upon completing the study. Older adults were volunteers recruited through the USC Davis
School of Gerontology as well as from the greater LA area. Older adults were offered $15
compensation in exchange for their participation. A total of 5 younger and 2 older adults were
excluded from analyses due to technical issues (e.g., inaudible audio recordings of responses,
technical issues resulting in unsaved data). An additional two participants were excluded (1
24

younger, 1 older) based on their low performance on a screening measure of cognitive
impairment (scores less than 26 out of 30 on the MMSE). Lastly, 1 older participant was
excluded for reportedly being younger than 65 years old. Therefore, 101 younger adult
participants between the ages of 18 and 24 (M = 19.76, SD = 1.28), and 77 older adult
participants between the ages of 65 and 87 (M = 71.73, SD = 4.81) were included in the analyses.
See Table 2 for demographic information.
Out of the 101 younger adults, 34 were Caucasian, 9 were African American, 11 were
Latino, 4 were Asian, and 6 were indicated as “other”. Out of the 77 older adults, 5 were
Caucasian, 11 were African American, 7 were Latino, 5 were Asian, and 4 were indicated as
“other”. Older adults in our sample were more educated than younger adults in our sample, an
expected finding given that our younger adults were still attending college and had not yet
received their degree (see Table 2). Younger adults also rated their overall health to be higher
than older adults’ ratings. Younger adults performed better on a cognitive screening measure of
general cognitive functioning (MMSE) and immediate verbal recall (Digit-Span) than older
adults; however no significant age differences were found with regard to baseline verbal
functioning (WTAR). Consistent with the literature (e.g., Flint, 1994; Hasin, Goodwin, Stinson,
& Grant, 2005), baseline measures of mood revealed that older adults reported fewer depressive
symptoms (CES-D) and less anxiety (STAI-T) than younger adults.
Procedure
The experiment was conducted within a 70 minute session, and all participants were
scheduled to participate at their preferred time of day. After reviewing the information sheet and
agreeing to participate in the study, participants completed a basic demographics questionnaire
that included a measure of self-reported overall health (“how would you rate yo ur overall health”
25

– see Table 2 for results). Questions related to flying on a plane were also asked to determine
whether there would be age differences in general attitudes towards the airport scenario as well
as recent experiences flying on a plane. The Positive and Negative Affective Schedule (PANAS),
a self-report measure of mood state (Watson, Clark, & Tellegen, 1988), was then given to
participants to fill-out (see Appendix A for flow chart). Participants were subsequently asked to
complete the Center for Epidemiological Studies – Depression scale (CES-D; Radloff, 1977), the
Spielberger State-Trait Anxiety Inventory – Trait (STAI-T; Spielberger, Gorsuch, & Lushene,
1970), and the Future Time Perspective scale (FTP; Carstensen & Lang, 1996). Next, the Mini-
Mental State Exam (MMSE), a screening measure of cognitive impairment, was administered,
followed by a test of immediate verbal recall (WAIS - Digit Span; Wechsler, 1981) and a general
measure of pre-morbid verbal functioning (Weschler Test of Adult Reading –WTAR; Wechsler,
1997a).
Upon completion of the aforementioned measures and before the ATSS scenario,
participants were asked to describe an important and meaningful event that they would like to
attend (see Appendix B). In addition, research assistants asked participants to think of two people
with whom they could imagine going to the event and were subsequently asked to rate how
disappointed they would be if they missed that event, on a scale from 1 (not at all disappointed)
to 9 (extremely disappointed). articipants who rated their disappointment level as a “4” or lower
were asked to think of another event that would be more disappointing to miss. They were then
asked to re-rate their disappointment level with the new event (this was necessary only once).
The event that participants rated as the more disappointing-to-miss was chosen as the event to
imagine during the ATSS scenario.
26

Participants were seated in front of a computer and presented with instructions on the
ATSS procedure in both written and audio formats (Davison, Vogel, & Coffman, 1997). See
Appendix C for instructions and Appendix D for scenario scripts. All participants were led
through a neutral practice scenario (imagining themselves at the supermarket) in order to
familiarize themselves with the response format. Examiners provided feedback to participants
after the practice scenario and addressed any questions or concerns they had at the time. By the
end of the practice scenario, participants acknowledged that they understood the procedure and
were able to verbalize their thoughts and feelings as per the instructions. Each participant was
randomly assigned to one of two conditions: the choice scenario or the no choice (control)
scenario. Examiners left the room while participants listened and responded to the audio-
recorded ATSS scenario. In addition to verbalizing their thoughts and feelings during each 30-
second pause, participants were asked to rate their general mood and level of regret on a Likert
scale from 1 – 9 after each 30-second pause, starting from the segment after participants made a
choice, or in the corresponding segment for the control condition. When the scenario concluded,
participants opened the door and indicated to the examiner that they had completed the ATSS
procedure.
Participants then completed a second PANAS questionnaire to assess post-scenario mood
state. A post-scenario questionnaire was also given, which included an assessment of feelings of
responsibility toward the unfavorable situation, as well as a manipulation check questionnaire
assessing the believability of the scenario. Lastly, participants were debriefed and examiners
addressed any questions or concerns regarding the experiment.
27

Measures of Emotional Response
ATSS. The Articulated Thoughts in Simulated Situations (ATSS) paradigm is a think-
aloud approach to cognitive assessment designed to capture the immediate thoughts and feelings
of participants in experimenter-controlled situations (Davison, Robins, & Johnson, 1983). The
ATSS paradigm allows for an unstructured response format and great flexibility in the creation
and delivery of the simulated situations. Situations and coding schemes have been tailored to
study a wide selection of contexts and populations, including smoking cessation, anger, anxiety,
depression, and marital conflict. Age-differences in response to unpleasant situations have also
been studied with the ATSS paradigm (Charles & Carstensen, 2008). For a recent conceptual and
empirical review, see Zanov & Davison (2010). The ATSS paradigm was chosen as a way to
assess older and younger adults’ cognitive and emotional responses to the same controlled
situation in near-real time, without having to rely on retrospective reporting.
For the current study, all participants engaged in a hypothetical scenario that takes place
at an airport. Both ATSS scenarios (one for the choice condition and one for the no-choice
“control” condition) used in the present study involved a delayed plane flight and were nearly
identical in nature except for whether the respondent believed he/she had a choice in changing
flights with ensuing negative consequences. Feedback from both a pilot study and an older adult
focus group supported that this type of situation is realistic and relevant for both younger and
older adults. In addition, analysis on post-ATSS measures did not reveal significant age
differences in how realistic and easily imaginable the scenario was to them, F(1, 176) = .68, p =
.41.
In the choice scenario participants imagined themselves at an airport with two
companions, ready to fly to an important event which they did not want to miss. Early on in the
28

scenario, participants in this condition were given a choice between staying on their existing
flight boarding in 60 minutes and switching to an alternative flight boarding in 15 minutes.
Regardless of their choice, participants experienced an unfavorable outcome within the scenario,
namely an overcrowded waiting area, rude gate attendants, and a delayed flight that ultimately
resulted in their missing the event. In addition to these frustrating factors, the flight option that
they did not choose was presented throughout the scenario as the option with the more desirable
outcome (counterfactual).
The no-choice control condition unfolded in the same way as the choice condition;
however participants in the control condition were not presented with a choice and therefore
were not asked to make a decision regarding their flight (see Appendix D for full scripts). Aside
from this difference, the main events that occurred throughout the choice scenario (e.g. rude gate
attendants) also occurred in the same manner throughout the no-choice control scenario. Thus,
using the ATSS, it was possible to manipulate whether participants did or did not make a choice
that resulted in a negative outcome.
ATSS coded response variables. For the present study, open-ended responses were coded
for the following emotions: 1) regret, 2) guilt, 3) anger at others, 4) anger at the self, 5) anxiety,
and 6) sadness. Relative problem-focused coping and emotion-focused coping statements were
also coded for and analyzed. As with all ATSS procedures, each segment of the presented
scenario was followed by a 30-second pause allowing participants to verbally express the
thoughts and feelings they were experiencing in reaction to the previous segment. Our scenario
included 11 30 to 45-second segments and 11 30-second pauses in which participants were asked
to verbally express themselves. Given that the choice point differentiating the two conditions was
29

introduced after the second segment, only the 9 segments following the decision point were
analyzed.
ATSS coding procedure. Audio recordings of participants’ articulated thoughts and
feelings were transcribed and coded by four trained research assistants (coders). Each participant
was coded by two coders, with the combination of coder pairs distributed equally throughout our
sample. Therefore, all four coders were assigned to code about half of our sample from the
transcription only. Coders were blind to the study’s main hypotheses. Respons es were coded by
segment for the presence and intensity of the response variables. Each of these categories was
coded on a four-point Likert scale: 0 – not at all (complete absence of code); 1 – slightly /
somewhat (low presence of code); 2 – moderately (moderate presence of code); 3 – very (high
presence of code). See Appendix E for coding procedure. For each category, the average of the
two coders’ ratings was taken for each segment and averaged across the nine segments to
produce a total category score for each participant. For the dependent variable coping style, each
participant’s code for problem -focused coping was subtracted by their code for emotion-focused
coping within each segment, thus creating a difference score indicating the relative usage of
problem-focused to emotion-focused coping.
We were conscious of the potential difficulty that could occur with coding responses
when listening to participants’ audio recordings. While listening to audio recordings of
participants’ responses has typically been thought of as useful for coding nonverbal aspects of
emotional expression, coders may be able to tell from the voices whether the participant is
younger or older. In an attempt to keep raters blind to the participants’ age, think-aloud
responses were coded using the transcribed responses only (without audio) and these codes were
used to test our main hypotheses. We believe this method allows for a “fairer” comparison
30

between age groups, as younger and older adults may have different vocal characteristics that
could bias coding of emotions. Coding ATSS responses using transcriptions (without audio) has
been employed in a previous study for similar reasons (Charles & Carstensen, 2008).
Intraclass correlations (ICCs) were computed as an estimate of the interrater reliability
between coders for each of the ATSS coded response variables. A two-way random effects
model with consistency agreement (average measures) was considered to be most appropriate to
calculate these estimates, as each participant’s resp onses were coded by two raters but not all
participants were coded by the same two raters. Coders’ ratings of each of the response variables
for each participant were averaged across the nine segments to create a new “average” rating for
each variable in order to calculate the appropriate ICC based on the primary analyses. ICC’s
ranged from .74 - .93, indicating that the seven ATSS coded response variables were rated in a
reliable manner and represent reliable constructs. Therefore, all coded variables were deemed
appropriate for further analysis.
Self-reported regret and mood. Along with free-flowing unconstrained verbal
responses to the ATSS scenario, participants were asked to rate their general mood on a scale
from 1 (most pleasant) – 9 (most unpleasant) after each 30-second pause of each segment.
Ratings of regret level were also collected on a 1 (no regret) – 9 (extreme regret) scale starting
from the segment after the participant made a choice (or in the corresponding segment for the
control condition). This quantitative measure taken during the ATSS procedure is a unique
addition to the typical ATSS method, which primarily uses coded qualitative responses to assess
variables of interest. Participants read the questions on the computer screen and indicated their
response using the computer mouse. The average response ratings across all segments for regret
and general mood were calculated separately for analysis.
31

Results
Differences between younger and older adults were assessed for the overall presence and
intensity of cognitive-emotional responses and coping style to the ATSS scenario. Baseline
perceptions of the airport scenario were analyzed between groups using one way analysis of
variance (ANOVA). Post-ATSS measures as well as manipulation check measures were
analyzed using one-way ANOVAs. Primary analyses tested age differences, differences in
experimental condition and their interaction regarding the level of negative emotional reactions
and coping strategies employed. ATSS variables were analyzed using either a 2 x 2 between
subjects ANOVA or a 2 x 2 logistic regression. A Holms-Bonferroni statistical adjustment was
used on hypothesized analyses (for ATSS coded emotions) to account for inflated Type 1 error
(Holm, 1979). The Holms-Bonferroni correction is similar to the Bonferroni correction in that it
takes into account the number of analyses conducted and therefore limits the possibility of
finding a significant effect by chance. However, it is a more statistically powerful technique than
the family-wise Bonferroni correction because it readjusts the corrected p-value after each
hypothesis is tested and the null hypothesis is rejected. All unadjusted and adjusted p-values are
reported in Table 6. To prevent redundancy, adjusted p-values are presented in text only when
significant unadjusted p-values are found (when significant results are not found, adjusted p-
values will also be non-significant).
As in prior studies comparing younger and older adults (Flint, 1994; Hasin et al., 2005),
age differences were found in baseline depressive symptoms as well as symptoms of anxiety.
Therefore, to adjust for these baseline differences, CES-D and STAI-T scores were used as
covariates within the main analyses. For thoroughness, primary findings from this study are
presented two ways: structured by dependent variable/emotions (to lay out analytic approach)
32

