University of Southern California Dissertations and Theses

Aging and emotion regulation in the judgment of facial emotion

(USC Thesis Other)

Aging and emotion regulation in the judgment of facial emotion

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

AGING AND EMOTION REGULATION IN THE JUDGMENT OF FACIAL
EMOTION

by

Jennifer L. Kellough

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

December 2009

Copyright 2009 Jennifer L. Kellough

ii

Acknowledgments
I gratefully acknowledge Bob G. Knight for his careful guidance and valuable
contributions to this paper. In addition, I am grateful to Margaret Gatz and Mara Mather
for their helpful input. The contributions of the participating adults and students are also
very much appreciated

iii

Table of Contents
Acknowledgments ii
List of Tables iv
List of Figures v
Abstract vi
Chapter 1: Introduction 1
Chapter 2: Methods 12
Chapter 3: Results 24

Chapter 4: Discussion 46
Bibliography 59

iv

List of Tables
Table 1: Demographic Measures by Age Group and FTP Induction 13
Group

Table 2: Expressions Comprising the Low, Medium, and High Levels 22
of Ambiguity for All Analyses

Table 3: Percent of Accurate Recognition of Clearly Expressed Emotions 25

Table 4: Mean Index Scores of Positive Affect by Age Group and FTP 26
Induction Group

Table 5: Mean Index Scores of Negative Affect by Age Group and FTP 27
Induction Group

Table 6: Mean Index Scores of Mixed Affect by Age Group and FTP 30
Induction Group

Table 7: Correlations for Executive Functioning and Emotion 43
Regulation in Older Adult Controls

Table 8: Correlations for Future Time Perspective and Emotion 44
Regulation in Older Adult Controls

Table 9: Correlations for Future Time Perspective and Emotion 45
Regulation in Young Adult Controls

v

List of Figures

Figure 1: Age x Ambiguity Interaction for Positive Affect in the Clearly 28
Negative, Neutral, and Morphed Expressions

Figure 2: Age x Ambiguity Interaction for Mixed Affect in the Clearly 31
Negative, Neutral, and Morphed Expressions

Figure 3: Age x Time Perspective Interaction for Positive Affect 33
Collapsed Across the Clearly Positive, Neutral, and Morphed
Expressions

Figure 4: Age x Time Perspective Interaction for Positive Affect 34
Collapsed Across the Clearly Negative, Neutral, and Morphed
Expressions

Figure 5a: Age x Ambiguity x Time Perspective Interaction for Angry 37
and Happy/Angry Morphs

Figure 5b: Ambiguity x Time Perspective Interaction in Younger Adults 37
For Angry and Happy/Angry Morphs.

Figure 5c: Ambiguity by Time Perspective Interaction in Older Adults 38
for Angry and Happy/Angry Morphs

Figure 5d: Age x Ambiguity Interaction in Controls for Angry and 38
Happy/Angry Morphs

Figure 6: Age x Ambiguity Interaction for Mixed Affect in the Angry 39
and Happy/Angry Morphs

Figure 7: Age x Ambiguity Interaction for Positive Affect in the Fear 40
and Happy/Fear Morphs

Figure 8: Ambiguity x Time Perspective Interaction for Negative Affect 41
in the Fear and Happy/Fear Morphs

Figure 9: Age x Ambiguity Interaction for Mixed Affect in the Fear and 42
Happy/Fear Morphs

Figure 10: Ambiguity x Time Perspective Interaction for Mixed Affect 42
in the Fear and Happy/Fear Morphs

vi

Abstract

Prior research has established differences between older and younger adults’ (1)
attention for emotional stimuli, (2) retrieval of emotional memories, and (3) appraisal of
their own emotional experiences. These age differences are seen as evidence of emotion
regulation, which is thought to service shifting emotional goals in later life. The current
study sought to explore emotion regulation further by assessing age differences in the
interpretation of others’ emotions. Specifically, we examined age differences in the
judgment of emotion from facial expressions. Our judgment paradigm allowed for the
evaluation of two prominent forms of emotion regulation described in the literature,
affect optimization and affect complexity. Results showed that older adults did exhibit
affect optimization and affect complexity in their judgments of the facial expressions.
Specifically, older adults perceived more positive affect and more mixed affect in the
expressions compared to younger adults. Additionally, when future time perspective was
experimentally manipulated for each age group, age differences in positive affect were
reduced and perception of negative affect increased across age groups. Findings are
discussed within the framework of developmental lifespan theories of emotion.

1

Chapter 1: Introduction
Age-related changes in emotion and cognition have become a focus of research as
our aging population grows. Early perspectives characterizing later life as a period of
declining satisfaction and well being have largely been unsupported by more recent data
(Carstensen, Pasupathi, Mayr, & Nesselroade, 2000; Charles, Reynolds, & Gatz, 2001;
Mroczek & Kolarz, 1998). Such research has revealed developmental patterns in
information processing that appear to serve as vehicles for maintaining, or even
improving, well being in later life. Much of this work has focused on how age groups
differ in attention and memory for emotional information, and in the complexity of
emotions they experience in daily life (Ersner-Hershfield, Mikels, Sullivan, &
Carstensen, 2008; Charles, Mather & Carstensen, 2003; Magai, Consedine,
Krivoshekova, Kudadjie-Gyamfi, & McPherson, 2006; Mather & Carstensen, 2003).
These age differences have been explained as representing forms of emotion regulation
that are integral for adaptive functioning in later life (Carstensen & Mickels, 2005;
Labouvie-Vief & Medler, 2002). Additional areas of research investigating age-related
changes in information processing have focused on accuracy of emotion recognition (see
Isaacowitz, et al., 2007 for a review). Although the ability to recognize specific
emotional signals accurately is crucial for successful functioning, the current literature
lacks a broader understanding of one’s judgment or interpretation of emotional
information in the environment. With the domains of attention and memory showing age-
related changes, it is also important to understand how older adults may be interpreting
emotional stimuli differently as they age and what factors influence their judgments. To

2

clarify these questions, the current study explores the effect of age on judgment of
emotion in facial expressions.
Aging, Information Processing, & Emotion Regulation
Several developmental life span theories have shed light on the seemingly
paradoxical findings of preserved well being in later life despite declines in physical
health, death of peers and spouses, and restriction of more youthful pursuits. (Carstensen,
1991, 1995; Labouvie-Vief & Blanchard-Fields, 1982; Lawton, 1996). These theories
propose that in later stages of life, a motivational shift occurs making emotional
information more salient. Carstensen’s socioemotional selectivity theory (SST) posits
that when the future is seen as expansive, interest is focused on information seeking. In
contrast, when the future is seen as more limited, emotionally meaningful goals such as
maintaining close relationships become more prominent. As a means for achieving these
new goals, people are motivated to monitor their emotions and arrange their
environments to optimize their emotional experience. In turn, this saliency or enhanced
monitoring of emotion allows for better emotion regulation. As the future naturally
becomes constrained as we age, SST suggests that emotion regulation improves with age
in service of these shifting goals.
SST has been examined in several empirical studies of information processing in
older adults. The argument driving research on attention proposes that older adults’
greater focus on regulating emotion will likely change where they direct their attention.
Mather and Carstensen (2003) examined this possibility using a dot-probe task of
attention. In this task, one emotional and one neutral face is displayed simultaneously for

3

a brief time. When the faces disappear, a dot appears behind one of the faces and the
participant must indicate the location of the dot probe. Attentional biases are then
inferred from response latencies in probe identification for the different emotional faces.
The results of this study indicate that older adults, when compared to younger adults,
demonstrate an attentional bias away from negative facial expressions. These results
were confirmed in several studies using a more robust measurement of attention
involving eye tracking technology (Rösler, et al., 2005; M. Knight, Seymour, Gaunt,
Baker, Nesmith, & Mather, 2007). Although younger and older adults showed no
differences in their sustained viewing of positive images in positive-neutral pairings,
older adults attended less to the negative images in negative-neutral pairings than did
younger adults. More recently, Allard and Isaacowitz (2008) found that older adults
demonstrated an attentional preference for positive relative to negative images, where as
younger adults’ preference was for negative relative to positive images.
Several studies of memory have revealed similar age differences on a variety of
tasks. Recent evidence shows a decrease in the recall and recognition of negative images
compared to positive images with age (Charles, Mather, & Carstensen, 2003). Older
adults in a neutral mood state recalled fewer negative and more positive words on a
delayed recall task than did younger adults (B. Knight, Maines, & Robinson, 2002).
Additionally, in a study of autobiographical memory, older adults tended to reappraise
negative events in a more positive light (Comblain, D’Argembeau, & Van der Linden,
2005). These changes in information processing, frequently referred to as the positivity
effect or affect optimization, are purported to allow for the maximization of positive

4

affect and minimization of negative affect in older adults. Mather and Knight (2005)
have further elaborated on these age differences in emotional memory and found that
executive functioning is an important factor in the emergence of the positivity effect.
Specifically, their study revealed that older adults draw on cognitive control processes to
make their memories more positive, whereas younger adults showed no signs of using
cognitive resources to modulate the emotional content of their memories. Furthermore,
they found that older adults who demonstrated higher cognitive control abilities exhibited
the positivity effect to a greater degree in their memory for emotional pictures.
Although current evidence suggests that older adults are regulating their emotions
by modifying (1) what they attend to and (2) how they selectively retrieve events in
memory, relatively little attention has been paid to possible age differences in how older
adults judge emotional information in their environment. Given that attention and
memory have been influenced by motivational shifts in aging, it is reasonable to consider
that the nature and range of emotions interpreted from a stimulus may also vary with age.
Just as attention allocation influences what gets encoded from the environment, changes
in judgment can influence the types of emotions encoded from the environment. This
may potentially reveal yet a further domain where emotion regulatory processes could be
operating. Another area of the life span development literature proposes an additional
mechanism by which emotional judgments may change with advancing age.
In addition to affect optimization, the existing literature describes a second form of
emotion regulation. Affect complexity has been described as the ability to coordinate
positive and negative affect into flexible and differentiated structures, which may present

5

as a mix of positive and negative emotions (Labouvie-Vief, 2000). Other researchers
have defined affect or emotional complexity as an increase in range of emotions or co-
occurrences of positive and negative affective states (Carstensen, et al., 2000;
Chipperfield, Perry, & Weiner, 2003; Ong & Bergeman, 2004). Carstensen and her
colleagues contend that the pursuit of emotionally meaningful goals, which becomes a
primary task in later life, necessitates more complex emotional states. Researchers have
further suggested that interpretations of emotional experience that include both positive
and negative emotions may be a feature of adaptive emotional functioning (Carstensen, et
al., 2000; Labouvie-Vief & Medler, 2002; Mayer & Salovey, 1997; Ong & Bergeman,
2004). Labouvie-Vief and Medler (2002) argue that when affect optimization and affect
complexity work in concert, they result in highly effective emotion regulation and in turn
improved well being.
Studies involving daily ratings of emotional experience have shown that older
adults demonstrate more complex mixes of emotions, as well as reduced frequency of
negative affect (Carstensen, et al., 2000; Ong & Bergeman, 2004). Using an experience-
sampling procedure that allowed participants to characterize their daily experiences by
endorsing an unlimited number of emotional descriptors from a list of 19 emotions,
Carstensen, et al. (2000) found differences in affect optimization and affect complexity
by age, and found these differences to be associated with a positive profile of
characteristics. Specifically, they found that age was associated with the frequency of
experiencing negative affect, such that frequency decreases from 18 to 60 years and then
ceases its downward trend from age 60 onward. Results showed that more dimensions or

6

factors are required to reflect the structure of older compared to younger adults’
emotions, as defined by within-person factor analyses. This reduction in negative affect
and increase in differentiation, as labeled by Carstensen and colleagues, was further
associated with less neuroticism and better emotional control.
Ong and Bergman (2004) confirmed these findings in their investigation of
individual difference variables that predict affect complexity using a sample of only older
adults. Using similar analytic techniques, this study also found reduced frequency of
negative affect as well as increased complexity or differentiation in emotional experience
of their older adult sample. In addition, Ong and Bergman also found a greater degree of
co-occurrence of positive and negative feeling states over the 30-day sampling period,
evidenced by a positive within-subjects correlation between positive and negative feeling
states. Furthermore, they found that individuals who were higher in emotional
complexity also showed higher levels of resilience, and lower levels of both neuroticism
and psychological stress. These studies of emotional experience are the few
investigations whose methodologies have allowed for the observation of affect
optimization and complexity simultaneously, and have uniquely displayed the two forms
of emotion regulation at work. Furthermore, they provide evidence that both
optimization and complexity are associated with adaptive characteristics and functioning.
While these studies suggest that affect optimization and complexity are being
utilized at a reflective stage of emotional evaluation of life experiences, a question
emerges as to whether these regulatory processes might be implemented at earlier stages
of an experience, such as the beginning of a social interaction. Given that older adults

7

exhibit signs of emotion regulation when appraising their own emotions, it seems
reasonable to hypothesize that they might use similar processes in the appraisal of others’
emotions. The present literature has explored the related topic of accuracy of emotion
recognition in older adults as one meter of their judgment of emotions encountered in the
environment.
Aging and Emotion Recognition
The current literature provides us with a wealth of information on how well older
adults can accurately recognize specific emotions. This literature was driven by the
important question of whether our ability to accurately recognize emotional stimuli
remains intact as we age. The vast majority of these studies have been conducted
examining emotion recognition in posed facial expressions of specific emotions. For
example, these evaluations consisted of correctly classifying an expression as “happy”
when the stimulus depicts a happy face, or incorrectly classifying it by responding with
any other emotion when the stimulus depicts a happy face (Calder, et al., 2003;
McDowell, Harrison, & Demaree, 1994; Moreno, Borod, Welkowitz, & Alpert, 1993;
Phillips, MacLean, & Allen, 2002; Sullivan & Ruffman, 2004). In general, participants in
these studies viewed faces expressing anger, disgust, fear, happiness, sadness, or surprise
and were instructed to select one emotion from those listed that best describes the face.
The number of emotions the participants had to choose from was limited to the specific
emotions being tested in the study, which typically ranged from four to six emotions.
Although some variability in the results exists, most prior evidence suggests
consistent age differences such that older adults demonstrate less accuracy in recognizing

8

the negative emotions of anger, sadness, and fear relative to young adults. Although age
differences were typically not found in the recognition of happiness and surprise, ceiling
effects were present in the recognition scores of happiness in over half of the studies.
Most recently, Isaacowitz et al. (2007) examined accuracy of emotion recognition in
lexical stimuli as well as facial stimuli. This study included an assessment of emotionally
neutral stimuli and error analyses to address the possibility of emotion-specific response
biases. Specifically, results showed that older adults were less accurate than younger
adults in recognizing facial expressions of fear, anger, and happiness, and lexical stimuli
depicting emotions of sadness, disgust, anger, happiness, surprise, and neutral emotion.
Taken together, this body of research has demonstrated a small but consistent age
difference in the ability to recognize a variety of emotions.
This research informs us on the accuracy of judgments made by older adults when
choices are limited and selection is constrained to a single emotion. Although this
provides us with information concerning the stability of emotion recognition across the
life span, it does not inform us as to the true nature and complexity with which we
interpret emotional information. The developmental life span literature suggests that we
adjust our processing of emotional information to serve changing goals, and that we
experience emotions more complexly than forced choice paradigms allow for. Therefore,
the next important step is to examine how these changes might emerge in our
interpretations of emotion in the environment. By assessing judgment of emotion in an
unconstrained manner, we can capture a more naturalistic view of how these processes

9

are affected by age, while furthering developmental life span theories by examining a
domain of emotion processing that has yet to be explored.
The Current Study
Prior research provides strong evidence of age-related changes in attention and
memory in the processing of emotional information. These changes have been explained
in the context of life span theories that suggest the observed changes are serving new
emotional goals targeted at regulating emotion. Additional differences by age have been
found in the complexity and affective valence of emotions used to describe one’s own
emotional experiences. However, little is known regarding how these new goals
influence the judgment of emotion from one’s social environment. Evidence concerning
accuracy of emotion recognition under forced choice conditions provides us with a
limited view of judgment of emotion from facial expressions. In light of the saliency of
emotion judgment in interpersonal relations and the importance of these relations in later
life, an evaluation of differences in emotion judgments by age is warranted.
The proposed study seeks to explore age differences in the judgment or
interpretation of emotion from facial expressions. Facial expressions have been chosen,
as they are arguably one of the most relevant social cues in the environment, and their
interpretation can be variable. Our assessment will allow for naturalistic judgments to be
measured by allowing participants to select any number of emotions they infer from the
stimulus from a broader list of emotions and feeling states than has been used in previous
research. In an effort to replicate the emotional variety and lack of clarity that
characterizes much of what one encounters in daily life, judgment of emotion will be

10

assessed from facial expressions depicting clearly expressed emotions as well as faces
portraying more ambiguous emotions. As emotional cues in the environment are
frequently unclear to the perceiver, it is critical to assess how one judges these cues. This
unique assessment will permit the evaluation of both affect optimization and affect
complexity in younger and older adults’ judgments of the clear and ambiguous facial
expressions. As the main tenet of SST emphasizes perceived time left in life as the
driving factor in the shift to pursue emotionally meaningful goals in later life, we will
experimentally manipulate future time perspective in both older and younger adults to
test whether it influences their judgments of the facial expressions. Additionally, as
existing research has shown that older adults draw on executive functions to strengthen
positive and diminish negative information in memory (Mather & M. Knight, 2005), we
will measure executive functioning to explore its relation to affect optimization and
complexity in the older adult participants.
Hypotheses
Hypothesis 1. On the basis of developmental life span theories and their accounts
concerning increased emotion regulation in later life, we predict a main effect for age
such that older adults will demonstrate greater affect optimization and complexity in their
judgment of emotion in facial expressions compared to younger adults. Affect
optimization will be represented by increased frequency of positive emotions and/or
reduced frequency of negative emotions; affect complexity will be represented by greater
frequency of co-occurrence of positive and negative emotions,

11

Hypothesis 2. However, because ambiguity in emotional meaning allows for
greater flexibility in judgment, we predict an interaction by age and stimulus type.
Specifically, we expect older adults to display affect optimization and complexity to a
greater degree in their judgment of emotion in the ambiguous facial expressions as
compared to the clearly expressed emotions.
Hypothesis 3. As previous research has shown younger adults to have a more
expanded time perspective and older adults to have more limited time perspective, we
predict that our manipulation (limited perspective for younger adults and expanded
perspective for older adults) will result in a reduction of the age differences in affect
optimization and complexity.
Hypothesis 4. Because emotion regulation has been tied to higher-order cognitive
processes in older adults, we anticipate our measure of executive functioning to be
positively correlated with affect optimization and complexity in the older adult group.
Hypothesis 5. In accordance with SST, we predict that future time perspective
will be related to our measures of emotion regulation. As lower scores on the future time
perspective scale reflect more limited time perspective, we predict that future time
perspective will be negatively correlated with emotion regulation.

12

Chapter 2: Methods
Participants
Younger adult participants (n = 127) were recruited from undergraduate
psychology classes at the University of Southern California (USC). Young adults ranged
in age from 18 to 24 (M = 19.92, SD = 1.38) with 71 percent of them being female.
Thirty-six percent were White (non-Latino), 39 percent were Asian American, 9 percent
were African American, 7 percent were Latino, and 9 percent were from other ethnic
backgrounds. Young adult participants received extra course credit for participation in
the study.
Older adult participants (n = 111) were recruited from the USC Emeriti Center,
the USC Andrus Gerontology Center volunteers, and the USC Healthy Minds research
volunteers. Older adults ranged in age from 64 to 91 (M = 75.88, SD = 6.44). Twenty-
nine percent of the older adult participants were female, and most (94%) were White
(non-Latino). All participants received detailed information regarding the nature of the
study and their participation via standard mail or electronic mail. Interested participants
responded to the invitation via telephone or e-mail at which point the investigator ensured
the participant understood what would be expected of them and scheduled an
appointment for the participant. English language proficiency was determined to be
adequate based on the participant’s ability to understand and respond to the initial
invitation, as well as the subsequent communications regarding their participation and
scheduling matters.