and then by hypothesis (to summarize combined analyses by our topics of interest). Results by
hypothesis are presented with p-values and effect sizes.
Perceptions Regarding the Airport Scenario
To assess whether younger and older adults were similarly comfortable with the
experimental scenario itself (being at an airport, going to fly on an airplane), we collected
information on how anxious participants were about flying on an airplane as well as the last time
they had flown on an airplane. No significant age differences in anxiety towards flying (1 = not
at all anxious; 5 = extremely anxious), or when they had last flown on an airplane (1 = < 1month
ago; 5 = > 5 years ago) were found (See Table 3). In addition, no significant age differences
were found regarding their level of anticipated disappointment in missing their described event
(1 = not at all disappointed; 7 = extremely disappointed), indicating that younger and older adults
viewed their event as similarly important and disappointing to miss. This finding was expected,
given that participants who initially rated this question lower than 4 out of 7 were prompted to
think of another event that would induce a higher level of disappointment if missed.
Furthermore, no significant differences were found between the choice and no choice conditions
on these variables (p’s > .05), indicating that on average, participants in each experimental group
were similarly comfortable with being at an airport and flying on a plane.
Manipulation and Scenario Check
articipants’ mood state was tested both before the ATSS scenario and again immediately
after engaging in the ATSS scenario using the PANAS measure. Paired t-test analyses revealed
that on average, participants endorsed significantly more negative mood and significantly less
positive mood after the scenario compared to their baseline measures of mood (p’s < . 5) .
Specifically, younger adults reported feeling more negative, t(100) = -10.53, p < .001, and less
33

positive t(100) = 2.81, p = .006 after the scenario compared to baseline. Older adults were also
reporting more negative mood, t(76) = -4.47, p < .001, and less positive mood, t(76) = 2.09, p =
.04, after the scenario compared to baseline. These results suggest that the scenarios were able to
successfully elicit negative affect (and diminish positive affect) in our participants. Furthermore,
participants in the choice condition reported feeling significantly more responsible for them and
their friends missing the described event than participants in the no choice control condition (see
Table 4). These results support the efficacy of the experimental condition, both with regard to the
ability to induce negative affect as well as the effectiveness of the experimental manipulation.
At the end of the study, participants were asked additional questions related to the
external validity of the scenarios. Overall, the scenarios used in our study were rated as highly
realistic and easily imaginable (0 = not at all realistic to 10 completely realistic; M = 8.13; SD =
1.80). Participants also reported that they believed their responses to the scenarios were highly
similar to what they thought they would have been in a real life situation (0 = not at all similar to
10 extremely similar; M = 8.48; SD = 1.59). Additionally, no significant differences were found
in the reported realness of the two scenarios (choice and control), p = .58, or with regard to how
similar their responses would have been in real life for the two scenarios, p = .74 (see Table 4).
Furthermore, younger and older adults also reported the scenarios as similarly realistic and easily
imaginable, p = .41, and both age groups reported that their responses were similar to what they
would have been in a real life situation, p = .16 (see Table 3).
ATSS Coded Variables
The current study examined both age-related differences and experimentally manipulated
effects of choice on participants’ responses, measured by the following ATSS coded variables:
regret, guilt, anger at others, anger at self, anxiety, sadness, and P-E coping (problem-focused –
34

emotion-focused). Results are presented in this section by coded dependent variable, as analyses
were conducted separately for each ATSS coded variable.
Distinctiveness of measures. To assess inter-item correlations among the ATSS coded
measures, earson’s bivariate correlations were conducted among the seven codes of interest. A
correlational value greater than .70 was used to screen for variables to be examined further for
multicollinearity. Correlations among dependent variables ranged from r = -.03 to .42, which
represent small to moderate correlations. Given that the correlations were not greater than .70,
we did not suspect that the ATSS coded measures were at risk of multicollinearity and therefore
all seven coded measures were subjected to further independent analysis.
Checking for outliers. Distributions of the residuals for each of the seven coded
variables were inspected for extreme outliers. In a process analogous to winsorization, values of
>3 SD were adjusted to the closest non-outlier response within their same age group. A close
inspection revealed two outliers for the coded measure of regret and 3 outliers for the coded
measure of sadness. Results for these measures are reported for the outlier-adjusted
distributions
1
. Inspection of residual distributions for all other coded measures did not reveal
extreme outliers and therefore these distributions were left unadjusted for analyses.
Normality of residuals. A visual inspection of the residual plots was conducted to
determine residual normality and model fit for each of the ATSS coded measures. Assessing this
factor enabled us to determine the most appropriate type of analysis that should be implemented
for each distribution. Clear patterns deviating from the normal probability plot and an abnormal

1
Analyses were also run on unadjusted distributions for both regret and sadness. Results for
regret were consistent with analyses conducted on adjusted distribution (with and without
covariates included in the model). Results for sadness were consistent with analyses conducted
on adjusted distribution; however when covariates (baseline depression, anxiety, word count)
were included, age differences in sadness did not reach statistical significance (p = .09).

35

distribution shown on the histogram of residuals represent a violation of the assumption of
normality.
A careful inspection of the residuals showed that the coded ATSS measures of guilt and
anger at self were non-normally distributed. Closer examination of these distributions revealed
that a significant number of participants were rated as not experiencing any guilt or anger at
themselves throughout the entire scenario. Specifically, 45% of younger and 70% of older
participants were rated a “ ” on expressions of guilt across all segments, and 5 % of younger
and 7 % of older participants were rated a “ ” on expressions of anger at themselves across all
segments. As more than 50% of participants were rated as expressing no guilt or anger at
themselves, we determined that the most appropriate way to analyze these two variables was to
convert the responses into binary outcomes with 0 representing an absence of that variable (e.g.
guilt) and 1 representing presence of that variable. Therefore, hypotheses related to guilt and
anger at the self were analyzed using logistic regression. A visual inspection of the outlier-
adjusted distributions indicated that the remaining 5 coded measures (regret, anger at others,
anger at self, anxiety, coping) were approximately normal and given that the F-test in ANOVA is
very robust against non-normal data, we decided to analyze the remaining 5 coded measures
using two-way between group analyses.
Word count as a covariate. The number of words verbalized by participants in each
coded segment was thought to be a potential confounding variable in our study since, on average,
older adults speak more slowly than younger adults (Smith, Wasowicz, & Preston, 1987).
Specifically, older adults in our sample used significantly fewer words overall than younger
adults, F(1, 176) = 20.85, p < .001. Given this finding, we believed it was possible that the
amount of words verbalized may be a contributing factor in coding for the frequency and
36

intensity of certain ATSS variables. For example, if younger adults used more words, it is
possible if not likely that they talked faster than older adults in general and therefore had more
opportunity to express feelings of regret or anger or any other emotion they were feeling at the
moment. earson’s bivariate correlations were conducted to determi ne to what extent word count
was associated with ratings of the seven ATSS coded variables. Results indicated that word
count was statistically significantly correlated with ratings of anger at others (r = .26, p = <.001),
anxiety (r = .29, p < .001), and sadness (outlier-adjusted distribution; r = .15, p = .05). Coded
measures of regret, guilt, anger at the self, and P-E coping were not significantly correlated with
word count (p’s > .05). For all variables significantly correlated with word count, the higher the
word count, the higher the ATSS coded rating of that measure (e.g., more verbalization was
associated with higher ratings of anger at others). Given these findings, analyses related to the
ATSS coded variables anger at others, anxiety, and sadness are reported with word count as a
covariate. Holms-Bonferroni corrected p-values were based on the analyses with the covariates
included.
Examining homogeneity of variance. Homogeneity of variance assumption was
examined for each of the continuous dependent variables for both groups using Levene’s test .
Although considered to be conservative, this test found that measures of anger at others, anxiety,
sadness, and coping were not in violation of this assumption (p’s > . 5). Analysis of the regret
variable showed a significant Levene’s test result ( p = .002); however the ratio of the largest
group variance to the smallest group variance was less than 10 (.06 / .02 = 3), therefore
differences in variances across groups were not expected to significantly affect results
(Tabachnick & Fidell, 2007). The omnibus F-test is also very robust against heterogeneity of
37

variances. Taken together, regret, anger at others, anxiety, sadness, and coping measures were
considered appropriate for analysis using ANOVA.
Emotion variables. The following six ATSS coded emotions were analyzed using either
a between-sub ects ANOVA or logistic regression depending on the individual measure’s
distribution. To account for multiple comparisons, the Holms-Bonferroni procedure was
implemented. Results are presented first with their original p-value and then with the adjusted p-
value when the original p-value is significant. See Table 5 for means and standard deviations for
expressed emotions by age group and choice condition.
Regret. A 2 x 2 between groups ANOVA was conducted to determine the extent to which
age and condition affected ratings of verbalized regret. Baseline level of depression and anxiety
were included as covariates. Consistent with hypothesis 1, results indicated a significant main
effect of age, F(1, 172) = 11.15, p = .001, 
p
2
= .06 in that younger adults expressed more regret
than older adults overall. Furthermore, this effect also remained significant after adjusting for
multiple comparisons (p = .004). A significant main effect of condition was also found, which
was consistent with hypothesis 2, and revealed that participants in the choice condition expressed
more regret than those in the no choice control condition, F (1, 172) = 4.94, p = .03, 
p
2
= .03.
This effect remained significant after adjusting for multiple comparisons (p = .03). While a
significant age by condition interaction was hypothesized for regret (hypothesis 3), results did
not reveal the interaction to be statistically significant, F(1, 172) = .07, p = .79, 
p
2
= .00. See
Figure 1 for a graphical representation of these results and Table 6 for the results of the ATSS
coded variables grouped by hypothesis.
Guilt. A 2 x 2 logistic regression was conducted to determine the impact of age,
condition, and their interaction on whether participants expressed guilt during the scenario (see
38

Table 5 for percentages by age group and condition). The initial model including only covariates
(CES-D, STAI-T, that is, depression and anxiety) was statistically significant, χ
2
(2, N = 178) =
6.47, p = .04, correctly classifying 57.9% of cases and explaining between 3.6% (Cox and Snell
R square) and 4.8% (Nagelkerke R Square) of the variance in expressed guilt. Both baseline
measures of depression (Exp(B) = .94, 95% CI: .88-.999, p = .047) and anxiety (Exp(B) = 1.07,
95% CI: 1.01 – 1.13, p = .01) were found to be significant covariates in the initial model. Adding
our main independent variables (age and condition) significantly improved model fit, χ
2
(4, N =
178) = 53.30, p < .001, correctly classifying 73.6% of cases and explaining between 25.9% (Cox
and Snell R square) and 34.7% (Nagelkerke R Square) of the variance in expressed guilt.
Baseline depression remained a significant covariate within this model (Exp(B) = .92, 95% CI:
.86-.995, p = .04; however baseline anxiety was no longer a significant covariate, (Exp(B) =
1.06, 95% CI: .99 – 1.13, p = .08).
Consistent with our hypotheses, both age (hypothesis 1) group (Exp(B) = 3.29, 95% CI:
1.49 – 7.23, p = .003) and condition (hypothesis 2) (Exp(B) = 8.71, 95% CI: 4.17 – 18.23, p <
.001) significantly contributed to model fit after taking into account baseline scores of depression
and anxiety. Both predictors remained significant after adjusting for multiple comparisons (p =
.009 and .003 respectively). These results indicate that younger adults were 3.3 times more likely
to express feelings of guilt than older adults and participants in the choice condition were 8.71
times more likely to express feelings of guilt than participants in the no choice control condition.
Hypothesis 3 was not supported for expressed guilt, as adding the interaction variable did not
improve model fit and was not significant (Exp(B) = .79, 95% CI: .17 – 3.62, p = .76).
Anger at others. A 2 x 2 between subjects ANOVA was conducted to determine the
extent to which age and condition affected ratings of verbalized anger at others. Baseline levels
39

of depression and anxiety, as well as word count were included as covariates. Consistent with
hypothesis 1, significant age differences were found such that younger adults expressed more
anger at others in the scenario than older adults, F(1, 171) = 12.79, p < .001, 
p
2
= .07. Age
differences remained significant after adjusting for multiple comparisons (p = .005). Main effects
of condition were not significant, indicating that participants in the choice condition did not
express significantly more or less anger at others compared to those in the control condition, F(1,
171) = 2.34, p = .13, 
p
2
= .01. A significant age by condition interaction was not found, F(1,
171) = .38, p = .54, 
p
2
= .00. Thus, older adults expressed less anger overall compared to
younger adults, but the difference did not depend on the type of condition they were in. See
Figure 2 for a graphical representation of results.
Anger at self. A 2 x 2 logistic regression was conducted to determine the impact of age
and condition on whether participants expressed anger at themselves during the scenario. The
initial model including only covariates (CES-D, STAI-T) was marginally significant, χ
2
(2, N =
178) = 5.64, p = .06, correctly classifying 62.9% of cases and explaining between 3.1% (Cox and
Snell R square) and 4.2% (Nagelkerke R Square) of the variance in expressed guilt. Neither
baseline depression (Exp(B) = .99, 95% CI: .93-1.05, p = .67) nor anxiety (Exp(B) = 1.05, 95%
CI: .99 – 1.10, p = .09) was found to be significant. Adding our main independent variables (age
and condition) significantly improved model fit, χ
2
(4, N = 178) = 53.30, p < .001, correctly
classifying 65.7% of cases and explaining between 10.7% (Cox and Snell R square) and 14.4%
(Nagelkerke R Square) of the variance in expressed anger at the self.
Consistent with hypotheses 1 and 2, both age group (Exp(B) = 2.17, 95% CI: 1.17 – 1.08,
p = .03) and condition (Exp(B) = 2.75, 95% CI: 1.45 – 5.22, p = .002) significantly contributed
to model fit after taking into account baseline scores of depression and anxiety. When adjusting
40

for multiple comparisons, both age (p = .03) and condition (p = .004) remained statistically
significant predictors. These results indicate that younger adults were 2.17 times more likely to
express anger at the self than older adults and participants in the choice condition were 2.75
times more likely to express anger at the self than participants in the no choice control condition.
Adding the interaction variable did not improve model fit and was not significant (Exp(B) =
1.25, 95% CI: .34 – 4.65, p = .74).
Anxiety. A 2 x 2 between groups ANOVA was conducted to determine the extent to
which age and condition were associated with ratings of verbalized feelings of anxiety. Baseline
levels of depression and anxiety, as well as word count were included as covariates. While not
hypothesized, a significant age x choice interaction was found F(1, 171) = 4.09, p = .045, 
p
2
=
.02. Consistent with hypothesis 1, younger adults were found to express more anxiety than older
adults, F(1, 171) = 5.87, p = .02, 
p
2
= .03. Age differences remained significant after adjusting
for multiple comparisons (p = .03). No prediction was made with regard to statistically
significant differences in expressed anxiety by condition and no statistically significant
differences were found, F(1, 171) = 2.45, p = .12, 
p
2
= .01.
Analyses revealed that younger adults expressed significantly more anxiety than older
adults in the choice scenario F(1, 81) = 10.42, p = .002, 
p
2
= .11 with no age differences found
in the control scenario F(1, 87) = .39, p = .53, 
p
2
= .00. A graphical inspection (see Figure 3)
and further analyses show this to be due to differences in the older adult groups, whereby older
adults in the choice scenario expressed significantly less anxiety than those in the control
scenario, F(1, 72) = 8.34, p = .005, 
p
2
= .10; with no differences in anxiety level by condition
for younger adults, F(1, 96) = .40, p = .53, 
p
2
= .00. In other words, older adults in general were
41

less anxious in the choice condition compared to the control condition, whereas younger adults
were similarly anxious overall regardless of the condition they were in.
Sadness. A 2 x 2 between groups ANOVA was conducted to explore the effects of age
group and choice condition on coded expressions of sadness. Baseline levels of depression and
anxiety as well as word count were included as covariates. Consistent with hypothesis 1, results
indicated that younger adults expressed more sadness than older adults overall, F(1, 171) = 4.01,
p =.047, 
p
2
= .02. No significant differences in condition, F(1, 171) = .01, p = .93, 
p
2
= .00, or
interaction effects, F(1, 171) = .25, p = .62, 
p
2
= .00, were found (see Figure 4). In other words,
participants in the choice condition did not differ in expressed sadness compared to those in the
control condition. Furthermore, while younger adults expressed more sadness than older adults,
this difference did not depend on the experimental condition they were in.
Coping language. A 2 x 2 between subjects ANOVA was conducted to determine the
effects of age group and choice condition on ATSS-coded expressions of coping style. Baseline
levels of depression and anxiety were included as covariates. We anticipated that younger adults
would use more problem-focused coping language in general compared to older adults
(hypothesis 4). Results did not support hypothesis 4, as no significant overall main effects of age
were found, F(1, 172) = .02, p = .89, 
p
2
= .00. However, while not hypothesized, results
revealed a significant age by condition interaction, F(1, 172) = 7.68, p = .006, 
p
2
= .04,
indicating that younger and older adults reacted differently to the choice condition in comparison
to the control condition. An overall main effect of condition was also found, F(1, 172) = 29.69, p
< .001, 
p
2
= .15 revealing that in general, participants in the control condition expressed more
problem-focused than emotion-focused coping relative to those in the choice condition. See
Figure 5 for a graphical representation.
42

Analyses revealed that both younger, F(1, 97) = 4.39, p = .04, 
p
2
= .04, and older, F(1,
73) = 27.01, p < .001, 
p
2
= .27 adults on average used relatively more problem-focused than
emotion-focused coping language in the control condition compared to in the choice condition. A
graphical inspection shows a reversal in the amount of problem-focused to emotion-focused
coping language used by younger and older adults, in that older adults used more P-E coping
than younger adults in the control condition, but less than younger adults in the choice condition.
However, analyses did not find significant age differences in the choice condition F(1, 82) =
1.48, p = .23, 
p
2
= .02 or the control condition F(1, 88) = 1.32, p = .25, 
p
2
= .02. Taken
together, these findings indicate that both younger and older adults expressed more problem-
focused coping language relative to emotion-focused coping language in the control condition
compared to the choice condition, with the strength of this difference being larger for older
adults than younger adults.
Emotion-focused coping. In addition to testing the relative use of emotion-focused to
problem-focused coping language, we also examined the effects of age group and condition on
emotion-focused coping language alone as a post-hoc analysis. Word count was not significantly
correlated with this measure (r = .03, p = .66) and therefore was not used as a covariate. Baseline
measures of depression and anxiety were included as covariates. A significant interaction effect
was found, indicating that older and younger adults’ use of emotion -focused coping language
was dependent on the type of condition they were in, F(1, 172) = 4.50, p = .04, 
p
2
= .03. A
significant main effect for condition was also found F(1, 172) = 13.34, p < .001, 
p
2
= .07
indicating that overall, participants in the choice condition expressed more emotion-focused
coping language compared to those in the control condition. No significant main effect of age
was found, F(1, 172) = .54, p = .46, 
p
2
= .00 (see Figure 6).
43

Inspection of this interaction revealed that older adults in the choice condition used more
emotion-focused coping language than older adults in the control condition, F(1, 73) = 12.02, p =
.001 
p
2
= .14, whereas no differences in emotion-focused coping language were found by
condition for younger adults, F(1, 97) = 1.68, p = .20, 
p
2
= .02. Age differences were not found
to be significant for both the choice, F(1, 82) = 2.04, p = .16 
p
2
= .02, and control conditions,
F(1, 88) = .11, p = .73 
p
2
= .00. Therefore, results indicate that the age by choice interaction
found for emotion-focused coping language was due to differences in older adults’ usage,
whereby older adults expressed more emotion-focused coping when in the condition where
choice was salient compared to the control condition.
Problem-focused coping. Post-hoc analyses examined the effects of age group and
condition on the use of problem-focused coping language. Word count was found to be
significantly correlated with this measure (r = .26, p = .001) and was therefore used as a
covariate in the main analyses, along with baseline measures of depression and anxiety. A
significant age by condition interaction was found, indicating that younger and older adults’ u se
of problem-focused coping language depended on which experimental condition they were in,
F(1, 171) = 4.93, p = .03, 
p
2
= .03 (see Figure 7).
Overall, older adults expressed more problem-focused coping language compared to
younger adults F(1, 171) = 5.79, p = .02, 
p
2
= .03, but this difference depended on the condition.
Specifically, within the control condition, older adults expressed more problem-focused coping
language than younger adults, F(1, 87) = 6.16, p = .02, 
p
2
= .07; however no age differences in
problem-focused coping language were found within the choice condition, F(1, 81) = .37, p =
.55, 
p
2
= .01. Overall differences in condition were also found, F(1, 171) = 13.78, p < .001, 
p
2
= .08, whereby those in the control condition expressed more problem-focused coping language
44

than those in the choice condition. When comparing condition differences by age group, older
adults in the control condition were shown to have used significantly more problem-focused
coping language compared to older adults in the choice condition, F(1, 72) = 13.45, p < .001, 
p
2
= .16. Younger adults showed a similar pattern; however, differences between condition for this
age group were not statistically significant, F(1, 96) = 1.79, p = .18, 
p
2
= .02. Taken together,
these findings reveal that the age by condition interaction effect for problem-focused coping
language was driven primarily by older adults’ usage, whereby older adults expressed more
problem-focused coping when in the control condition compared to the choice condition.
Self-reported Feelings of Regret
In addition to articulated thoughts data, we collected Likert scale ratings of self-reported
general mood and regret collected throughout the scenario. To review this procedure: Following
each audio segment (starting after the choice point or the equivalent timeframe in the control
condition), participants were asked to rate their general mood at the moment “right now” (1 =
extremely pleasant; 9 = extremely unpleasant) as well as their level of regret at the moment (1 =
not at all regretful; 9 = extremely regretful). After rating their mood and level of regret,
participants moved on to the next segment and continued with the scenario.
An initial analysis indicated that the two self-report measures were strongly correlated
with each other (r = .75). In addition to having conceptual overlap, the high correlation between
the two variables may have been due to the fact that both questions were always presented at the
same time and had similarly scaled response options. In light of this, we decided to look
primarily at scores of regret in order to avoid redundancy and issues of collinearity.
A 2 (age) by 2 (condition) between groups ANOVA was conducted on self-rated regret.
Similar to ATSS analyses, baseline depression and anxiety scores were included in the model as
45

covariates. Results revealed a significant age by condition interaction, F(1, 172) = 4.17, p = .04,

p
2
= .02. Analyses revealed that in the choice condition, younger adults rated their level of
regret as higher than older adults, F(1, 82) = 8.10, p < .01, 
p
2
= .09. In contrast, no age
differences in self-reported regret were found within the control condition, F(1, 88) = .04, p =
.85, 
p
2
= .00. These findings were somewhat consistent with hypothesis 3, as we anticipated age
differences in the choice condition to be larger than in the control condition. On the other hand,
hypothesis 2 was not supported for this measure, as no significant condition differences were
found for younger adults, F(1, 97) = 1.49, p = .23, 
p
2
= .02 or older adults, F(1, 73) = 2.57, p =
.11, 
p
2
= .03. This indicates that neither younger nor older adults significantly differed in their
regret ratings by condition. An inspection of means revealed that on average, while not
statistically significant (p’s > . 5) , younger adults rated their regret as higher in the choice
condition compared to the control condition and, older adults, on average, rated their regret as
higher in the control condition compared to the choice condition (see Figure 8). Taken together,
unlike ATSS coded regret analyses which did not find a significant age by condition interaction
effect, self-rated regret measures did find that younger adults were more regretful than older
adults in the choice condition, with no age differences found in the control condition.
Future Time Perspective and Use of Coping Strategies
While not specifically hypothesized, we were interested in understanding the relationship
between future-time perspective and use of problem-focused and emotion-focused coping
language. These exploratory analyses were conducted based on Carstensen’s Socioemotional
Selectivity Theory (SST), which proposes that with a more limited time horizon comes an
increased focus on present-oriented goals such as regulating emotions. We initially ran earson’s
bivariate correlations to determine the relationship between scores on a measure of future-time
46

perspective and expressions of emotion-focused and problem-focused coping language
separately. Interestingly, a significant positive correlation was found (r = .15, p = .04) indicating
that increased time perspective was associated with more expressed emotion-focused coping.
This finding is somewhat contradictory to what would be anticipated according to the SST. No
significant relationship was found for FTP and problem-focused coping language (r = .02, p =
.75). Further inspection revealed that the positive correlation between FTP and emotion-focused
coping was driven by older adults (r = .33, p = .003), with no significant relationship found in
the younger adult group (r = .15, p = .13). Therefore, within our older adult sample, more
expanded time horizon was associated with more emotion-focused coping language, with no
correlation found between time-perspective and problem-focused coping language.
Results by Hypothesis
We have just reported the findings organized around the several ATSS-coded cognitions
and emotions as well as some self-report questionnaire measures, mentioning along the way how
the findings relate to our hypotheses. For comprehensiveness and clarification purposes, findings
from this study will now be organized around the hypotheses. To simplify the results and as
these findings are described and structured by analysis (emotion) above, statistical findings are
represented here simply as adjusted p-values and effect sizes. All analyses include baseline
scores of depression and anxiety as covariates. Word count is also included as a covariate for
ATSS coded variables anger at others, anxiety, and sadness.
Hypothesis 1 (age differences in expressed emotion). We anticipated that across
conditions, younger adults would express more regret, guilt, anger at others, anger at themselves,
and anxiety compared to older adults. Analyses revealed significant main effects of age in the
hypothesized direction for all ATSS-coded emotions (p < .05). Specifically, younger adults
47

expressed more regret (p = .004, 
p
2
= .06), guilt (p = .009, OR = 3.29), anger at the self (p = .03,
OR

= 2.17), anger at others (p = .005, 
p
2
= .07), and anxiety (p = .03, 
p
2
= .03). Consistent with
results from articulated thoughts of regret, responses from a self-report Likert measure of regret
also showed that younger adults reported feeling more regretful than older adults (p = .04, 
p
2
=
.02). While not hypothesized, ATSS coded responses revealed that younger adults also expressed
more sadness compared to older adults (p = .047, 
p
2
= .02). Thus, there is considerable support
for Hypothesis 1. See Table 6.
Hypothesis 2 (condition differences). We anticipated that compared to those in the
control condition, participants in the choice condition would express more regret, guilt, and
anger at themselves based on ATSS-coded responses. Findings were consistent with this
hypothesis. Specifically, participants in the choice condition expressed more ATSS-coded regret
(p = .03, 
p
2
= .03), guilt (p = .003, OR = 8.71), and anger at the self, (p = .004, OR = 2.75)
compared to those in the control condition. We also anticipated participants in the choice
condition to rate their regret level as higher than participants in the control condition on a Likert
measure of regret. This hypothesis was not supported, as significant differences were not found,
(p = .58, 
p
2
= .00). Differences in expressed anger at others, anxiety, and sadness by condition
were not hypothesized and no significant differences were found (p > .12).
Hypothesis 3 (age by condition interaction). We hypothesized an interaction between
age group and choice condition for ATSS-coded expressions of regret (and self-reported Likert
measure of regret), guilt, and anger at the self. Our findings did not support this hypothesis for
ATSS-coded emotions, (p > .79). In other words, age differences in expressed regret, guilt, and
anger at the self were not found to be significantly larger in the choice condition relative to the
control condition as anticipated. In contrast to the ATSS-coded measure of regret and in support
48