13

The groups differed significantly with regard to gender, !
2
(1, N = 237) = 40.65, p
< .001, ethnicity, !
2
(4, N = 237) = 79.73, p < .001, and education, t(235) = 12.42, p <
.001. Older and younger adults did not differ significantly with regard to self-reported
health, t(235) = 0.61, ns. See Table 1 for means of demographic information by age
group.
Table 1: Demographic Measures by Age Group and FTP Induction Group.
Note. CES-D = Center for Epidemiologic Studies Depression Scale; FTP = Future Time Perspective Scale.

Measures
Demographic Information. Personal information on age, gender, ethnicity, education,
and self-reported health was obtained from the participants.

Age Group
Younger Older

Control
(n = 64)
Manipulation
(n = 63)
Control
(n = 58)
Manipulation
(n = 53)
Demographic M SD M SD M SD M SD
Age (years) 19.87 1.20 19.97 1.54 76.03 6.53 75.72 6.40
Education (years) 14.81 1.16 14.76 1.30 17.28 2.18 17.91 2.10
Self-rated health 8.51 1.03 8.57 1.00 8.18 1.57 8.74 1.02
CES-D 12.06 7.14 12.48 8.12 6.47 5.54 6.94 6.15
FTP 40.23 4.93 40.54 5.19 32.11 8.89 33.20 7.94

14

Executive Functioning Measures.
Verbal Fluency. The Controlled Oral Word Association Test (COWAT; Benton
& Hamsher, 1976, 1989) and Animal Naming are measures of phonemic and semantic
fluency that were used as a measurement of executive functioning. Participants were
asked to write down as many different words as possible that begin with the letters F, A,
and S in 60 second increments for each letter. They were also asked to write down as
many animals as possible in 60 seconds. This test shows a strong relationship to general
intellectual ability and has been used extensively with older adults (Lindenberger, Mayr,
& Kliegl, 1993). Correct responses across all three letters (F, A, and S) were summed for
the measurement of phonemic fluency. Correct responses for Animal Naming were
summed for the measurement of semantic fluency.
Design Fluency. The Figural Fluency Test (Ruff, Light, & Evans, 1987) is a non-
verbal analogue to word fluency tasks and was also used as a measure of executive
functioning. The task requires participants to generate as many different designs as
possible based on a set of rules. Designs were generated by connecting dots provided in
an array on a sheet of paper. There were three trials of design making that involved a
slightly different set of rules for each trial. Participants had 60 seconds for each trial to
make as many designs as possible. Unique designs that adhered to the rules were given 1
point. Points were summed across the three trials for the final score.
Raw scores from each of the three tasks were converted to Z scores. Each
participant’s three Z scores were summed to form a composite score of executive
functioning, which was used to address Hypothesis 4. Additionally, the composite score

15

was used as an exclusion criterion; any participant scoring 3 standard deviations below
the mean of their age group would be excluded from the study. However, no participant
met this criterion, thus no one was excluded from the analysis on this basis.
Mood Measure. The Center for Epidemiological Studies—Depression Scale
(CES-D; Radloff, 1977) was used to assess the presence of depressive symptoms. The
CES-D is widely used in research with adults of all ages, and its reliability and validity
have been well established (Radloff, 1977; Radloff & Teri, 1986). The internal
consistency for this measure in the current study was ! = .78 for the older adult group
and ! = .86 for the young adult group. The CES-D asks individuals to indicate on a 4-
point scale, with responses ranging from 0 (rarely or none of the time) to 3 (most of the
time), how frequently they experienced certain symptoms within the past week. Items 4,
8, 12, and 16 were reversed scored and item scores were summed to create a total score,
with higher scores indicating higher frequency of depressive symptoms. This measure
was used to examine the potential influence of mood on the judgment of emotion.
Younger adults reported significantly more symptoms of depression compared to older
adults: for younger adults, M = 12.27, SD = 7.62; for older adults, M = 6.69, SD = 5.82;
t(235) = 6.27, p < .001.
Future Orientation. The Future Time Perspective Scale (FTP; Carstensen &
Lang, 1996) is a 10-item measure assessing perceived limitations on time. The FTP asks
participants to rate on a scale from 1 (very strongly) to 7 (not at all) the degree to which
they agreed with each of 10 items. Sample items are “Many opportunities await me in the
future,” “Most of my life (still) lies ahead of me,” “I have the sense that time is running

16

out,” “As I get older, I begin to experience that time is limited,” and “My future seems
infinite to me.” The internal consistency of this measure has ranged from ! = .76 to .92
in prior studies with adults of all ages (Fung, Lai, & Ng, 2001; Lang & Carstensen, 2002;
Yeung, Fung, & Lang, 2007). Internal consistency of this measure in the current study
was ! = .89 for the older adult group and ! = .78 for the young adult group. A total score
was created by summing the ratings on all items focusing on an expansive future and the
reverse scores of items focusing on a more limited future. Higher scores indicated
perception of the future as more expansive relative to lower scores. As expected,
younger adults scored significantly higher on the FTP than older adults indicating a more
expanded time perspective compared to older adults: for younger adults, M = 40.38, SD =
5.04; for older adults, M = 32.63, SD = 8.43; t(235) = 8.55, p < .001
Based on Cate and John’s (2007) factor analysis revealing a two-factor model of
future time perspective, we explored the FTP subscales of “focus on opportunities” and
“focus on limitations” as well as the total scale in our examination of Hypothesis 5. The
focus on opportunities subscale consists of items 1, 2, 3, 6, 7, 8, and 9, while the focus on
limitations subscale consists of items 4, 5, and 10. The subscale scores were calculated
so that higher scores reflected a greater focus on opportunities and limitations
respectively.
Stimuli
Clearly Expressed Emotions. The following four clearly expressed emotions
were evaluated in this study: happiness, sadness, anger, and fear. Images of facial
expressions of these emotions were selected from the collection titled Pictures of Facial

17

Affect (Ekman & Friesen, 1976). These photo stimuli are well standardized for their
emotional content and have been used extensively in research with young and older
adults (e.g. Calder, et al., 2003). Photos of four different actors (2 males, 2 females)
posing each of the four emotions were selected from this collection.
Ambiguous Emotions. Two different types of facial stimuli were used for the
assessment of ambiguous emotions. We selected the neutral faces from the Pictures of
Facial Affect collection as one type of ambiguous facial stimuli. Although these
expressions were intended to be non-emotional, they have not been normed for neutrality
by the creators. On the contrary, the standardization procedure used by Ekman and
Friesen (1976) demonstrates the frequency with which individuals judged the neutral
faces as emotional. Moreover, the standardization shows the variability of the emotions
endorsed for these neutral faces, with most of the faces being rated as potentially
expressing any of six possible emotions. These data support the notion that these
expressions are not only emotional, but ambiguous as well. Additionally, these “neutral”
faces have been used to represent ambiguous expressions of emotion in previous research
(Chepenik, Cornew, & Farah, 2007; Meyer, Pilkonis, & Beevers, 2004; Yoon & Zinbarg,
2007). Photos of neutral expressions from twelve different actors, balanced for gender,
were selected for this purpose.
In addition to the neutral faces, images of two specific emotions posed by the
same actor were morphed or blended to generate another set of ambiguous emotional
stimuli. The morphs were generated using FantaMorph 4.0; this software allows for
identifying pertinent anatomical areas such as the mouth, eyes, nose, chin, and hairline to

18

use as control areas to facilitate smooth morphing. An important feature of this procedure
is that control points are shifted by an equal percentage of the total distance between their
initial and final positions. This produces images that morph in 10% increments from the
first prototype to the last prototype. The resulting images with 50% of each emotion
were used in the current study as this degree of morphing produces the most ambiguity.
The following emotional blends were used: happy/sad, happy/anger, and happy/fear.
Morphs were generated on these 3 continua for four different actors (2 male, 2 female),
yielding 12 morphed faces.
Emotion Judgment Task
The judgment task was performed via a paper and pencil format. Photographs of
facial expressions were presented one per page with a list of emotions below each
expression. Participants indicated the emotion they interpreted from the expression and
their level of confidence in that judgment on a 9-point scale ranging from 1 (not
confident) to 9 (extremely confident). Any rating of 1 to 9 indicated the presence of the
emotion and the confidence in that rating, thus capturing both in a single rating.
Participants were not limited in the number of emotions they could endorse for a single
expression. The list of emotions includes happy, calm, hopeful, serene, content, excited,
angry, fearful, sad, annoyed, bored, nervous, ashamed, embarrassed, worried, guilty, and
neutral. Additional emotions beyond the four emotions of happiness, anger, sadness, and
fear were selected in order to generate a more comprehensive list of both positive and
negative emotions. These emotions were selected to be consistent with previous studies
examining optimization and complexity in older and younger adults appraisals of daily

19

life experiences (Carstensen, et.al., 2000; Ong & Bergeman, 2004). Several “other”
blanks and corresponding confidence scales were provided on the response sheet to
accommodate other emotions participants could have interpreted from the expressions
that were not included in the list. Three independent raters coded all write-in responses
as positive, negative, or not applicable. Examples of responses coded as positive
emotions were amused, peaceful, and proud. Examples of responses coded as negative
emotions were frustrated, intimidated, and hopeless. Several participants wrote in
responses that the coders deemed not emotion terms like tired, pain, mesmerized, and
ready; responses like these were dropped from the analyses.
Procedure
Participants were tested in groups ranging from 2 to 8 individuals. The study was
conducted in quiet rooms on the USC campus. Upon their arrival, participants were
provided with a consent form approved by the USC Institutional Review Board. After
reading the consent form, participants were given the instructions for the cognitive tasks.
Participants then completed the demographics questionnaire, the CES-D, and the FTP.
After the questionnaires were completed, instructions for the judgment task were read
aloud. Young adults in the limited-time condition were given the following additional
instruction: “Imagine that you are a graduating senior, and today is the last day that you
will be a student at USC. Tomorrow you will be leaving Los Angeles and your life as a
college student is coming to a close.” Older adults in the expanded-time condition were
given the following instruction: “Imagine that last week you found out from your doctor
about a new medical advance that insures you will enjoy 20 more years beyond the age