of our hypothesis, an age by condition interaction effect was found for a Likert measure of self-
rated regret level (p = .04 , 
p
2
= .02). Analyses revealed that in the choice condition, younger
adults rated their level of regret as higher than older adults, (p < .01, 
p
2
= .09). In contrast, no
age differences in self-rated regret were found within the control condition, (p = .85, 
p
2
= .00).
No significant condition differences were found for younger adults (p = .23, 
p
2
= .02) or older
adults (p = .11, 
p
2
= .03).
Interestingly, while not hypothesized, an age by condition interaction effect was also
found for ATSS-coded expressions of anxiety (p = .045, 
p
2
= .02). Analyses revealed that
younger adults expressed significantly more anxiety than older adults in the choice scenario (p =
.002, 
p
2
= .11) with no age differences found in the control scenario (p = .53, 
p
2
= .00). A
graphical inspection and further analyses show this to be due to differences in the older adult
groups, whereby older adults in the choice scenario expressed significantly less anxiety than
those in the control scenario (p = .005, 
p
2
= .10); with no differences in anxiety level by
condition for younger adults (p = .53, 
p
2
= .00).
Hypothesis 4 (coping measures). We hypothesized that younger adults would express
relatively more problem-focused to emotion-focused coping strategies compared to older adults.
Results did not support this hypothesis, with no significant age differences found in relative
expressions of coping strategies (see Table 7). Having said this, findings revealed an interesting
age by condition interaction effect (p = .006, 
p
2
= .04). Both younger (p = .04, 
p
2
= .04) and
older adults (p < .001, 
p
2
= .27) expressed more P-E coping (difference between average
problem and emotion focused coping scores) in the choice condition compared to the control
condition. Effect sizes show that the difference between older adults’ expression of -E coping
in the choice condition compared to the control condition was stronger than the difference
49

between younger adults’ expression of -E coping for the two conditions. Further analyses did
not find significant age differences in the choice condition (p = .23, 
p
2
= .02) or the control
condition (p = .25, 
p
2
= .02). Taken together, these coping findings indicate that overall,
relatively more problem-focused coping language was used in the control condition and
relatively more emotion-focused coping language was used in the choice condition. This
difference in coping language usage by condition was larger for older adults than for younger
adults. Post-hoc analyses were conducted to explore age x condition effects for emotion-focused
coping and problem-focused coping separately, with results showing age by condition interaction
effects for both measures. Analyses revealed that the effects were driven by differences in older
adults’ use of coping style by condition, where older adults in the choice condition expressed
more emotion-focused coping language and less problem-focused coping language than older
adults in the control condition. Younger adults, on the other hand, used relatively similar
amounts of problem-focused and emotion-focused coping language regardless of condition.
Discussion
The present study examined whether the positivity effect in aging extends to reactions to
unfavorable post-decision outcomes, particularly regret for a decision that resulted in negative
consequences. In addition, the study examined the role of choice in the nature of these cognitive
and emotional reactions. Using the ATSS paradigm, all participants listened and responded to the
same unfolding scenario, with one group having to make a decision between two options
(choice-salient condition) and the other group having no decision to make (control condition). In
all instances the outcome was undesirable. We hypothesized that (1) older adults would react less
negatively in general compared to younger adults, (2) participants in the choice-salient condition
would express more regret, guilt, and anger at the self compared to those in the control condition,
50

(3) the positivity effect in older adults would be larger in the choice-salient condition compared
to the control condition, and (4) younger adults would express more problem-focused to
emotion-focused coping language relative to older adults. Results fully supported our first two
hypotheses. Hypothesis 3 was not supported, however, in that age differences in regret, guilt, and
anger at the self were not significantly greater in the choice condition relative to the control
condition. An unexpected age by condition interaction effect was found for coping language and
showed that choice saliency (experimental condition) had a significant impact on older adults’
use of problem-focused and emotion-focused coping language. In contrast, younger adults were
shown to express similar amounts of both coping strategies when analyzed separately regardless
of condition. These results highlight older adults’ increased ability to attune to contextual aspects
of the situation and to adapt their coping strategies to most effectively regulate their emotions.
Taken together, our findings show that the positivity effect in older adults extends to cognitive
and emotional reactions to negative decision outcomes, and are consistent with older adults’
goal-directed focus on emotion regulation. We turn now to a discussion of each of these findings.
Hypothesis 1: Younger adults will express more regret, guilt, anger at others, anger at
themselves, and anxiety compared to older adults
We hypothesized that older adults would express less regret, guilt, anger at others, anger
at themselves, and anxiety compared to younger adults across conditions (hypothesis 1). We
measured these cognitions and emotions using ATSS-coded responses, with regret measured also
with a self-rated Likert scale measure. Our results supported this hypothesis for all of the
variables. As guilt and anger at the self were analyzed dichotomously, results revealed that
younger adults were more likely to express these emotions than older adults. The overall results
are consistent with previous studies examining the positivity effect in older adults and show that
51

this effect also extends to the way they respond to negative decision-related outcomes. In
addition to the emotions that we hypothesized would differ by age, we also found that younger
adults expressed more sadness than older adults in our study.
Older adults in prior studies have self-reported experiencing fewer regrets than younger
adults (Timmer et al., 2005); however it is unclear whether this age difference is due primarily to
older adults’ avoidance and risk -averseness or whether it is due to differences in reaction to an
unfavorable event. The latter is a response-focused style of coping and was tested in the present
study by controlling the environment. Specifically, this study was designed so that every
participant listened to and imagined themselves in the same scenario. Therefore, older adults in
our study were not less regretful or less anxious because they avoided the situation or had
different interactions than younger adults, but rather because of the way they reacted to the
ongoing negative events that were occurring. This is an important finding, as some researchers
have attributed the positivity effect to be primarily due to avoidance of negative situations (e.g.,
Charles, Piazza, & Luong, 2009). The use of ATSS permitted this examination.
Furthermore, we were conscious of potential baseline differences in depressive and
anxiety symptoms by age group and controlled for these factors within our analyses. Therefore,
even after controlling for initial differences in mood symptoms, younger adults were more
negatively affected by the unpleasant situation than older adults in our sample. These results
support the extension of the positivity effect in older adults to negative post-decision outcomes
when compared across similar settings.
Hypothesis 2: Choice salient condition will elicit more regret, guilt, and anger at the self
than the control condition
52

Based on previous literature on the emotional consequences of decision-making and
predictors of negative post-decision outcomes such as regret, we anticipated that participants in
the choice condition would express more regret, guilt, and anger at the self compared to those in
the control condition (hypothesis 2). Our results fully supported this hypothesis for these ATSS
coded cognitions and emotions. Specifically, in the condition where participants imagined
themselves making a choice about changing flights, they expressed more regret relative to those
exposed to the same unpleasant hypothetical events but without having made a choice. As guilt
and anger at the self were analyzed dichotomously, findings indicate that individuals in the
choice condition were more likely to express guilt and anger at the self relative to those in the
control (non-choice) condition.
To recall, the scenarios presented for each condition were virtually identical in the
content and unfolding of the negative events that occurred. The only major difference was that
the scenario in the choice condition included a “choice point” early on that asks participants to
either keep their existing flight or change to an alternative flight. While this was not directly
measured, we believe that this choice point affected the perspective of the scenario in that the
choice condition was seen more as a “post -decision outcome” scenario whereas the scenario in
the control condition was more simply a negative situation with no saliency of choice presented
(and more similar to previous studies examining age differences in reacting to unpleasant
situations, e.g. Charles & Carstensen, 2008). The condition difference was subtle, and the fact
that differences in negative reactions were found by condition speaks to the impact of having a
choice point on participants’ reactions to a negative situation.
Support for the effectiveness of this manipulation was also demonstrated by responses of
participants’ feelings of responsibility. When asked after the scenario how responsible they felt
53

for the situation they just encountered, participants in the choice condition rated their feelings of
responsibility as much higher than those in the control condition. The present study contributes
to the decision-making and regret literature by highlighting the importance of choice saliency on
negative post-decision reactions within an experimentally controlled setting.
Hypothesis 3: Age differences will be larger in the choice condition compared to the control
condition
We anticipated that younger adults would be more negatively affected by aspects of the
choice condition than older adults (more regretful, guilty, angry at the self) and that these
differences would be larger than any age differences found in the control condition; however
results did not support this (hypothesis 3). Specifically, no significant age by condition
interaction effects were found for regret, guilt, or anger at the self. Lack of significant interaction
effects indicate that age differences in the choice condition were not significantly larger or
smaller than age differences found in the control condition.
While a lack of significant interaction effects does not support hypothesis 3, results can
be taken to support the interpretation for hypothesis 1, specifically that significant age
differences in negative reactions (regret, guilt, anger at self) found in the control condition did
not significantly differ from reactions in the choice condition. If a significant interaction effect
revealed age differences in the control condition (which was expected) but no differences in the
choice condition, results would signify that the positivity effect for these emotions did not extend
to situations where choice was salient in affecting the negative scenario. Instead, this finding (or
rather, lack of significant finding) provides further support for the positivity effect extending to
negative post-decision making reactions similar to the positivity effect found in older adults
responding to other unpleasant situations (e.g. Charles & Carstensen, 2008). No significant
54

interaction effects were found for expressions of anger at others or sadness, which was not
hypothesized but is also consistent with interpretations for hypothesis 1.
Taken together, these results indicate that younger adults expressed more regret, guilt,
anger at themselves, anger at others, and sadness compared to older adults, and that these age
differences in emotions did not differ depending on whether they were in the choice condition or
the control condition.
Older adults’ anxiety differed by condition whereas younger adults’ did not: the potential
role of emotion-focused strategies
While not hypothesized, results from our study found a significant age by condition
interaction effect for ATSS-coded expressions of anxiety. Specifically, younger adults were
found to be more anxious compared to older adults in the choice condition but no age differences
were found in the control condition. Therefore, hypothesis 1 was partially supported when
broken down by condition, in that the positivity effect was found for anxiety within the choice
condition but no age differences were found in the control condition, which is inconsistent with
our hypothesis. Baseline depressive and anxiety symptoms were controlled for within these
analyses as well, lending more support for these findings as appropriately capturing reactions to
the experimental situation.
These results suggest that for expressed anxiety, the positivity effect in aging is more
prominent in reacting to situations where there is a decision-making component. It is possible
that the added feeling of responsibility within the choice condition affected younger adults more
significantly than older adults, which is what we had anticipated for post-decision making
emotions (regret, guilt, anger at self). A graphical inspection, however, shows that the significant
interaction effect was not due to younger adults feeling more anxious in the choice compared to
55

the control condition, but rather that older adults in the choice condition expressed significantly
less anxiety than older adults in the control condition. Notably, anxiety was the only measured
emotion out of all hypothesized emotions that was significantly lower in the choice condition
relative to the control condition, and this was only found within the older adult group. Less
anxiety in the choice condition is consistent with the perceived control literature (cf. Geer,
Davison, & Gatchel, 1970; Taylor, Helgeson, & Reed, 1991), which postulates that increased
feelings of control reduce experienced anxiety and that this type of reaction is generally adaptive.
Hypothesis 4: Younger adults will express more problem-focused coping language in
general compared to older adults.
Previous research has shown younger adults to use more problem-focusing coping than
older adults (strategies with the goal of solving a problem), especially when reacting to
interpersonal challenges (e.g. Blanchard-Fields, Mienaltowski, & Baldi, 2007). We hypothesized
that older adults would express relatively more emotion-focused (strategies to alleviate negative
emotions) to problem-focused coping language compared to younger adults, who we anticipated
would use more problem-focused coping language in general (hypothesis 4).
Interestingly, we found that age differences in coping language interacted with condition
type. Both younger and older adults expressed more problem-focused relative to emotion-
focused coping in the control condition compared to the choice condition. The difference in
coping language by condition for older adults, however, was larger than the difference for
younger adults ( 
p
2
= .27 vs .04). In fact, when broken down by coping style and analyzed
separately, it was apparent that older adults were driving the interaction effect for both emotion-
focused coping language as well as problem-focused coping language. Graphical representations
of these analyses illustrate the strong influence that choice saliency (experimental condition) had
56

on older adults’ use of problem -focused and emotion-focused coping language. In contrast,
younger adults were shown to express similar amounts of both coping strategies regardless of
condition.
We interpret these findings to support older adults’ increase d ability to attune to
contextual aspects of the situation and to adapt their coping strategies to most effectively regulate
their emotions. In the choice condition, negative unfolding events that occur throughout the
scenario can be viewed as a result of their decision to either stay on their existing flight or switch
to the alternative flight. Therefore, hearing that the chosen flight is delayed but the alternative
flight (counterfactual) is leaving on time is likely to elicit feelings of regret. In this situation, if
regulating emotions is a high priority and feelings of regret are apparent, expressing strategies to
make oneself feel better about the situation is a realistic goal. Older adults in the choice
condition used much more emotion-focused coping language than those in the control condition.
On the other hand, without the choice point, hearing that the chosen flight is delayed and that
another flight is leaving earlier may elicit more feelings of anxiety than regret – which is what
was found with our older adult group (more regret and less anxiety in the choice condition
compared to control condition). Without the earlier choice point, older adults in the control
condition may have focused more on solving the problem as a way of maintaining well-being
and therefore expressed more problem-focused coping language.
These results are consistent with prior research showing older adults’ increased flexib ility
in their use of coping strategies when dealing with unpleasant situations compared to younger
adults (e.g. Blanchard-Fields, Jahnke, & Camp, 1995). Specifically, when already in a conflict
situation, unlike younger adults, who appear more inclined to use problem-focused strategies in
57

both emotional and non-emotional situations, older adults appear to tailor their strategies to the
contextual features of the problem at hand (Blanchard-Fields, Mienaltowski, & Baldi, 2007).
Our results support this notion: older adults in our study appeared to be using specific
types of coping strategies to match the situation, whereas younger adults on average used
relatively similar strategies across situations. From a goal-oriented perspective, older adults’
potential focus on maintaining emotional well-being may have influenced their increased usage
of emotion-focused coping strategies when reacting to a negative choice outcome. On the other
hand, older adults in the control condition were not anchored by a decision point at the beginning
of the scenario. Therefore, it is plausible that instead of viewing the situation as coping with their
“bad” decision (thinking back) in the way that those in the choice condition may be viewing the
situation, individuals in the control condition may have instead been anticipating a negative
outcome to occur (thinking forward). This would explain older adults’ increased usage of
problem-focused language in the control condition as well as their higher levels of anxiety, as
they may have been trying to manage the situation in order to minimize the chances of emotional
distress in the near future. Younger adults on the hand, may have been focused more on solving
the problem, which would explain why their use of individual coping strategies did not
significantly differ by condition (the problems were the same in both situations). These findings
are particularly exciting, as they show younger and older adults’ coping style to the same
situations. Therefore, differences in emotional reactions and coping style were not due to
antecedent coping strategies (e.g., avoiding potentially unpleasant situations) within this study,
but instead explicitly show older adults’ use of response focused coping styles to maintain
emotional well-being.
58