20

you expected to live, in reasonably good health.” Both groups of participants in the
experimental condition were instructed to: “Keep this new perspective in mind while you
judge the following facial expressions.” All participants completed the judgment task
independently. No time limit was set for this task, completion time ranged from 20 to 40
minutes.
As assignment to condition was by group as opposed to by individual, we
analyzed baseline differences across the two conditions for both older and younger
adults. No significant differences were found in any of the demographic or baseline
mood and FTP measures across conditions.
Analytic Plan and Data Reduction
In order to examine our specific hypotheses, we performed repeated measures
ANOVA analyses with two between-subjects factors, age and time perspective, and one
within-subjects factor, ambiguity. The between-subjects factor of age consisted of two
levels, older and younger adults. The time perspective factor also had two levels
consisting of a control group for both younger and older adults in which their time
perspective was not manipulated, and a manipulation group where younger adults were
induced into a limited time perspective and older adults were induced into an expanded
time perspective. To begin the analyses, 3 levels of ambiguity were analyzed for the
within-subjects factor. Low ambiguity was represented by the clearly expressed facial
expressions, medium ambiguity consisted of the neutral facial expressions, and high
ambiguity consisted of the morphed facial expressions.

21

We derived indices of positive and negative affect for the 3 levels of ambiguity by
first calculating the individual frequencies of positive and negative affect terms endorsed
for each category of expression (clearly expressed, neutral, or morphed). We then
divided the frequency by the number of expressions in the corresponding category in
order to account for the varying number of stimuli in each category. For example, if a
participant had endorsed a total of sixteen positive affect terms and eight 8 negative affect
terms in their judgments of the twelve morphed expressions, their indices of positive and
negative affect for the morphed expressions would be 1.33 (16 endorsed emotions / 12
expressions) and 0.67 (8 endorsed emotions / 12 expressions) respectively. The index of
mixed affect was similarly derived by tallying the number of expressions from each
category that received endorsements of both positive and negative affect. This frequency
was then divided by the number of expressions in the corresponding category to account
for the discrepant number of stimuli per category. For example, if a participant had
endorsed both a positive and negative affect term for one of the twelve morphed faces,
their index of mixed affect would be 0.08 (1 / 12).
The analyses were conducted in two stages. The first stage of the analysis
consisted of examining the three dependent measures of positive, negative, and mixed
affect across the three levels of ambiguity. Due to results of previous research revealing
differences in the processing of positive and negative stimuli, the clear expressions of
positive and negative affect comprising the low level of ambiguity were analyzed
separately. As such, we conducted two sets of ANOVAs, first examining the positive-
clearly expressed (comprised of happy expressions), neutral, and morphed expressions

22

(comprised of all three morph types), then examining the negative-clearly expressed
(comprised of sad, angry, and fearful expressions), neutral, and morphed expressions. In
the second stage of the analysis, we conducted more emotion specific comparisons using
only two levels of ambiguity, low and high. In this analysis we used analysis of variance
to examine each clearly expressed negative emotion with its corresponding morph type
(e.g. clear sad expression with happy/sad morph). This allowed us to explore the age and
manipulation effects across the different negative emotions individually. See Table 2 for
a description of the different levels of analyses including the various expressions used in
each analysis.
Table 2: Expressions comprising the low, medium, and high levels of ambiguity for all
analyses.
Note. Each set of ANOVAs conducted on these groups of expressions included analyses of positive affect, negative
affect, and mixed affect.
a
Medium level of ambiguity was not used in the emotion-specific analyses.

Level of Ambiguity
Analysis Low Med High
Stage 1 - Positive
Happy
(N = 4)
Neutral
(N = 12)
Morphs – All Types
(N = 12)
Stage 1 - Negative
Sad, Anger, Fear
(N = 12)
Neutral
(N = 12)
Morphs – All Types
(N = 12)
Emotion-Specific:
Sad
Sad
(N = 4)
N/A
a

Happy/Sad Morphs
(N = 4)
Emotion Specific:
Anger
Anger
(N = 4)
N/A
a

Happy/Anger Morphs
(N = 4)
Emotion-Specific:
Fear
Fear
(N = 4)
N/A
a

Happy/Fear Morphs
(N = 4)

23

The results will be presented in the following order. Although accuracy of
emotion recognition is not the subject of this investigation, we will first briefly present
the results of the accuracy analyses for recognition of the four clearly expressed emotions
of happiness, sadness, anger, and fear. We will then present the results for affect
optimization (levels of positive and negative affect) and affect complexity (mixed affect)
for the comparisons of the three levels of ambiguity (Stage 1) followed by the results of
the time perspective manipulation. Then we will present the results of the emotion-
specific comparisons (Stage 2). Lastly we will report the results of the individual
difference analyses regarding executive functioning and future time perspective.
Missing Data
All participants completed the protocol according to instructions resulting in a
complete data set. Two participants’ data were excluded from the analyses due to an
improper administration of the FTP induction. All other participants’ data were included
in the analyses with no missing data.

24

Chapter 3: Results
Accuracy of Emotion Recognition for Clearly Expressed Emotions
No age differences were found in the accuracy of recognition for happy faces.
Over 90% of both younger and older adults perceived happiness in the happy
expressions, which is consistent with a number of studies on emotion recognition.
However, analyses revealed significant age differences in the accuracy of judgments for
the sad, angry, and fearful faces. Younger adults were more accurate in their judgments
of the sad and angry faces than the older adults. However, younger adults were less
accurate than older adults in their judgments of the fearful faces. Previous research on
emotion recognition has found similar results of reduced accuracy for older adults in the
recognition of sadness and anger. Although no previous research has found increased
accuracy in older adults for recognition of fear, almost half of the studies that have
examined recognition of fear have found no age differences for this emotion. Specific
accuracy results for each age group can be found in Table 3. While most studies of
emotion recognition find reduced accuracy for negative emotions across age groups, the
present findings revealed a more pronounced reduction in accuracy for the expressions of
sadness.

25

Table 3: Percentage of Accurate Recognition of Clearly Expressed Emotions.
Emotion Younger Adults Older Adults
Happy 92.5 % 90.8 %
Sad 69.5 % 57.4 %
Angry 87.0 % 81.5%
Fear 71.3 % 78.2 %

Hypothesis 1 and 2. Older adults will exhibit emotion regulation (either increase in
positive affect, decrease in negative affect, or increase in mixed affect) in their judgments
of the emotional content of facial expressions, particularly when the expressions are
more ambiguous.
Affect Optimization
Positive affect in the Clearly Positive, Neutral, and Morphed Expressions. We
found a significant main effect for age, F(1, 234) = 6.32, p = .013, "
p
2
= .026 such that
older adults judged more positive affect in the facial expressions across all levels of
ambiguity compared to younger adults (see Table 4). This finding demonstrated support
for Hypothesis 1 regarding affect optimization.

26

Table 4: Mean of Index Scores of Positive Affect by Expression Group.
Note. HS Morph = Happy/Sad Morph, HA Morph = Happy/Angry Morph, HF Morph = Happy/Fearful Morph; All
Negative = Sad, Angry, and Fearful faces combined; All Morphs = HS, HA, and HF morphed faces combined.

Negative Affect in the Clearly Positive, Neutral, and Morphed Expressions. A
significant main effect for ambiguity was found, F(2, 468) = 401.55, p < .001, "
p
2
= .632
(see Table 5). All subjects judged negative affect in the expressions similarly, with the
highly ambiguous expressions receiving the most endorsements of negative affect.
Within-subjects contrast revealed all three levels of ambiguity to be significantly
different from one another, low versus high F(1, 234) = 720.88, p < .001, "
p
2
= .755 and
medium versus high F(1, 234) = 44.62, p < .001, "
p
2
= .160. Hypothesis 1 and 2 for
affect optimization were not supported with these findings.

Age Group
Younger Older

Control
(n = 64)
Manipulation
(n = 63)
Control
(n = 58)
Manipulation
(n = 53)
Expression M SD M SD M SD M SD
Happy 1.80 0.76 2.03 0.78 2.03 0.88 1.81 0.74
Sad 0.00 0.03 0.01 0.06 0.01 0.07 0.00 0.03
Angry 0.00 0.03 0.00 0.00 0.00 0.00 0.01 0.08
Fearful 0.01 0.04 0.00 0.00 0.01 0.05 0.01 0.06
Neutral 0.35 0.25 0.33 0.26 0.54 0.34 0.45 0.38
HS Morph 0.55 0.45 0.58 0.53 0.75 0.58 0.55 0.51
HA Morph 0.75 0.49 0.82 0.55 1.16 0.72 0.88 0.60
HF Morph 0.20 0.30 0.26 0.37 0.55 0.55 0.50 0.52
All Negative 0.01 0.02 0.00 0.02 0.01 0.03 0.01 0.04
All Morphs 0.50 0.30 0.55 0.36 0.82 0.46 0.64 0.39

27

Table 5: Mean of Index Scores of Negative Affect by Expression Group.
Note. HS Morph = Happy/Sad Morph, HA Morph = Happy/Angry Morph, HF Morph = Happy/Fearful Morph; All
Negative = Sad, Angry, and Fearful faces combined; All Morphs = HS, HA, and HF morphed faces combined.

Positive affect in the Clearly Negative, Neutral, and Morphed Expressions. We
found a significant between-subjects main effect for age such that older adults judged
more positive affect in the facial expressions across all levels of ambiguity compared to
younger adults, F(1, 234) = 26.10, p < .001, "
p
2
= .100 (see Table 4). We also found a
significant Age x Ambiguity interaction effect, F(2, 468) = 9.64, p < .001, "
p
2
= .040 (see
Figure 1). The effect of age was greater for the medium and high levels of ambiguity
than for the low level of ambiguity. While both older and younger adults exhibited an
increase in their judgments of positive affect from the low to high levels of ambiguity,

Age Group
Younger Older

Control
(n = 64)
Manipulation
(n = 63)
Control
(n = 58)
Manipulation
(n = 53)
Expression M SD M SD M SD M SD
Happy 0.00 0.03 0.01 0.09 0.00 0.00 0.01 0.07
Sad 1.91 0.80 2.34 1.00 2.13 1.05 2.16 1.07
Angry 1.70 0.57 1.80 0.74 1.67 0.86 1.56 0.84
Fearful 1.73 0.82 1.97 0.84 1.56 0.83 1.79 1.08
Neutral 0.60 0.32 0.75 0.40 0.66 0.45 0.82 0.60
HS Morph 0.88 0.58 1.10 0.80 0.97 0.69 1.02 0.70
HA Morph 0.61 0.46 0.69 0.63 0.65 0.58 0.77 0.63
HF Morph 1.34 0.95 1.37 0.81 1.12 0.73 1.09 0.93
All Negative 1.78 0.57 2.03 0.70 1.78 0.71 1.84 0.82
All Morphs 0.94 0.51 1.05 0.62 0.91 0.48 0.96 0.58

28

older adults exhibited a significantly greater shift in their judgments of positive affect as
ambiguity increased than did younger adults. Post-hoc contrasts revealed a significant
difference between the low and high level of ambiguity, F(1, 234) = 16.37, p < .001, "
p
2

= .065, however, the difference between the medium and high level of ambiguity was
found to be not significant. These results provide support for Hypothesis 1 and 2
regarding affect optimization.
Figure 1. Age x Ambiguity interaction for positive affect in the clearly negative, neutral,
and morphed expressions.

Negative Affect in the Clearly Negative, Neutral, and Morphed Expressions. A
significant main effect for ambiguity was found, F(2, 468) = 444.61 p < .001, "
p
2
= .655
(see Table 5). All subjects judged negative affect in the expressions similarly, with the
expressions of low ambiguity receiving the most endorsements of negative affect.
Within-subjects contrast revealed all three levels of ambiguity to be significantly

29

different from one another, low versus high F(1, 234) = 527.38 p < .001, "
p
2
= .693 and
medium versus high F(1, 234) = 44.62, p < .001, "
p
2
= .160. Hypothesis 1 and 2
pertaining to affect optimization were not supported by these results.

Affect Complexity
Mixed Affect in the Clearly Positive, Neutral, and Morphed Expressions. A
significant main effect for ambiguity was found, F(2, 468) = 84.756, p < .001, "
p
2
= .266
(see Table 6). All subjects judged mixed affect in the expressions similarly, with the
expressions of low ambiguity receiving the least endorsements of mixed affect. Within-
subjects contrasts revealed a significant difference between the low and high level of
ambiguity, F(1, 234) = 13.66, p < .001, "
p
2
= .327, but not between the medium and high
levels of ambiguity. Hypothesis 1 and 2 regarding affect complexity were not supported
by these results.

30

Table 6: Mean of Index Scores of Mixed Affect by Expression Group.
Note. HS Morph = Happy/Sad Morph, HA Morph = Happy/Angry Morph, HF Morph = Happy/Fearful Morph; All
Negative = Sad, Angry, and Fearful faces combined; All Morphs = HS, HA, and HF morphed faces combined.

Mixed affect in the Clearly Negative, Neutral, and Morphed Expressions. There
was a significant main effect for age, F(1, 234) = 4.89, p = .028, "
p
2
= .020, suggesting
that older adults judged more mixed affect in the facial expressions across all levels of
ambiguity compared to younger adults (see Table 6). There was also a significant Age x
Ambiguity interaction, F(2, 468) = 13.39, p < .001, "
p
2
= .054 (see Figure 2). The effect
of age was greatest for the low level of ambiguity, with older adults judging significantly
more mixed affect in the clear expressions of negative emotion than younger adults. Post-
hoc contrasts revealed a significant difference between the low and high level of

Age Group
Younger Older

Control
(n = 64)
Manipulation
(n = 63)
Control
(n = 58)
Manipulation
(n = 53)
Expression M SD M SD M SD M SD
Happy 0.02 0.09 0.05 0.13 0.02 0.06 0.02 0.07
Sad 0.04 0.13 0.03 0.08 0.08 0.15 0.04 0.10
Angry 0.02 0.07 0.02 0.09 0.20 0.27 0.16 0.25
Fearful 0.04 0.11 0.04 0.09 0.22 0.29 0.14 0.22
Neutral 0.17 0.18 0.17 0.18 0.18 0.22 0.18 0.17
HS Morph 0.16 0.25 0.19 0.24 0.16 0.22 0.20 0.26
HA Morph 0.13 0.19 0.16 0.20 0.09 0.19 0.08 0.17
HF Morph 0.15 0.22 0.18 0.22 0.18 0.22 0.19 0.23
All Negative 0.04 0.08 0.03 0.06 0.17 0.19 0.11 0.15
All Morphs 0.15 0.15 0.18 0.18 0.14 0.17 0.16 0.16

31

ambiguity, F(1, 234) = 28.19, p < .001, "
p
2
= .107, but not between the medium and high
levels of ambiguity. Hypothesis 1 pertaining to affect complexity was supported by these
results, while Hypothesis 2 received mixed support. Although we predicted an Age x
Ambiguity interaction for affect complexity, we hypothesized the effect of age to be
greatest at the high level of ambiguity as opposed to the low level of ambiguity.
Figure 2. Age x Ambiguity interaction for mixed affect in the clearly negative, neutral,
and morphed expressions.

Hypothesis 3. Future Time Perspective is related to emotion regulation, and
manipulating time perspective can alter the degree to which emotion regulation emerges
in the judgment of facial emotion.
Affect Optimization
Positive affect in the Clearly Positive, Neutral, and Morphed Expressions. We
found a significant between-subjects interaction effect of Age x Time Perspective, F(1,

32

234) = 6.51, p = .011, "
p
2
= .027 such that for both older and younger adults, the amount
of positive affect in their judgments is dependent on whether they were in the control
group or manipulation group (see Table 4 and Figure 3). Across all three levels of
ambiguity, older adults in the control group perceived significantly more positive affect
in the images overall than those older adults who where induced into an expanded time
perspective. In contrast, younger adults in the control group judged similar amounts of
positive affect in the images overall compared to those who were induced into a limited
time perspective. The manipulation of the older adult group’s future time perspective to
that of a younger person served to eliminate the age differences in judgment of positive
affect. These results provide partial support for Hypothesis 3.

33

Figure 3. Age x Time Perspective interaction for positive affect collapsed across the
clearly positive, neutral, and morphed faces.

Negative Affect in the Clearly Positive, Neutral, and Morphed Expressions. We
found a significant main effect for Time Perspective in the judgment of negative affect
F(1, 234) = 5.60, p = .019, "
p
2
= .023, such that across both age groups, individuals
whose time perspective had been manipulated perceived more negative affect in the
expressions overall compared to those in the control group (see Table 5). When induced
into an expanded time orientation, older adults as expected exhibited more negative affect
in their judgments across all the facial expressions. However, contrary to our predictions,
younger adults when induced into a limited time perspective exhibited more negative
affect in their judgments of the expressions compared to their counterparts in the control
group. These findings provide mixed support for Hypothesis 3.
Positive affect in the Clearly Negative, Neutral, and Morphed Expressions. Again
we found a significant Age x Time Perspective interaction effect in the perception of

34

positive affect F(1, 234) = 4.14, p = .043, "
p
2
= .017 (see Table 4 and Figure 4). When
induced into an expanded time orientation, older adults exhibited significantly less
positive affect in their judgments of the expressions compared to their counterparts in the
control group. However, younger adults’ judgment of positive affect remained relatively
unchanged when they were induced into a limited time perspective. These results provide
mixed support for Hypothesis 3. Additionally, the Age x Ambiguity x Time Perspective
3-way interaction approached significance, F(1.80, 421.82) = 3.02, p = .055, "
p
2
= .013
(see Table 4).
Figure 4. Age x Time Perspective interaction for positive affect collapsed across the
clearly negative, neutral, and morphed faces.

Negative affect in the Clearly Negative, Neutral, and Morphed Expressions. We
found a significant main effect for Time Perspective in the judgment of negative affect
F(1, 234) = 5.04, p = .026, "
p
2
= .021, such that across both age groups, individuals
whose time perspective had been manipulated perceived more negative affect in the

35

expressions overall compared to those in the control group (see Table 5). When induced
into an expanded time orientation, older adults exhibited more negative affect in their
judgments across all the facial expressions. However, contrary to our predictions,
younger adults when induced into a limited time perspective exhibited more negative
affect in their judgments of the expressions compared to their counterparts in the control
group. These results provide mixed support for Hypothesis 3.
Affect Complexity
Mixed affect in the Clearly Positive, Neutral, and Morphed Expressions. No
significant effects were found for Time Perspective in perception of mixed affect.
Hypothesis 3 was not supported for affect complexity.
Mixed affect in the Clearly Negative, Neutral, and Morphed Expressions. No
significant effects were found for Time Perspective in this analysis. Hypothesis 3 was
not supported for affect complexity.
Emotion-Specific Comparisons for 2 Levels of Ambiguity
Positive Affect in the Sad and Happy/Sad Morphs. There was a significant main
effect for ambiguity, F(1, 234) = 326.73, p < .001, "
p
2
= .583 (see Table 4). The
happy/sad morphed expressions were rated with significantly more positive affect than
the sad expressions overall. No differences were found between age or time perspective
group. No hypotheses were supported by the results.
Negative Affect in the Sad and Happy/Sad Morphs. There was a significant main
effect for ambiguity, F(1, 234) = 353.77, p < .001, "
p
2
= .602 (see Table 5). The sad
expressions were rated with significantly more negative affect than the morphed

36

expressions overall. No differences were found between age or time perspective group.
No hypotheses were supported by the results.
Mixed Affect in the Sad and Happy/Sad Morphs. There was a significant main
effect for ambiguity, F(1, 234) = 64.69, p < .001, "
p
2
= .22 (see Table 6). The morphed
expressions were rated with significantly more mixed affect than the sad expressions
overall. No differences were found between age or time perspective group. No
hypotheses were supported by the results.
Positive Affect in the Angry and Happy/Angry Morphs. We found a significant
Age x Ambiguity x Time Perspective 3-way interaction effect, F(1, 234) = 8.59, p = .004,
"
p
2
= .035. The amount of positive affect judged in these faces was dependent on age
group, level of ambiguity, and time perspective (see Table 4 and Figures 5a - d). Across
all groups, almost no positive affect was judged from the angry faces, leaving the age
and time perspective group differences to emerge in the judgments of the happy/angry
morphed expressions. Older adults in the control group exhibited significantly more
positive affect in their judgments than younger adults in the control group. Additionally,
older adults in the control group perceived significantly more positive affect in the
expressions than did their counterparts who where induced into an expanded time
perspective. Younger adults’ judgments of positive affect did not differ significantly
between the control and manipulation group. These results demonstrated partial support
for hypotheses 2 and 3.