Taken together, these findings support the notion of age-related differences in goal-
directed cognitive and emotional processes. On the other hand, results did not support the general
lifespan theory of control, which suggests that with increasing age comes an increased need to
enact secondary control processes such as regulating emotions. Instead, our results revealed older
adults’ adaptive coping abilities in reaction to negative situations that includes both emotion -
focused and problem-focused coping strategies. In fact, older adults were shown to express more
problem-focused coping language in the control condition compared to younger adults.
Examining future-time perspective and coping
The positivity effect in older adults has been hypothesized to be due to a motivational
shift and increased focus on emotionally meaningful goals, which typically results in more
complex emotional experiences as well as better emotion regulation (Carstensen, Fung, &
Charles, 2003). This typically refers to maintaining emotional well-being by minimizing
negative mood and maximizing positive mood. The Socioemotional Selectivity Theory would
suggest that a more limited time horizon would increase one’s motivation to focus on
maintaining well-being in the present moment.
In contrast to this notion, we found that more limited future time-perspective was
associated with less emotion-focused coping language, not more as would be expected given this
theory. This was particularly true for older adults in our sample, where more expanded time
perspective was positively correlated with more emotion-focused coping language. Future time
perspective and coping language were not significantly correlated for younger adults, possibly
because of the restricted range of scores for the future time perspective measure with this group.
No significant correlation between future time perspective and problem-focused coping language
were found.
59

We believe that older adults who felt they had a more expanded time horizon may have
been more able to console themselves by thinking they could attend the event at another time or
somehow make up for the missed event in the future. These statements would have been coded
as emotion-focused coping strategies, as they would be in service of improving their current
mood state. On the other hand, those who believed they had a more limited time horizon may
have been less able to comfort themselves for missing an event that they found very meaningful
to them, as they may have believed that they have fewer chances to experience these meaningful
events.
Differences between Likert regret and ATSS coded expressions of regret
In contrast to the ATSS coded expression of regret, a significant age by condition
interaction was found for Likert regret, where younger adults rated themselves as feeling
significantly more regretful than older adults in the choice condition but no age difference found
in the control condition. This finding partially supports hypothesis 3, where larger age
differences are expected in the choice condition compared to the control condition. However, we
anticipated that younger adults would report higher levels of regret than older adults in both
conditions, but that younger adults would be significantly more regretful than older adults in the
choice compared to the control condition. Interestingly, a post-hoc inspection of means indicates
that this interaction was due to the combination of younger adults in the choice condition rating
themselves as more regretful than younger adults in the control condition, and older adults in the
choice condition rating themselves as less regretful than older adults in the control condition.
While these differences were not large enough to be statistically significant on their own, the
combination of the two effects was substantial enough to result in a significant interaction effect.
60

This pattern of results differs from the results found for ATSS-coded expressions of
regret, where significant main effects of age and condition were found but no significant
interaction effects were revealed. It is unclear why we found differences between these two
measures. One potential reason for this could be the nature of the measures themselves,
specifically the difference in information gained from an open-ended modality of measurement
(ATSS) compared to the restricted response format garnered from a Likert scale measurement.
This may be especially pertinent for older adults, as their expressed level of regret appears to
differ more so depending on the mode of measurement. In various ATSS publications, the basic
differences between ATSS and more conventional self-report methods have been emphasized
(e.g., Davison & Zighelboim, 1987; Davison et al., 1997).
Limitations and Strengths
The current study tested age differences in emotional responses and coping strategies
within a controlled lab-based paradigm. Limitations to this study may be beneficial in guiding
future research in this area.
As with all cross-sectional designs, results from the present study support an association
between age and post-decision reactions; however, factors such as potential cohort effects
preclude clear conclusions regarding causation, and of course age is a correlational variable.
Analyzing two distinct age groups also limits the ability to detect potential curvilinear effects of
a middle-aged group. Given the time and resources needed to include another group, we made a
conscious decision to focus on recruiting a sufficient number of younger and older participants
rather than to disperse resources in order to create three groups with fewer participants in each.
Participants were also required to come to the USC campus in order to participate, which may
have led to a selected sample of healthier older adults compared to the general population. To
61

allow for a more representative sample of younger and older adults, the present study also
recruited participants from the greater Los Angeles area.
A more specific limitation of the current study relates to the decision not to use audio
recording when content-analyzing participants’ data. The ATSS data were coded using transcript
only to limit coders’ awareness of whether the participant was younger or older. The benefit of
using only transcripts was also to ensure a “fairer” comparison between younger and older
adults, as older adults tend to speak slower than younger adults and this general difference could
have inappropriately affected coders’ ratings of how emotional these individuals appeared. On
the other hand, using only transcripts to code verbalizations may have led us to miss important
nonverbal cues that could have affected coder ratings in a valid way. Much of the information we
receive when communicating with others is through paralinguistic features such as voice volume
and tone, unavailable in transcripts. Having said this, intraclass correlations between raters were
sufficiently high and significant age differences were found in the hypothesized direction.
Therefore, the decision to use transcripts only did not prevent the emergence of predicted
findings.
In addition, emotional reactions and coping style were measured using participants’
verbalized thoughts and emotions. The potential downside to this is that participants must be
aware of the emotions they are feeling in order to verbalize them. This is not unlike many other
measures of emotion, however, including questionnaires and other self-report measures. This
study also employed a hypothetical scenario and asked participants to imagine themselves in that
scenario. Therefore, intensity of expressed emotions may have been affected by participants’
ability and/or willingness to engage in the scenario. Again, this is not significantly different than
other self-report measures. That is to say, reports on interviews and on questionnaires are in
62

response to verbally presented material that is no more “real” than the scenarios in ATSS
research.
While this study has certain limitations, it also provides a number of important
contributions to the aging and decision-making literature. First, the present study’s use of the
ATSS method allowed us to capture thoughts and emotions to the same environmental situation.
Therefore, as mentioned earlier, we were able to limit participants’ use of antecedent coping
strategies such as avoiding potentially unpleasant interactions, which older adults have been
shown to do in their day to day lives (Charles et al., 2009). By asking all participants to engage
in the same scenario with negative events unfolding in the same way, we could more reliably
compare younger and older adults’ reactions to negative events, and specifically to unfavorable
post-decision outcomes. Second, along with controlling for environment, we were also able to
control for time. Young and older adults were asked to verbalize their thoughts and feelings at
the same structured times throughout the scenario. They were given the same amount of time to
express their reactions and this was done in a near-immediate fashion. Therefore, unlike
retrospective reporting where memory differences are likely to play a role, the use of the ATSS
paradigm limits the added memory component and captures data at the same times for all
participants. These aspects allowed us to more reliably compare “apples to apples”. Third, we
were mindful of accounting for potential baseline measures of depression and anxiety between
age groups. We were also aware of differences in speech rate between younger and older adults
and accounted for this by adding word count as a covariate within our ATSS coded analyses.
Therefore, analyses within this study were conducted in a way that most effectively compares
older and younger adults’ reactions to the experimental scenarios. Taken together, we believe
63

that the design and analytic approach taken to study age-related reactions to negative post-
decision outcomes were strengths of this study and allow for confidence in the study’s findings.
Implications for Future Research
The present study adds to both the literature on age-related positivity effects in attention
and memory as well as the decision-making literature by focusing on age differences in
responding to the negative consequences of a decision. Our study provided some initial
understanding of the relationship between negative post-decision emotions such as regret and
guilt and the different types of coping strategies employed, as well as how these relationships can
differ by age. Future studies may extend this line of research by focusing on the effects of regret
and coping style on future behavior. For example, does feeling regret decrease the likelihood of
or confidence in making future decisions in the same domain, or even a different domain? What
are the mediating factors affecting potential behavior change and do these effects depend on age-
related factors?
Another interesting area of future research is understanding the role of self-efficacy and
feelings of control on coping style as it relates to negative post-decision reactions. Theories have
suggested that problem-focused coping strategies are employed more often when individuals feel
they can control the situation, whereas emotion-focused coping strategies are more likely to be
used when the situation feels uncontrollable. Future studies could determine whether negative
feelings such as regret and guilt increase or decrease these feelings of self-efficacy or control and
whether these impact willingness to make future decisions and/or the types of decisions made.
Summary and Conclusions
The present study examined younger and older adults’ responses to an unfavorable
decision outcome using a controlled hypothetical scenario (employing the ATSS paradigm). We
64

found age-related positivity effects in feelings of regret, guilt, anger at the self and others,
anxiety, and sadness. This effect extended to negative post-decision reactions, where younger
adults reacted more negatively than older adults to the same situation. Both older and younger
adults expressed more problem-focused compared to emotion-focused coping strategies in the
control condition relative to the choice condition. Older adults’ use of different coping strategies
(specifically emotion-focused and problem-focused) were clearly shown within our experimental
study when these strategies were examined separately, as older adults expressed more problem-
focused coping language in the control condition compared to the choice condition, and more
emotion-focused coping language in the choice condition compared to the control condition.
Younger adults on the other hand, tended to use similar amounts of each strategy regardless of
condition. These results are consistent with prior research showing older adults’ increa sed
flexibility in their use of coping strategies when dealing with unpleasant situations compared to
younger adults. We interpret these findings to support older adults’ superior ability to attune to
contextual aspects of the situation and to adapt their coping strategies to most effectively regulate
their emotions.

65

References

Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging: The
model of selective optimization with compensation. In P.B. Baltes & M. M. Baltes (Eds.),
Successful aging: Perspectives from the behavioral sciences (pp. 1-34). New York:
Cambridge University Press.
Birditt, K. S., & Fingerman, K. L. ( 3). Age and gender differences in adults’ descriptions of
emotional reactions to interpersonal problems. Journal of Gerontology, Series B:
Psychological Sciences and Social Sciences, 58, 237-245.
Birditt, K. S., & Fingerman, K. L. (2005). Do we get better at picking our battles? Age group
differences in descriptions of behavioral reactions to interpersonal tensions. Journal of
Gerontology, Series B: Psychological Sciences and Social Sciences, 60, 121-128.
Birditt, K. S., Fingerman, K. L., & Almeida, D. M. (2005). Age differences in exposure and
reactions to interpersonal tensions: A daily diary study. Psychology and Aging, 20, 330 –
340.
66

Bjalkebring, P., Vastfjall, D., & Johansson, B. (2013). Regulation of experienced and anticipated
regret for daily decisions in younger and older adults in a Swedish one-week diary study.
GeroPsych, 26, 233-241.
Blanchard-Fields, F. (2007). Everyday problem solving and emotion - An adult developmental
perspective. Current Directions in Psychological Science, 16, 26 – 31.
Blanchard-Fields, F. & Coats, A. H. (2008). The experience of anger and sadness in everyday
problems impacts age differences in emotion regulation. Developmental Psychology, 44,
1547-1556.
Blanchard-Fields, G., Jahnke, H. C., & Camp, C. (1995). Age differences in problem-solving
style: The role of emotional salience. Psychology and Aging, 10, 173-180.
Blanchard-Fields, F., Mienaltowski, A., & Baldi, R. (2007). Problem solving effectiveness in
older adulthood. Journal of Gerontology, Series B: Psychological Sciences and Social
Sciences, 62, 61-64.
Blanchard-Fields, F., Stein, R., & Watson, T. L. (2004). Age differences in emotion-regulation
strategies in handling everyday problems. Journal of Gerontology, Series B:
Psychological Sciences and Social Sciences, 59, 261–269.
Blazer, D. G. (2003). Depression in late life: review and commentary. Journal of Gerontology,
Series A: Biological Sciences and Medical Sciences, 58, 49-65.
Burger, J. M. (1989). Negative reactions to increases in perceived personal control. Journal of
Personality and Social Psychology, 56, 246-256.
Cacioppo, J. T., Berntson, G. G., Klein, D. J., & Poehlmann, K. M. (1998). Psychophysiology of
emotion across the life span. In K. W. Schaie & M. P. Lawton (Eds.), Annual review of
67

gerontology and geriatrics, Vol. 17: Focus on emotion and adult development (pp. 27 –
74). New York: Springer.
Carstensen, L. L. (2006). The influence of a sense of time on human development. Science, 312,
1913-1915.
Carstensen, L. L., & Fredrickson, B. L. (1998). Influence of HIV status and age on cognitive
representations of others. Health Psychology, 17, 494-503.
Carstensen, L. L., Fung, H. H., & Charles, S. T. (2003). Socioemotional Selectivity Theory and
the Regulation of Emotion in the Second Half of Life. Motivation and Emotion, 27, 103-
123.
Carstensen, L. L., Isaacowitz, D. M., Charles, S. T. (1999). Taking time seriously. American
Psychologist. 54, 165-181.
Carstensen, L. L., & Lang, F. R. (1996). Future Time Perspective Scale. Unpublished
manuscript, Stanford University, Palo Alto, CA.
Carstensen, L. L., Mikels, J. A., & Mather, M. (2006). Aging and the intersection of cognition,
motivation and emotion. In J. E. Birren & K.W. Schaie (Eds.), Handbook of the
psychology of aging (pp. 343-362). San Diego, CA: Academic Press.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). Emotional experience in
everyday life across the adult life span. Journal of Personality and Social Psychology, 79,
644–655.
Charles, S. T. (2010). Strength and vulnerability integration (SAVI): A model of emotional well-
being in later adulthood. Psychological Bulletin, 136, 1068-1091.
Charles, S. T., & Carstensen, L. L. (2008). Unpleasant situations elicit different emotional
responses in younger and older adults. Psychology and Aging, 23, 495 – 504.
68