37

Figure 5a. Age x Ambiguity x Time Perspective interaction for angry and happy/angry
morphs.

Figure 5b. Ambiguity x Time Perspective interaction in younger adults for angry and
happy/angry morphs.

38

Figure 5c. Ambiguity x Time Perspective interaction in older adults for angry and
happy/angry morphs.

Figure 5d. Age x Ambiguity interaction in controls for angry and happy/angry morphs.

Negative Affect in the Angry and Happy/Angry Morphs. We found a significant
main effect for ambiguity, F(1, 234) = 401.50, p < .001, "
p
2
= .632 (see Table 5). The
angry expressions were rated with significantly more negative affect than the

39

happy/angry morphed expressions overall. No differences were found between age or
time perspective group. No hypotheses were supported by the results.
Mixed Affect in the Angry and Happy/Angry Morphs. We found a significant
main effect for age suggesting that older adults perceived more mixed affect in the
expressions overall compared to younger adults, F(1, 234) = 6.23, p = .013, "
p
2
= .026
(see Table 6). There was also a significant Age x Ambiguity interaction effect, F(1, 234)
= 50.80, p < .001, "
p
2
= .178 (see Figure 6). The effect of age was greater at the low
level of ambiguity such that older adults judged significantly more mixed affect for the
clear expressions of anger compared to younger adults. Hypothesis 1 was supported by
these results, and Hypothesis 2 received mixed support.
Figure 6. Age x Ambiguity interaction for mixed affect in the angry and happy/angry
morphs.

Positive Affect in the Fearful and Happy/Fearful Morphs. We found a significant
Age x Ambiguity interaction effect, F(1, 234) = 24.67, p < .001, "
p
2
= .095 (see Figure
7). The effect of age was greater at the high level of ambiguity such that older adults

40

perceive more positive affect in the happy/fearful morphed expressions compared to
younger adults. These results provide support for Hypothesis 2, regarding affect
optimization.
Figure 7. Age x Ambiguity interaction for positive affect in the fear and happy/fear
morphs.

Negative Affect in the Fearful and Happy/Fearful Morphs. We found a main
effect for age, F(1, 234) = 4.87, p = .028, "
p
2
= .020 such that older adults exhibited less
negative affect in their judgments of the expressions overall compared to younger adults
(see Table 5). Results also showed a significant Ambiguity x Time Perspective
interaction effect, F(1, 234) = 3.88, p = .050, "
p
2
= .016 (see Figure 8) such that the effect
of time perspective was greatest at the low level of ambiguity. Across age groups,
individuals whose time perspective had been manipulated judged more negative affect in
the fear faces compared to those in the control group. These results provide support for
Hypothesis 1, and mixed support for Hypothesis 3.

41

Figure 8. Ambiguity x Time Perspective interaction for negative affect in the fear and
happy/fear morphs.

Mixed Affect in the Fearful and Happy/Fearful Morphs. We found a significant
main effect for age, F(1, 234) = 4.87, p = .028, "
p
2
= .020 such that older adults perceived
more mixed affect in their judgments of the expressions overall compared to younger
adults (see Table 6). This main effect was clarified by an Age x Ambiguity interaction
effect, F(1, 234) = 14.16, p < .001, "
p
2
= .057 (see Figure 9). The effect of age was
greatest at the low level of ambiguity with older adults judging significantly more mixed
affect in the clear expressions of fear. There was also a significant Ambiguity x Time
Perspective interaction F(1, 234) = 4.44 p = .036, "
p
2
= .019 (see Figure 10) such that the
effect of time perspective was greatest at the low level of ambiguity. Across age groups,
individuals whose time perspective had been manipulated judged less mixed affect in the
clearly expressed fear faces compared to those in the control group. These results
provide support for Hypothesis 1, and mixed support for Hypothesis 3.

42

Figure 9. Age x Ambiguity interaction for mixed affect in the fear and happy/fear
morphs.

Figure 10. Ambiguity x Time Perspective interaction for mixed affect in the fear and
happy/fear morphs.

43

Hypothesis 4. Executive Functioning will be positively correlated with emotion
regulation in the judgment of facial expressions in the older adult group.
Our composite score of executive functioning was not significantly correlated
with any of the three measurements of emotion regulation. Furthermore, no significant
correlation was found between the three individual measurements of executive
functioning and the measurements of emotion regulation. See Table 7 for complete
results.
Table 7: Correlations for Executive Functioning and Emotion Regulation in Older Adult
Controls.

Note. Within cells, Comp = composite executive functioning; DF = design fluency; PF = phonemic fluency; SF =
semantic fluency; PA = positive affect; NA = negative affect; MA = mixed affect.
*p < .05; **p < .01

Variables
Comp DF PF SF PA NA MA
Executive Functioning
Comp --
Variables DF .682** --
PF .693** .260* --
SF .621** .071 .163 --
Emotion Regulation PA -.121 -.242 -0.76 .082 --
Variables NA .206 .126 .165 .123 .345** --
MA .007 -.175 .009 .187 .591** .220 --

44

Hypothesis 5. Future Time Perspective will be negatively correlated with emotion
regulation within each age group.
Future time perspective as measured by the full scale and the two subscales was
not significantly correlated with any of the three measurements of emotion regulation in
either older or younger adults. The correlation between the focus on limitations subscale
and the perception of mixed affect approached significance in the older adult group, r = -
.254, p = .053. See Tables 8 and 9 for complete results.
Table 8: Correlations for Future Time Perspective and Emotion Regulation in Older
Adult Controls.
Variables FTP FO FL PA NA MA
Future Time Perspective FTP --
Variables FO .963** --
FL -.835** -.657** --
Emotion Regulation PA .189 .147 -.224 --
Variables NA -.214 -.198 .186 .345** --
MA .113 .023 -.259
†
.591** .220 --
Note. Within cells, FTP = future time perspective; FO = FTP subscale – focus on opportunities; FL = FTP subscale –
focus on limitations; PA = positive affect; NA = negative affect; MA = mixed affect.
*p < .05; **p < .01

45

Table 9: Correlations for Future Time Perspective and Emotion Regulation in Young
Adult Controls.
Variables FTP FO FL PA NA MA
Future Time Perspective FTP --
Variables FO .881** --
FL -.774** -.382** --
Emotion Regulation PA .012 .068 .075 --
Variables NA .082 .010 -.156 .361** --
MA .158 .131 -.149 .478** .488** --
Note. Within cells, FTP = future time perspective; FO = FTP subscale – focus on opportunities; FL = FTP subscale –
focus on limitations; PA = positive affect; NA = negative affect; MA = mixed affect.
*p < .05; **p < .01

46

Chapter 4: Discussion

The current study examined age differences in the interpretation of emotion from
facial expressions of clear and ambiguous emotions. Specifically, we were looking for
evidence of emotion regulation that has been seen in older adults’ memory for and
attention to emotional information. Our first hypothesis predicted that older adults would
demonstrate emotion regulation as evidenced by affect optimization and complexity
across all image types; this hypothesis was partially confirmed. We found main effects
for age with positive affect such that older adults perceived more positive affect, a
measure of affect optimization, across all expression types. While affect optimization
can be displayed as an increase in positive emotion and/or a decrease in negative
emotion, most existing research has identified age differences in emotion processing in
the latter. In contrast, our findings showed virtually no age differences in the amount of
negative affect interpreted from the facial expressions. Only in the analysis of the fearful
expressions did we find a main effect for age with older adults interpreting significantly
less negative affect compared to younger adults. We also found evidence of affect
complexity in the comparison of the clearly negative, neutral, and morphed expressions,
with older adults perceiving more mixed affect in their judgments of these expressions
overall. However, there were no age differences in mixed affect for the comparison of
the clearly positive, neutral, and morphed expressions.
Our second hypothesis, based on the notion that ambiguous expressions of
emotion allow for more flexibility in their interpretation, predicted that as ambiguity
increased, affect optimization and complexity would increase in the older adults’