Charles, S. T., Carstensen, L. L., & McFall, R. M. (2001). Problem-solving in the nursing home
environment: Age and experience differences in emotional reactions and responses.
Journal of Clinical Geropsychology, Special Issue: Management of Behavioral Problems
in Late Life –Therapeutic Approaches and Related Issues, 7, 319-330.
Charles, S. T., & Luong, G. (2013). Emotional experience across adulthood: The theoretical
model of strength and vulnerability integration. Current Directions in Psychological
Science, 22, 443-448.
Charles, S. T., iazza, J. R., Luong, G. ( 9). Now you see it, now you don’t: Age
differences in affective reactivity to social tensions. Psychology and Aging, 24, 645-653.
Charles, S. T., Mather, M., & Carstensen, L. L. (2003). Aging and emotional memory: The
forgettable nature of negative images for older adults. Journal of Experimental
Psychology: General, 132, 310–324.
Charles, S. T., Reynolds, C. A., & Gatz, M. (2001). Age-related differences and change in
positive and negative affect over 23 years. Journal of Personality and Social Psychology,
80, 136-151.
Chipperfield, J. G., Perry, R. P., & Weiner, B. (2003). Discrete emotions in later life. Journals of
Gerontology, Series B: Psychological Sciences and Social Sciences, 58, 23-34.
Coats, A. H., & Blanchard-Fields, F. (2008). Emotion regulation in interpersonal problems: The
role of emotional complexity, emotion regulation goals, and expressivity. Psychology and
Aging, 23, 39-51.
Connolly, T., & Zeelenberg, M. (2002). Regret and decision making. Current Directions in
Psychological Science, 11, 212-216.
69

Davison, G. C., Robins, C., & Johnson, M. K. (1983). Articulated thoughts during simulated
situations: A paradigm for studying cognition in emotion and behavior. Cognitive
Therapy and Research, 7, 17-39.
Davison, G. C., Vogel, R. S., & Coffman, S. G. (1997). Think-aloud approaches to cognitive
assessment and the articulated thoughts in simulated situations paradigm. Journal of
Consulting and Clinical Psychology, 65, 950-958.
Davison, G. C., & Zighelboim, V. (1987). Irrational beliefs in the articulated thoughts of college
students with social anxiety. Journal of Rational-Emotive Therapy, 5, 238 – 254.
Dijkstra, P., & Barelds, D.P.H. (2008). What Dutch women of different ages regret most.
Journal of Research in Personality, 42, 1663-1667
Finucane, M. L., Slovic, P., Hibbard, J. H., Peters, E., Mertz, C. K., & MacGregor, D. G. (2002).
Aging and decision-making competence: An analysis of comprehension and consistency
skills in older versus younger adults considering health plan options. Journal of
Behavioral Decision Making, 15, 141–164.
Flint, A. (1994). Epidemiology and co-morbidity of anxiety disorders in late life. American
Journal of Psychiatry, 151, 640 – 649.
Folkman, S., Lazarus, R. S., Pimley, S., & Novacek, J. (1987). Age differences in stress and
coping processes. Psychology and Aging, 2, 171–184.
Fung, H. H., & Carstensen, L. L. ( 6). Goals change when life’s fragility is primed: Lessons
learned from older adults, the September 11 attacks and SARS. Social Cognition, 24,
248-278.
Fung, H. H., Carstensen, L. L., & Lutz, A. M. (1999). Influence of time on social preferences:
Implications for life-span development. Psychology and Aging, 14, 595 – 604.
70

Geer, J. H., Davison, G. C., & Gatchel, R. I. (1970). Reduction of stress in humans through
nonveridical perceived control of aversive stimulation. Journal of Personality and Social
Psychology, 16, 731 – 738.
Hasin, D. S., Goodwin, R. D., Stinson, F. S., Grant, B. F. (2005). Epidemiology of major
depressive disorder: Results from the National Epidemiologic Survey on Alcoholism and
Related Conditions. Archives of General Psychiatry, 62, 1097–1106.
Heckhausen, J., & Schulz, R. (1995). A life-span theory of control. Psychological Review, 102,
284 – 304.
Holm, S. (1979). A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian
Journal of Statistics, 6(2), 65–70.
Hosseini, S. M., Rostami, M., Yomogida, Y., Takahashi, M., Tsukiura, T., & Kawashima, R.
(2010). Aging and decision making under uncertainty: Behavioral and neural evidence
for the preservation of decision making in the absence of learning in old age.
Neuroimage, 52, 1514-1520.
Isaacowitz, D. M., Wadlinger, H. A., Goren, D., & Wilson, H. R. (2006a). Is there an age-related
positivity effect in visual attention? A comparison of two methodologies. Emotion, 6,
511-516.
Isaacowitz, D. M., Wadlinger, H. A., Goren, D., & Wilson, H. R. (2006b). Selective preference
in visual fixation away from negative images in old age? An eye tracking study.
Psychology and Aging, 21, 40-48.
Jokisaari, M. (2003). Regret appraisals, age, and subjective well-being. Journal of research in
Personality, 37, 487-503.
71

Kennedy, Q., Mather, M., & Carstensen, L. L. (2004). The role of motivation in the age-related
positivity effect in autobiographical memory. Psychological Science, 15, 208 – 214.
Kim, S., & Hasher, L. (2005). The attraction effect in decision making: Superior performance by
older adults. Quarterly Journal of Experimental Psychology, 58A, 120–133.
Kim, S., Healey, M. K., Goldstein, D., Hasher, L., & Wiprzycka, U. J. (2008). Age differences in
choice satisfaction: A positivity effect in decision making. Psychology and Aging, 23,
33–38.
Knight, M., Seymour, T. L., Gaunt, J. T., Baker, C. Nesmith, K., & Mather, M. (2007). Aging
and goal-directed emotional attention: Distraction reverses emotional biases. Emotion, 7,
705–714.
Kryla-Lighthall, N, Mather M. ( 9). The role of cognitive control in older adults’ emotional
well-being. In V. Berngtson, D. Gans, N. Putney, & M. Silverstein (Eds.), Handbook of
Theories of Aging, (pp. 323–344). New York: Springer. 2nd ed.
Labouvie-Vief, G. (2003). Dynamic integration: Affect, cognition, and the self in adulthood.
Current Directions in Psychological Science, 12, 201 – 206.
Landman, J. (1993). Regret: The persistence of the possible. New York: Oxford University
Press.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer.
Leigland, L. A., Schulz, L. E., & Janowsky, J. S. (2004). Age related changes in emotional
memory. Neurobiology of Aging, 25, 1117-1124.
Löckenhoff, C. E., & Carstensen, L. L. (2004). Socioemotional selectivity theory, aging, and
health: The increasingly delicate balance between regulating emotions and making tough
choices. Journal of Personality, 72, 1395-142.
72

Löckenhoff, C. E., & Carstensen, L. L. (2007). Aging, emotion, and health-related decision
strategies: Motivational manipulations can reduce age differences. Psychology and
Aging, 22, 134–146.
Mata, R. & Nunes, L. (2010). When less is enough: Cognitive aging, information search, and
decision quality in consumer choice. Psychology and Aging, 25, 289-298.
Mather, M. (2006). A review of decision-making processes: Weighing the risks and benefits of
aging. In L. L. Carstensen & C. R. Hartel (Eds.) When I’m 64, (pp. 145– 173).
Washington, DC: The National Academies Press.
Mather, M., Mazar, N., Gorlick, M. A., Lighthall, N. R., Burgeno, J., Schoeke, A., Ariely, D.
( ). Risk preferences and aging: The “certain ty effect” in older adults’ decision
making. Psychology and Aging, 27¸ 801-816.
Mather, M., & Carstensen, L. L. (2003). Aging and attentional biases for emotional faces.
Psychological Science, 14, 409-415.
Mather, M., & Carstensen, L. L. (2005). Aging and motivated cognition: The positivity effect in
attention and memory. Trends in Cognitive Sciences, 9, 496-502.
Mather, M., & Johnson, M. K. (2000). Choice-supportive source monitoring: Do our decisions
seem better to us as we age? Psychology and Aging, 15, 596-606.
Mather, M., & Knight, M. (2005). Goal-directed memory: the role of cognitive control in older
adults’ emotional memory. Psychology and Aging, 20, 554-570.
Mikels, J. A., Reed, A. E., & Simon, K. I. (2009). Older adults place lower value on choice
relative to young adults. Journal of Gerontology, Series B: Psychological Sciences and
Social Sciences, 64, 443-446.
73

Mroczek, D. K., & Kolarz, C. M. (1998). The effect of age on positive and negative affect: A
developmental perspective on happiness. Journal of Personality and Social Psychology,
75, 1333-1349.
Petrican, R., Moscovitch, M., & Schimmack, U. (2008). Cognitive resources, valence, and
memory retrieval of emotional events in older adults. Psychology and Aging, 23, 585 –
594.
Piazza, J. R., & Charles, S. T. (2006). Mental health and the baby boomers. In S. K. Whitbourne,
& S. L. Willis (Eds.), The Baby Boomers at Midlife: Contemporary Perspectives on
Middle Age, (pp. 111-146). Hillsdale, NJ: Erlbaum.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general
population. Applied Psychological Measurement, 1, pp. 385–401.
Ratner, R. K., & Herbst, K. C. (2005). When good decision have bad outcomes: The impact of
affect on switching behavior. Organizational Behavior and Human Decision Processes,
96, 23-37.
Reed, A. E., & Carstensen, L. L. (2012). The theory behind the age-related positivity effect.
Frontiers in Psychology, 3, 339-347.
Reed, A. E., Mikels, J. A., & Simon, K. I. (2008). Older adults prefer less choice than young
adults. Psychology and Aging, 23, 671 – 675.
Roese, N. J. (1997). Counterfactual thinking. Psychological Bulletin, 121, 133-148.
Rösler, A., Ulrich, J. ., illino, J., Sterzer, ., Weidauer, S., ernhardt, … Kleinschmidt. A.
(2005). Effects of arousing emotional scenes on the distribution of visuospatial attention:
changes with aging and early subcortical vascular dementia. Journal of Neurological
Sciences, 229-230, 109-116.
74

Samanez-Larkin, G. R., & Carstensen, L. L. (2011). Socioemotional functioning and the aging
brain. In J. Decety & J. T. Cacioppo (Eds.), The handbook of social neuroscience. (pp.
507-521). New York, NY: Oxford University Press.
Scheibe, S., & Carstensen, L. L. (2010). Emotional aging: recent findings and future trends.
Journal of Gerontology, Series B: Psychological Sciences and Social Sciences, 65, 135-
144.
Smith, B. L., Wasowicz, J., & Preston, J. (1987). Temporal characteristics of the speech of
normal elderly adults. Journal of Speech & Hearing Research, 30, 522-529.
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the state-trait anxiety
inventory. Palo Alto, CA: Consulting Psychologists Press.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. (5
th
ed). Boston, MA:
Allyn and Bacon.
Tanius, B. E., Wood, S., Hanoch, Y., & Rice, T. (2009). Aging and choice: Applications to
Medicare Part D. Judgment and Decision Making, 4, 92-101.
Taylor, S. E., Helgeson, V. S., & Reed, G. M. (1991). Self-generated feelings of control and
adjustment to physical illness. Journal of Social Issues,47,91 – 109.
Timmer, E., Westerhof, G. J., & Dittmann-Kohli, F. ( 5). “When Looking ack on My
ast Life I Regret…”: Retrospective Regret in the Second alf of Life. Death Studies,
29, 625-644.
V stf ll, ., eters, E., lkebring, . ( ). The experience and regulation of regret across
the adult life span. In I. Nyklicek, A. Vingerhoets, & M. Zeelenberg (Eds.), Emotion
Regulation and Well-Being, (pp. 165-180). New York: Springer.
75

Watson, D., Clark, L. A., & Tellegen, A. (1988). "Development and validation of brief measures
of positive and negative affect: The PANAS scales." Journal of Personality and Social
Psychology, 54, 1063-1070.
Wechsler, D. A. (1997a). Wechsler Adult Intelligence Scale (3rd ed.). San Antonio, TX: The
Psychological Corporation.
Wechsler, D. A. (1981). Wechsler Adult Intelligence Scale—Revised. New York: Psychological
Corporation.
Wrosch, C., Bauer, I., & Scheier, M. F. (2005). Regret and quality of life across the adult life
span: The influence of disengagement and available future goals. Psychology and Aging,
20, 657-670.
Wrosch, C., & Heckhausen, J. (2002). Perceived control of life regrets: Good for young and bad
for old adults. Psychology and Aging, 17, 340-350.
Yoon, C., Cole, C. A., & Lee, M. P. (2009). Consumer decision making and aging: Current
knowledge and future directions. Journal of Consumer Psychology, 19, 2-16.
Zanov, M., & Davison, G. C. (2010). A conceptual and empirical review of 25 years of cognitive
assessment using the articulated thoughts in simulated situations (ATSS) think-aloud
paradigm. Cognitive Therapy and Research, 34, 282-291.
Zeelenberg, M. (1999). Anticipated regret, expected feedback and behavioral decision-making.
Journal of Behavioral Decision Making, 12, 93-106.