47

judgments of the faces. This hypothesis was confirmed in only one of the facial
comparisons. The comparison of the clearly negative, neutral, and morphed faces
revealed age differences in positive affect that were greater at the level of medium and
high ambiguity. Indeed one would expect positive affect to increase as ambiguity
increased in this particular analysis, however the older adults’ increase in positive affect
from low to high ambiguity was significantly greater than that of younger adults. No
other comparisons showed an increase in emotion regulation as ambiguity increased. On
the contrary, the pattern of results for mixed affect showed age differences to be greater
at the low level of ambiguity in the comparison of the clearly negative, neutral, and
morphed faces. Older adults exhibited a rather consistent level of mixed affect in their
interpretation of the three groups of faces, while younger adults perceived little mixed
affect in the clear expressions and increased in their judgment of mixed affect for the
more ambiguous faces to that of a level similar to older adults. When we examined the
individual negative emotions, we found that older adults perceived mixed affect
specifically in the clearly expressed emotions of anger and fear.
The third hypothesis sought to directly test the assertion made by Carstensen’s
Socioemotional Selectivity Theory (SST) that observed age differences in emotion
processing are driven by age differences in future time perspective. This hypothesis was
confirmed for older adults, but received mixed support in the younger adult sample.
When older adults were induced into an expanded time perspective, like that of a younger
person, they decreased the amount of positive affect in their judgments of the expressions
compared to their counterparts that had not been manipulated. However, when younger

48

adults were induced into a limited time perspective, like that of older adults, their
judgments of positive affect remained largely unchanged when contrasted against their
counterparts who had not been manipulated. Regarding the perception of negative affect,
across both age groups, individuals whose time perspective had been manipulated
increased in their perception of negative affect compared to those in the control group.
Therefore, the evidence of emotion regulation typically present in older adults’
interpretations of emotional expressions was reduced when time perspective was
expanded. On the other hand, younger adults’ judgments were either not influenced by
the manipulation of time perspective as in the case of their perceptions of positive affect,
or influenced in the opposite direction as in the case of their perceptions of negative
affect.
The fourth and fifth hypotheses sought to examine individual difference variables
related to the emergence of emotion regulation. Based on previous research
demonstrating that older adults who performed best on tests of executive functioning
were most likely to show optimization in memory for affective information (Mather &
Knight, 2005), we explored this relationship in our sample of older adult controls to
determine if it generalized to interpretation of emotion. This hypothesis was not
supported as there was no significant correlation between either our composite measure
or individual measures of executive functioning and our measures of positive, negative,
and mixed affect. One possible explanation for these null findings is that our measures of
executive functioning were not tapping the same cognitive control mechanisms examined
in previous work. Specifically, studies examining the positivity effect in memory

49

measured one’s ability to inhibit goal-irrelevant information, and maintain selected
information in working memory in the presence of distracting information. These
inhibition and attentional control tasks are somewhat distinct from our verbal and non-
verbal fluency measures, which assess one’s cognitive flexibility, self-monitoring,
planning, and use of strategies. Despite the fact that these measures all fall under the
umbrella of executive functioning, our findings suggest that the various facets of
executive functioning may not uniformly impact emotion regulatory processes.
Additionally, it is possible that executive functioning does not influence emotion
regulation in all domains of information processing. While cognitive control mechanisms
have been shown to influence the positivity effect in older adults’ memory, recent work
has found no effect of divided attention on the positivity effect in older adults’ attention
(Allard & Isaacowitz, 2009). Therefore, future work should explore the various aspects
of executive functioning in relation to memory, attention, and perception of emotional
information to elucidate these questions.
Lastly, based on SST’s assertion that future time perspective is what underlies the
activation of emotion regulatory processes, we hypothesized that scores on the future
time perspective measure would be negatively correlated with emotion regulation. That
is, as time perspective becomes more limited, evidence of emotion regulation would
emerge in one’s interpretation of emotion. This hypothesis was also not supported, as
there was no significant correlation between the FTP score and the three measures of
emotion regulation. Within group variance in future time perspective was not related to
variance in emotion regulation in either the older or younger adult group. Despite the

50

fact that FTP was shown to have an effect on emotion regulation when experimentally
manipulated, the effects sizes were rather small. As such, it would require significantly
more power and a larger sample size to detect a significant relationship between an
individual difference variable like future time perspective and emotion regulation in
judgments of a rather broad stimulus set. For exploratory purposes, we examined two
proposed subscales of future time perspective in relation to our measures of emotion
regulation (Cate & John, 2007). While the subscales of “focus on opportunities” and
“focus on limitations” were not significantly correlated with our measures of emotion
regulation at the p < .05 level, we found a correlation between the focus on limitations
subscale and the perception of mixed affect that approached significance (p = .053).
Specifically, mixed affect was negatively correlated with this subscale in the older adult
group. Thus, as individuals become more focused on the limitations of their future, they
interpret less mixed emotion from facial expressions of emotion. As focusing on
limitations can be viewed as a negative feature of limited time perspective, it seems
reasonable that an adaptive interpretation style like affect complexity would decrease as
one’s focus on limitations increased. Despite the plausibility of this explanation,
interpretations of this marginal finding should be made with caution as these subscales
have undergone minimal examination. As such, further investigation is necessary to
explore this potential relationship in greater depth.
In summary, our study revealed that older adults do indeed exhibit emotion
regulation in the form of affect optimization and complexity when interpreting the
emotional content of facial expressions. These findings are consistent with past research

51

in demonstrating increased emotion regulation in older adults compared to younger
adults, while also expanding these age differences to another domain of information
processing. Therefore, not only do older adults manage their emotional state by altering
what they attend to, and how they recall emotional information, but also by varying the
way in which they interpret expressions of emotion encountered in their environment.
Our investigation also revealed that the degree to which affect optimization is
utilized could be modulated based on stimulus characteristics like ambiguity of the
expression. Affect complexity on the other hand appears to be a more constant form of
regulation that appears to be activated across all stimulus types. While we predicted that
increasing ambiguity would allow more freedom in interpretation, and thus allow for a
greater degree of regulation to emerge, older adults revealed that when faced with a clear
expression of negative emotion like anger or fear they buffer the experience of that
negative affect by interpreting mixed emotion in the expression. Ong and Bergeman
(2004) and Carstensen, et.al. (2000) have found affect complexity to be negatively
associated with detrimental personality traits like neuroticism, and positively associated
with characteristics like dispositional resilience and emotional control. In the context of
this past work, our findings of affect complexity in the interpretations of the clear
expressions of anger and fear could be interpreted as being in direct service of
maintaining a positive affective state and therefore be adaptive for older adults.
Despite the fact that these findings are in line with SST and our expectations of
increased emotion regulation in later life, their unanticipated nature requires some
consideration. One possible explanation for this unexpected finding pertains to older

52

adults’ vigilance toward negative, particularly threatening, information. Hahn, Carlson,
Singer, and Gronlund (2006) found that older adults, similar to younger adults, were
quicker at detecting an angry face rather than a happy face amongst an array of
nonemotional distracters. Similarly, Mather and Knight (2006) also found older adults to
be faster at locating angry relative to happy and sad faces when presented with neutral
distracter faces. While this automatic vigilance toward threatening information should be
adaptive for older adults, the current study suggests that once this information is detected
it is processed in a manner that reduces its impact on one’s positive affective state. While
fear in particular was not examined in these detection studies, it is reasonable to consider
an expression of fear as being a type of threat cue and thus responded to in a similar
manner as expressions of anger.
These detection studies also provide us with a possible rationale for the divergent
pattern of results we found for the sad expressions. Although sadness is a negative
emotion, it is not considered a threat cue and does not appear to engender the same
vigilance in older adults. Therefore, it is possible that the sad expressions did not
necessitate such regulatory processes in their interpretation, as did the expressions of fear
and anger. Additional explanation for the dissimilar results of the sad expressions was
sought by examining the relatively reduced accuracy both groups showed in judging the
sad faces. Previous research examining age differences in the accuracy of emotion
recognition has shown accuracies in the upper eighties for younger adults and lower
eighties for older adults in the recognition of sadness in facial stimuli. However, the
current study resulted in considerably lower accuracy in identifying sadness in the sad

53

expressions. By providing subjects with a broader range of emotions to choose from, the
current protocol allows for more variability in interpretation of the emotions to be
captured. As such, we might expect a certain amount of reduction in accuracy compared
to past research in this area. However, a reduction in accuracy of this magnitude is cause
for further exploration. While sadness received the highest percentage of endorsements
from younger adults with an average of 69 percent across all 4 faces, worry received the
second highest percentage of endorsements for the sad expressions with an average of 50
percent across the 4 faces. Older adults also judged the sad expressions as worried, but
to a greater degree than did younger adults with an equal percentage of endorsements for
sadness and worry (approximately 58 percent). Worry can be viewed as a more moderate
negative emotion compared to sadness, and one that is relatively common for people to
experience. It is possible that interpreting worry in the expressions of sadness is one way
for older adults to diminish the degree of negative affect in the face. Alternatively,
though worry is a negative emotion, older adults may not experience it as something that
they need to avoid like emotions of anger and fear, which could explain why we found
age effects for emotion regulation in the anger and fear faces but not the sad faces.
Of possibly greater interest, due to their theoretical implications, are the findings
on the experimental manipulation of future time perspective. Although future orientation
has been manipulated in previous research, these studies explored the assertion that future
time perspective influenced the saliency of personal relationships and choices individuals
make regarding those relationships. To the best of our knowledge, this study is the first
to examine the effects of an experimental manipulation of time perspective on emotional

54

information processing. Manipulating older adults’ time perspective to a more expanded
perspective served to reduce the emergence of emotion regulation, specifically affect
optimization, in their perception of emotion. This provides some of the strongest
evidence to date that the activation of emotion regulatory processing in older adults is
driven in part by their more limited future time perspective. While this finding does
support the underlying theory of emotion in later life, it seems paradoxical that a positive
notion like living 20 years longer for a person in older adulthood would have an overall
negative effect on their information processing. However, as SST suggests, shifting older
adults’ time perspective to be more expansive does indeed result in emotional processing
that looks more similar to younger adults.
The results for the manipulation of younger adults’ time perspective, on the other
hand, were not as conclusive. The analyses revealed that the manipulation of younger
adults’ time perspective to a more limited perspective changed their perception of
negative affect, but not positive affect. Furthermore, their change in negative affect was
in the opposite direction from what was hypothesized. One possible explanation for the
lack of change in positive affect is that the manipulation procedure itself may not have
been as effective for younger adults as for older adults due to the challenging nature of
selecting a situation of equal saliency and impact for each group. One might imagine that
for and 80 year old to envision living another 20 years beyond what they had expected
might have a greater impact on how they viewed their future than for a 20 year old to
envision their life as a college student coming to a close. While younger adults’
interpretations of positive affect were largely unchanged by the manipulation of time

55

perspective, the amount of negative affect they perceived in the expressions increased
when their time perspective was manipulated. Based on the assertions of SST, we
predicted that limiting younger adults’ time perspective would reduce the amount of
negative affect in the judgments making them more similar to older adults in the control
group. One possible explanation for this unexpected finding is that the situation posed to
the younger adults in the manipulation group could be viewed as a potentially stress
inducing situation for younger adults. Considering the current economic climate,
graduating from college and moving on to the “real world” at this time for many could
have been an unsettling prospect that conjured up more negative emotion for them.
Another possibility is that the specific situation used with the younger adults could be
viewed as either an ending of one phase of life, or as a beginning of a new phase of life.
This potentially dramatic variation in interpretation of the scenario could have served to
cancel out the effects. However, as we have never examined the effects of time
perspective manipulation on this type of information processing, it is possible that there
are some important underlying differences between a limited time perspective for a
younger adult and that of an older adult. These results would need to be replicated before
warranting an exploration into these potential differences.
Although SST’s assertions are specifically directed at future time perspective’s
influence on older adults’ management of their emotional well being, the theory implies
an understanding of how time perspective operates in younger adults. Past research has
demonstrated that perceived limitations on time manifest similar changes in internal
emotional states and preferences for social partners for both older and younger adults

56

(Carstensen, et. al., 2000; Fung, Carstensen, & Lutz, 1999). However, the current study
suggests that at least within the domain of perception, there is not a direct analogue in
younger adults for the changes in emotional information processing seen in older adults
in response to limitations on time. Therefore, it appears that the influence of perceived
constraints on time permeates older adults to such a degree that it triggers emotion
regulatory processes in decision making, appraisal of emotional experience, and
emotional information processing including attention, memory, and perception.
However, the current study raises questions regarding FTP’s influence on information
processing in younger adults. Further exploration into the effect of limited time
perspective on younger adults’ information processing is warranted to determine if the
pattern of results found in perception of emotion is replicated in attention and memory.
While the present study examined the effects of age, stimulus characteristics, and
future time perspective induction on emotion regulation in a single study, it is not without
its limitations. Most notable is the lack of a direct manipulation check for the future time
perspective induction. While some studies have used a subset of the FTP scale items as a
manipulation check, we wanted to use all ten items on the scale to examine the
relationship of individual differences in FTP and emotion regulation. Furthermore, we
felt that with the brevity of the induction procedure, a pre and post measure of FTP would
prime the participants to the nature of the manipulation and potentially introduce demand
characteristics. We also did not create a comparable situation for participants in the
control condition to imagine, as even seemingly benign situations like imagining your
last trip to the grocery store or what you ate for breakfast can introduce confounds to the

57

experiment. As such, we chose to read the instructions for the judgment task and then
insert the manipulation, which took an additional 60 – 90 seconds to complete, ensuring
that the only difference between the groups would be the brief induction procedure. We
were also unable to measure the return of FTP to baseline levels and cannot know if our
effects were attenuated due to the length of the judgment task.
Additionally, older adults were underrepresented among the actors portraying the
facial expressions. Age of posers has been examined experimentally with regard to its
influence on accuracy of emotion recognition and has been found to not influence the
perceptual accuracy scores for young, middle-aged, and older adults (Moreno, Borod,
Welkowitz, & Alpert, 1993). However, the current study was rather different from
recognition studies. While we found that older adults demonstrated emotion regulation in
the interpretation of the middle-aged and younger adult expressions used in the study, we
cannot be certain that this pattern of results would emerge if the posers had been of
similar age to the older adult sample. This limits the generalizability of our findings to
many older adults, particularly those living in retirement communities as they spend a
significant amount of their social time with individuals in their same age cohort.
The use of USC alumni for the majority of the older adult participant pool was
intended to provide a more equivalent comparison group for the younger adult sample
comprised of USC students. This strategy resulted in large differences between the
groups in both ethnicity and gender, with the older adult sample comprised mostly of
non-Hispanic White males. The current SST literature has not found gender and ethnicity
to be influential in its reported age by valence interactions (e.g. Charles, et.al., 2003), nor

58

have we found meaningful differences in reanalysis of subsamples in this study.
Additionally, the older adults in this sample had relatively low CES-D scores compared
to that found in the general population of older adults. While this somewhat limits the
generalizability of our findings, this is not an uncommon occurrence in experimental
research involving older adult populations. Nevertheless, future studies should explore
gender and ethnic differences more specifically, and use sampling strategies that will
improve generalizability of research findings.
Despite these limitations, the present study represents an important contribution to
the study of age differences in the processing of emotional information. We found that
older adults implement two forms of emotion regulation, affect optimization and affect
complexity, in their perception of emotional information. However, the emergence of
these forms of regulation is somewhat dependent on the nature of the information, with
optimization being more stimulus-dependent and complexity being more constantly
activated. The current findings also bolster the leading developmental lifespan theory of
emotion by demonstrating that future time perspective does play a direct role in the
observed age differences in emotion regulation. Future research is needed to explore
other methods for measuring and manipulating future time perspective to better
understand how future orientation may differentially effect older and younger adults’
emotion processing, as this may in turn reveal avenues for assisting younger adults with
the management of their emotional well being.

59

Bibliography
Allard, E. S., & Isaacowitz, D. M. (2008). Are preferences in emotional processing
affected by distraction? Examining the age-related positivity effect in visual fixation
within a dual-task paradigm. Aging, Neuropsychology, and Cognition, 15, 725-743.

Benton, A. L., & Hamsher, K.deS. (1989). Multilingual Aphasia Examination. Iowa City,
Iowa: AJA Associates.

Calder, A., Keane, J., Manley, T., Sprengelmeyer, R., Scott, S., Nimmo-Smith, I. &
Young, A. (2003). Facial expression recognition across the adult life span.
Neuropsychologia, 41, 195-202.

Carstensen, L. L. (1991). Selectivity theory: Social-Activity in life-span context. Annual
Review of Gerontology and Geriatrics, 11, 195-217.

Carstensen, L. L. (1995). Evidence for a life-span theory of socioemotional selectivity.
Current Directions in Psychological Science, 4, 151-155.

Carstensen, L. L., & Lang, F. R. (1996). Future time perspective scale. Unpublished
manuscript.

Carstensen, L. L., Mayr, U., Pasupathi, M., & Nesselroade, J. R. (2000). Emotional
experience in everyday life across the adult life span. Journal of Personality and Social
Psychology, 79, 644-644.

Carstensen, L. L., & Mikels, J. A. (2005). At the intersection of emotion and cognition.
Current Directions in Psychological Science, 14, 117-121.

Cate, R. A., & John, O. P. (2007). Testing models of structure and development of future
time perspective: Maintaining a focus on opportunities in middle age. Psychology and
Aging, 22, 186-201.

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.

Chepenik, L. G., Cornew, L. A., & Farah, M. J. (2007). The influence of sad mood on
cognition. Emotion, 7, 802-811.

60

Chipperfield, J. G., Perry, R. P., & Weiner, B. (2003). Discrete emotions in later life.
Journal of Gerontology: Psychological Sciences, 58B, 23-34.

Comblain, C., D’Argembeau, A., & Van der Linden, M. (2005). Phenomenal
characteristics of autobiographical memories for emotional and neutral events in older
and younger adults. Experimental Aging Research, 31, 173-189.

Ekman, P., & Friesen, W. V. (1976). Pictures of facial affect. Palo Alto, CA: Consulting
Psychologists Press; 1976.

Ersner-Hershfield, H., Mikels, J. A., Sullivan, S. J., & Carstensen, L. L. (2008).
Poignancy: Mixed emotional experience in the face of meaningful endings. Journal of
Personality and Social Psychology, 94, 158-167.

Fung, H. H., Lai, P. & Ng, R. (2001). Age differences in social preferences among
Taiwanese and Mainland Chinese: The role of perceived time. Psychology and Aging, 16,
351-356.

Gross, J. J., & John, O. P., (2003). Individual differences in two emotion regulation
processes: Implications for affect, relationships, and well-being, Journal of Personality
and Social Psychology, 85, 348–362

Hahn, S., Carlson, C., Singer, S., & Gronlund, S. D. (2006). Aging and visual search:
Automatic and controlled attentional bias to threat faces. Acta Psychologica, 123, 312-
336.

Isaacowitz, D. M., Löckenhoff, C. E., Lane, R. D., Wright, R., Sechrest, L., Riedel, R., &
Costa, P. T. (2007). Age differences in recognition of emotion in lexical stimuli and
facial expressions. Psychology and Aging, 22, 147-159.

Knight, B. G., Maines, M. L., & Robinson, G. S. (2002). The effects of sad mood on
memory in older adults: A test of the mood congruence effect. Psychology and Aging,
17, 653-661.

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.

Labouvie-Vief, G. (2000). Cognitive-emotional integration in adulthood: A personal
view. In J. E. Birren & H. Schroots (Eds.), History of geropsychology in autobiography
(pp. 173-184). Washington, DC: American Psychological Association.

Labouvie-Vief, G., & Blanchard-Fields, F. (1982). Cognitive aging and psychological
growth. Ageing and Society, 2, 183-209.

61

Labouvie-Vief, G., & Medler, M. (2002). Affect optimization and affect complexity:
Modes and styles of regulation in adulthood. Psychology and Aging, 17, 571-588.

Lang, F. R. & Carstensen, L. L. (2002). Time counts: Future time perspective, goals, and
social relationships. Psychology and Aging, 17, 125-139.

Lawton, M. P. (1996). Quality of life and affect in later life. In C. Magai & S. H.
McFadden (eds.), Handbook of emotion, adult development and aging (pp. 327-348). San
Diego, CA: Academic Press.

Lindenberger, U