76

Table 1:

Table of Conditions

Age

Experimental Condition

Younger

No Choice

Choice

Older

No Choice

Choice

77

Table 2:

Participant Characteristics at Baseline

Young
(N=101, 67F)

Old
(N=77, 54F)

Measure M SD M SD F(1,176) p
Years of Education 14.65 1.09 16.42 2.27 46.67 <.001
Self-rated Health 4.02 .84 3.68 .98 6.39 .01
CES-D 13.96 8.28 10.34 8.21 8.43 .004
STAI –Trait 40.05 9.70 32.57 9.10 27.39 <.001
Future Time
Perspective
38.67 5.92 30.71 8.64 53.04 <.001
MMSE 29.19 .91 28.65 1.19 11.69 <.01
Digit Span 12.32 1.69 11.74 2.06 4.20 .04
WTAR 41.17 5.37 41.70 6.74 .35 .56

78

Table 3:

Age Differences in Perceptions of Air Travel and ATSS scenario

Young

Old

Measure M SD M SD F(1,176) p
Before Scenario
Degree of flight anxiety 1.53 .72 1.73 .98 2.29 .13
Last flown? 2.27 1.22 2.51 1.33 1.55 .21
Disappointed to miss 6.24 .87 6.10 1.48 .68 .41

After Scenario
How realistic was scenario? 8.03 1.57 8.26 2.06 .68 .41
Similarity to past situation? 8.34 1.42 8.68 1.78 1.99 .16
How responsible did you feel? 5.90 2.60 4.99 2.79 5.00 .03*

79

Table 4:

Manipulation Check and External Validity of Scenario Conditions

Choice
(N = 86)
Control
(N = 92)

Measure M SD M SD F(1,176) p
How responsible did you feel? 6.45 2.39 4.63 2.71 22.15 <.001
PANAS Negative (T2 – T1) 7.69 9.88 7.86 9.73
PANAS Positive (T2 – T1) -2.05 6.46 -1.55 7.15
How realistic was scenario? 8.21 1.96 8.06 1.64 .31 .58
Similarity to past situation? 8.44 1.76 8.52 1.43 .11 .74

80

Table 5:

Means, SDs, Estimated Marginal Means, and SEs for Younger and Older adults’ Expressions of
ATSS coded emotions, Coping Language, and Likert-rated Regret by condition

Condition

Choice Control

DV Age Group M SD Est M SE M SD Est M SE

Regret Young .29 .24 .28 .03 .22 .22 .21 .03
Old .15 .19 .16 .03 .09 .15 .10 .03
AngerO Young 1.06 .53 1.04 .08 1.18 .51 1.11 .08
Old .58 .45 .66 .09 .78 .61 .83 .09
Anxiety Young .65 .42 .63 .06 .66 .40 .61 .06
Old .29 .26 .35 .07 .52 .44 .56 .06
Sadness Young .25 .20 .24 .03 .24 .25 .23 .03
Old .14 .18 .15 .04 .16 .23 .17 .04
P-E Coping Young -.28 .51 -.31 .09 -.05 .61 -.07 .08
Old -.57 .57 -.54 .10 .16 .65 .19 .10
Self-rated
Regret
Young 6.51 1.42 6.36 .26 6.12 1.81 5.97 .25
Old 5.05 2.00 5.23 .29 5.71 1.92 5.91 2.86

N no/yes % presence N no/yes % presence
Guilt Young 11 / 38 77.6% 35 / 17 32.7%
Old 18 / 19 51.4% 36 / 4 10%
AngerS Young 18 / 31 63.3% 33 / 19 36.5%
81

Old 23 / 14 37.8% 32 / 8 20%

Note: AngerO represents anger at others; AngerS represents anger at the self. Guilt and AngerS
are presented as N who did not express emotion / N who did express emotion with percentage of
each group who expressed emotion. Estimated means are means after baseline depression (CES-
D) and anxiety (STAI-T) scores are included as covariates, as well as word count for angerO,
anxiety, and sadness.

82

Table 6:

Results for ATSS Responses and Self-rated Regret (N = 178)

Effect

Dependent
Variable

F (or stated
otherwise)


p
2

or
Exp(B)

Unadjusted
p

Holms
Bonferroni
Adjusted p

Age
(young, old)

Regret 11.15 .06 .001 .001 x 4 .004*
Guilt Wald =8.72 3.29 .003 .003. x 3 .009*
Anger at Self Wald = 4.72 2.17 .03 .03. x 1 .03*
Anger at Others 12.79 .07 <.001 .001. x 5 .005*
Anxiety 5.87 .03 .016 .016. x 2 .032*
Sadness 4.01 .02 .047* --
Self-rated regret 4.24 .02 .04* --
Condition
(choice, control)

Regret 4.94 .03 .028 .028 x 1 .028*
Guilt Wald = 33.04 8.71 <.001 .001 x 3 .003*
Anger at Self Wald = 9.64 2.75 .002 .002 x 2 .004*
Anger at Others 2.34 .01 .13 --
Anxiety 2.45 .01 .12 --
Sadness .01 .00 .93 --
Self-rated regret .30 .00 .58 --
Age X Condition
Interaction

Regret .07 .00 .79 .79 x 1 .79
Guilt Wald = .09 .04 .76 .76 x 2 1.52
Anger at Self Wald = .12 1.25 .74 .74 x 3 2.22
Anger at Others .38 .00 .54 --
Anxiety 4.09 .02 .045* --
Sadness .25 .00 .62 --
Self-rated regret 4.17 .02 .04* --

Note: Guilt and anger at self are analyzed using logistic regression; all other dependent variables are analyzed
using 2x2 Between Subjects ANOVA.
83

Table 7:

Results for ATSS-coded Coping Language

Dependent Variable

Post-hoc Analysis

Effect

F


p
2

P

Coping
(Problem – Emotion)

Age X Choice 7.68 .04 .006*
Age .02 .00 .89
Choice 29.69 .15 <.001*
Problem-focused Age X Choice 4.93 .03 .03*
Age 5.79 .03 .02*
Choice 13.78 .08 <.001*
Emotion-focused Age X Choice 4.50 .03 .04*
Age .54 .00 .46
Choice 13.34 .07 <.001*

Note: Coping measure is calculated by subtracting mean emotion-focused coping from mean problem-
focused coping score for each participant to create a P-E difference score. Post-hoc analyses examined
problem-focused and emotion-focused coping language separately.
84

Figure 1:
ATSS-coded Expressed Regret for Younger and Older Adults by Condition

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Control Choice
ATSS Coded Regret
Condition
Young
Old
85

Figure 2:
ATSS-coded Expressed Anger at Others for Younger and Older Adults by Condition

0
0.2
0.4
0.6
0.8
1
1.2
1.4
Control Choice
ATSS Coded Anger at Others
Condition
Young
Old
86

Figure 3:
ATSS-coded Expressed Anxiety for Younger and Older Adults by Condition

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Control Choice
ATSS Coded Anxiety
Condition
Young
Old
87

Figure 4:
ATSS-coded Expressed Sadness for Younger and Older Adults by Condition

0
0.05
0.1
0.15
0.2
0.25
0.3
Control Choice
ATSS Coded Sadness
Condition
Young
Old
88

Figure 5:
ATSS-coded P-E Coping Language for Younger and Older Adults by Condition

-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
ATSS Coded P - E focused Coping
Condition
Young
Old
Control Choice
89

Figure 6:
ATSS-coded Emotion-focused Coping Language for Younger and Older Adults by Condition

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Control Choice
ATSS Coded Emotion-focused Coping
Condition
Young
Old
90

Figure 7:
ATSS-coded Problem-focused Coping Language for Younger and Older Adults by Condition

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Control Choice
ATSS Coded Problem-focused Coping
Condition
Young
Old
91

Figure 8:
Self-rated Feelings of Regret for Younger and Older Adults by Condition (Likert rated)

3
3.5
4
4.5
5
5.5
6
6.5
7
Control Choice
Self-rated Feelings of Regret
Condition
Young
Old
92

Appendix A: Flow chart of study procedure
Read through and agree to participate:
Information Sheet

Questionnaires/Measures to complete:
Demographic Information
PANAS (1
st
administration)
CES-D
STAI - trait
MMSE
Digit Span
WTAR
FTP

ATSS portion:
Collect information on an important event from participant
Provide instructions and administer practice scenario
Participants assigned to either choice or no choice condition:

Choice Condition No Choice Condition
Participant
choice
Alternative Flight Existing Flight Existing Flight

Outcome: Outcome: Outcome:
Delayed Flight Delayed Flight Delayed Flight
|_______________________|___________________________|

Post-ATSS measures to complete:
PANAS (2
nd
administration)
Post-scenario questionnaire
Manipulation Check / Debriefing questionnaire

93

Appendix B: Event Description

We would like you to think of an important and meaningful event that you would really want to
attend. The event should be something special to you and one that may require you to travel a
short distance by plane. Think of an event that may only occur once in your lifetime and one that
you can imagine involving two friends or family members. Examples include a family member’s
wedding, a close friend’s graduation, or a special type of reunion.

Event: ________________________________________________________________________
Location:______________________________________________________________________
Travel Companions______________________________________________________________

Please indicate how disappointed would you be if you were not able to attend this actual event?
(circle one)

1 2 3 4 5 6 7

Not at all Moderately Extremely
Disappointed Disappointed Disappointed

94

Appendix C: Instructions for the ATSS scenario
You are participating in a study of people’s thoughts and feelings while in a situation
involving air travel. Often, when people are going about their daily affairs, interacting with
others and so forth, they have a kind of internal monologue going through their heads, a constant
stream of thoughts or feelings, which reflect their reactions to something, which is happening.
What we’d like you to do is to play a part in a situation that we have taped. You will be
listening to an audio recording. Please imagine that this situation is unfolding right now and that
you are a part of it. Every so often the recording will stop, you will hear a tone, and you will be
asked to speak into a microphone for 30 seconds. Simply say out loud whatever is going through
your mind. Say as much as you can until you hear another tone. Of course, there are no right or
wrong answers, so please just say whatever comes to mind without judging whether it is
appropriate or not. The more you can tell us, the better.
Try to imagine as clearly as you can that it is really you in the situation right now. Note
that your task is not to speak back to any of the voices on the tape, as though you were having a
conversation with one of them. Rather, you should tune in to your own thoughts and say them
out loud. Everything that you say will be completely confidential. Your name will not be
associated with the recording in any way.

95

Appendix D: Scripts for choice and no choice (control) conditions
ATSS choice scenario

Segment 1:
[Narrator] Imagine that it is 5pm and you are at the airport right now with two companions. The three of you have
tickets to fly to the special event that starts later tonight. The three of you have been looking forward to this event
for some time now, and you can’t wait to see everyone else when you get there. Since all of you had things to do
earlier today, you planned the one hour flight so that you will all arrive right on time. You have all just passed
through security and are walking towards the airport monitors to look for your gate number.

Segment 2:
[Narrator] As you are looking for your gate number, you overhear a couple next to you.
[ erson ] Flight 6 … ok here we are - gate A1.
[ erson ] Oh it’s boarding in 5 minutes, perfect. And it’s right ov er there.
[Narrator] You realize that their flight is heading to the same destination as yours is and is boarding in 15 minutes,
whereas your flight is not scheduled to board for another hour. You realize that the gates are right across the way
from each other and would only take a minute to walk over. Since you only have carry-on luggage, it appears that
you would be able to change your flight if you want. You look to your travel companions, who appear unsure of
what to do. It appears there may be pros and cons to each. They ask you to make the decision for everyone of
whether to change flights or not.

Segment 3 (for choice condition only):
[Narrator] Now choose a flight for the three of you by clicking on the time and gate number.
Gate A1- 5:15pm (alternative flight) Gate A2 – 6:00pm (existing flight)

Segment 4:
[Narrator] You decide to take the (alternative / existing) flight, boarding at Gate (A1 / A2) in (15 / 60) minutes. The
three of you walk over to the gate for this flight and verify with the gate attendant.
[Attendant ] mm okay you’d all like to (change / check in) to this flight. Let’s see here, (okay we can switch you
to this flight at no extra charge. Upon switching though, you won’t be able to switch flights again due to our policy...
/ okay all of you are now checked in. Just to let you know, you won’t be able to change flights at this point since
you’ve already checked in due to our policy). Alright, you’re all set. lease have a seat until your boarding group is
called. We should be boarding… um… scheduled to board right here at Gate (A / A ) in (a little bit / a little while).
Please take a seat.

Segment 5:
[Narrator] As you are leaving the counter, you overhear two gate attendants talking behind the counter:
[Attendant 2] Well, we have a few minutes before we have to say anything, there’s no need to get people upset yet.
[Attendant 3] Yeah okay let’s ust give it a few minutes.
[Attendant ] Okay, then we’ll call the flight crew and do the update I guess.

Segment 6:
[Narrator] You and your travel companions walk around the crowded gate looking for empty seats. The air is warm
and stale. After walking around the seating area a few times, you finally find a spot by the trash cans. As you are
sitting down, the loudspeaker catches your attention.
[Attendant 1] Attention all passengers waiting at gate (A1 / A2). Your flight has been delayed. The status of your
flight will be updated in a few minutes. We appreciate your patience. Again, for passengers scheduled to leave at
(5:15pm / 6:00pm), gate (A1 / A2) your boarding time has been delayed.
[Narrator] You look around and see people shifting in their seats. You see the people traveling with you check their
watches and exchange looks.

96

Segment 7:
[Narrator] After a few minutes and no update, you see that some people have formed a line at the gate counter. You
decide to go up and check the status of the flight. As you are standing in line, you overhear a couple that has just
been helped as they walk away from the counter and towards the gate across the way.
[Person 3] Thank god we changed flights just in time. Who knows how long that flight is going to be delayed?
[ erson 4] I can’t believe seats opened up for us, we would’ve had to sit over there with everyone waiting.
[Person 3] Yeah looks like they'll be waiting awhile. Can you imagine, for just a one hour flight...

Segment 8:
[Narrator] You finally reach the front of the line and inquire about the status of your flight. The gate attendant
speaks to you:
[Attendant 1] I can’t give you a specific time at the moment. If you could ust take a seat and be patient like
everyone else. You know, you aren’t the only one delayed on this flight. The other flight going to the same
destination is booked. There’s nothing I can do for you.

Segment 9:
[Narrator] As you are heading back to your seat, you can see that both people traveling with you look unnerved and
anxious. You can tell they are worried about missing the event. You notice their eyes wandering and realize they are
looking at the gate across the way. You see that the plane you did not choose has arrived and people are already
standing in line to board. You can see the gate attendant at the other flight, gate (A2 / A1), preparing to make an
announcement:
[Attendant 4] Hello, this is an announcement for passengers on flight (20 / 60). Our plane has arrived into gate (A2 /
A1). As you can see people are de-boarding the plane now. We’d like to start boarding everyone soon and may even
be able to leave a little early since everyone has checked in.

Segment 10:
[Narrator] As you sit there in the crowded waiting area, you see the gate attendants for your flight whispering behind
the counter. They can't see you, but you are close enough where you can hear them talking.
[Attendant 1] Ugh this is going to be a long night.
[Attendant 2] Can you believe the nerve of that passenger trying to change their flight?
[Attendant 1] I know, it's so annoying when they try to get special treatment.
[Attendant] Yeah, I can't wait until my shift is over.

Segment 11:
[Narrator] Some time passes and no update regarding your flight. You decide to get up and walk around. On the way
back to your gate, you look over at the gate across the way, the one flying to your same destination. You see that
everyone has already boarded the plane. You hear the gate attendant over the loudspeaker:
[Attendant 4] Last call for passengers at Gate (A2 / A1), last call for boarding. Patty, shut the door and prepare to
leave.”
[Narrator] You see your travel companions looking the same way, and then both turn to you.

Segment 12:
[Narrator] You make eye contact with the two people you are traveling with and go back to wait in your seat. You
look at your watch and realize you will all miss the event if your plane is delayed much longer. As you watch the
flight you did not choose take off, you hear a message for your flight:
[Attendant 1] This is an announcement for passengers at gate (A1 / A2) with initial boarding time of (5:15pm /
6:00pm). Your flight has been delayed. The estimated time of departure is 10pm. We ask that you stay around the
area in case there are changes; once again, passengers at gate (A1 / A2). Your new estimated departure time is 10
o'clock. Thank you.

Note. Differences between the two choice scenarios (alternative / existing) appear in parentheses.

97

ATSS no choice (control) scenario

Segment 1:
[Narrator] Imagine that it is 5pm and you are at the airport right now with two of your friends. The three of you have
tickets to fly to the special event that starts later tonight. The three of you have been looking forward to this event
for some time now, and you can’t wait to see everyone else when you get there. Since all of you had things to do
earlier today, you planned the one hour flight so that you will all arrive right on time. You have all just passed
through security and are walking towards the airport monitors to look for your gate number.

Segment 2:
[Narrator] As you are looking for your gate number, you overhear a couple next to you.
[ erson ] Flight 6 … ok here we are - gate A1.
[ erson ] Oh it’s boarding in 5 minutes, perfect. And it’s rig ht over there.
[Narrator] You realize that your flight is right next to their gate, A2. However, your flight is not scheduled to
board until 6pm, 1 hour from now.

Segment 3:
Choice point does not exist for this scenario.

Segment 4:
[Narrator] The three of you walk over to your flight at gate A2 and verify with the gate attendant.
[Attendant ] mm okay you’d all like to check in to this flight. Let’s see here, okay all of you are now checked in.
Just to let you know, you won’t be able to change flights at this point since you’ve already checked in due to our
policy. Alright, you’re all set. lease have a seat until your boarding group is called. We should be boarding… um…
scheduled to board right here at Gate A2 in a little while. Please take a seat.

Segment 5:
[Narrator] As you are leaving the counter, you overhear two gate attendants talking behind the counter:
[Attendant ] Well, we have a few minutes before we have to say anything, there’s no need to get people upset yet.
[Attendant 3] Yeah okay let’s just give it a few minutes.
[Attendant ] Okay, then we’ll call the flight crew and do the update I guess.

Segment 6:
[Narrator] You and your friends walk around the crowded gate looking for empty seats. The air is warm and stale.
After walking around the seating area a few times, you finally find a spot by the trash cans. As you are sitting down,
the loudspeaker catches your attention.
[Attendant 1] Attention all passengers waiting at gate A2. Your flight has been delayed. The status of your flight
will be updated in a few minutes. We appreciate your patience. Again, for passengers scheduled to leave at 6:00pm,
gate A2 your boarding time has been delayed.
[Narrator] You look around and see people shifting in their seats. Your friends check their watches and exchange
looks.

Segment 7:
[Narrator] After a few minutes and no update, you see that some people have formed a line at the gate counter. You
decide to go up and check the status of the flight. As you are standing in line, you overhear a couple that has just
been helped as they walk away from the counter and towards the gate across the way.
[Person 3] Thank god we changed flights just in time. Who knows how long that flight is going to be delayed?
[ erson 4] I can’t believe seats opened up for us, we would’ve had to sit over there w ith everyone waiting.
[Person 3] Yeah looks like they'll be waiting awhile. Can you imagine, for just a one hour flight...

Segment 8:
[Narrator] You finally reach the front of the line and inquire about the status of your flight. The gate attendant
speaks to you:
98

[Attendant ] I can’t give you a specific time at the moment. If you could ust take a seat and be patient like
everyone else. You know, you aren’t the only one delayed on this flight. The other flight going to the same
destination is booked. There’s nothing I can do for you.

Segment 9:
[Narrator] As you are heading back to your seat, you can see that both of your friends look unnerved and anxious.
You can tell they are worried about missing the event. You notice their eyes wandering and realize they are looking
at the gate across the way. You see that there is another flight going to your same destination. Their plane has
arrived and people are standing in line to board. You can see the gate attendant at gate A3, the other flight
preparing to make an announcement:
[Attendant 4] Hello, this is an announcement for passengers at Gate A3. Our plane has arrived. As you can see
people are de-boarding the plane now. We’d like to start boarding everyone soon and may even be able to leave a
little early since everyone has checked in.

Segment 10:
[Narrator] As you sit there in the crowded waiting area, you see the gate attendants for your flight whispering behind
the counter. They can't see you, but you are close enough where you can hear them talking.
[Attendant 1] Ugh this is going to be a long night.
[Attendant 2] Can you believe the nerve of that passenger trying to change their flight?
[Attendant 1] I know, it's so annoying when they try to get special treatment.
[Attendant] Yeah, I can't wait until my shift is over.

Segment 11:
[Narrator] Some time passes and no update regarding your flight. You decide to get up and walk around. On the way
back to your gate, you look over at the gate across the way, the one flying to your same destination. You see that
everyone has already boarded the plane. You hear the gate attendant over the loudspeaker:
[Attendant 4] Last call for passengers at Gate A3 last call for boarding. atty, shut the door and prepare to leave.”
[Narrator] You see your friends looking the same way, and then both turn to you.

Segment 12:
[Narrator] You make eye contact with your friends and go back to wait in your seat. You look at your watch and
realize you will all miss the event if your plane is delayed much longer. As you watch the other flight take off, you
hear a message for your flight:
[Attendant 1] This is an announcement for passengers at gate A2 with initial boarding time of 6:00pm. Your flight
has been delayed. The estimated time of departure is 10pm. We ask that you stay around the area in case there are
changes; once again, passengers at gate A2. Your new estimated departure time is 10 o'clock. Thank you.

Note. Additions/alterations from the choice scenario appear in bold (deletions from choice scenario are not marked).

99

Appendix E: ATSS Coding Procedure
Please use the following 4-point Likert scale to rate categories 1-6 described below.
0 – not at all
1 – slightly/somewhat
2 – moderately
3 – very

Coding Categories:

1. Regret (refers to wishing that they had made another choice)
This coding category captures statements referring to thoughts or feelings of having made the wrong
choice. Statements indicating sorrow or remorse for an act will be included in this category.

Examples
I made the wrong choice
Damn -- I should have taken the earlier flight
I wish I made a different choice
Why oh why did I make the decision to change flights?

2. Guilt (refers more to feeling responsible for making the decision, feeling bad especially with regard to
how it affects others)
This coding category captures statements referring to the feeling of responsibility or remorse.

Examples
I feel bad making my friends go through this; I’m sorry
Maybe I’ll offer to buy my friends dinner as an apology for my bad decision
It’s my fault this happened

3. Anger/Frustration at others
This coding category captures statements referring to feelings of strong displeasure, antagonism and/or
annoyance. It also denotes frustration and resentment towards other individuals.

Examples
I’m pissed
This flight attendant is rude. So annoying
I’m furious as a result of this
Why does this always happen to me? (with no indication that they are upset at themselves)

4. Anger/Frustration at self
This coding category captures statements referring to feelings of strong displeasure towards the self. It
also denotes self-loathing and frustration towards the self. (Think of this code as “things that someone
who’s always upset and blaming themselve s would say”)

Examples
I’m so stupid
I should have known better
Why am I always doing things wrong?
I can’t get anything right

100

5. Anxiety/Distress (focus more on the “anxiety” aspect, anxiety-related distress)
This coding category captures statements referring to feelings of worry, nervousness, and unease.
Statements indicating uncertainty are also included in this category.

Examples
I am anxious until I know that we’ll be there on time
Slightly nervous at the attendant’s being unsure of something
I’m feeling anxious about my decision at this point
I wish someone would ust tell us what’s going to happen

6. Sadness/Disappointment focused outward
This coding category captures statements referring to feelings of unhappiness or sorrow. Statements
indicating feeling downhearted or dejected are also included in this category. Participants stating they are
sad about what is going on or disappointed in others should be reflected in this category. This code does
not include emotions indicating guilt or regret.

Examples
This is such a bummer!
I’m feeling resigned that I can’t do anything to change this
Damn it this sucks
Disappointed in the airline and their workers

Rate for intensity of PFcoping and EFcoping: 0=none, 1=a little/some, 2=moderate, 3=a lot

7. Problem-focused coping
This coding category captures statements referring to intentions of acting on the environment to solve the
problem (being delayed on a flight). Statements indicating a plan of action and entertaining ways to get to
the event are included in this category.

Examples
I see this as an opportunity to make a run for that flight and get on
Are there other flights with other airlines?
The flight is only an hour so maybe we can rent a car
I’m going to try and work the other gate ag ent into letting us on

8. Emotion-focused coping
This coding category captures statements indicating use of strategies to improve internal state. Statements
indicating use of emotion regulation techniques such as positive reappraisal, acceptance, and
disengagement are included in this category.

Examples
More time to hang out with my friends
It’s not the end of the world
I’ll ust have to make the best of it
It is what it is, there’s no use getting upset
Just relax
Let’s ust go to the bar and have a drink

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Age differences in jealousy: an ATSS examination of the role of attributions in the emotional reactions of older and younger adults in relationship-threatening situations

PDF

Emotional reactivity and memory biases in older and younger adults

PDF

Personalized normative feedback applied to undergraduates with problem drinking: a comparison with psychoeducation and an examination of cognitive-affective change mechanisms via the articulated ...

PDF

The role of age, mood, and time- perspective on the interpretation of emotionally ambiguous information

PDF

Measuring narcissistic cognitions in response to relationship threat using articulated thoughts in simulated situations

PDF

The articulated thoughts of heterosexual male college students in reaction to anti-gay hate speech

PDF

Mechanisms of stress effects on learning and decision making in younger and older adults

PDF

Zooming in on mindfulness and gaming: the application of a brief, standardized, zoom-based, mindfulness intervention on videogame play as compared to a treatment as usual approach

PDF

Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

PDF

Breaking it down to make it stronger: examining the role of source credibility and reference group specificity in the influence of personalized normative feedback on perceived alcohol use norms a...

Asset Metadata

Creator Feng, Michelle C. (author)

Core Title Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 06/16/2014

Defense Date 04/18/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aging,ATSS,coping,decision‐making,OAI-PMH Harvest,positivity effect,regret

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Davison, Gerald C. (committee chair), Clark, Florence (committee member), Lopez, Steven R. (committee member), Mather, Mara (committee member), Walsh, David A. (committee member)

Creator Email michelcf@usc.edu,michelle.feng@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-420328

Unique identifier UC11285928

Identifier etd-FengMichel-2550.pdf (filename),usctheses-c3-420328 (legacy record id)

Legacy Identifier etd-FengMichel-2550.pdf

Dmrecord 420328

Document Type Dissertation

Format application/pdf (imt)

Rights Feng, Michelle C.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA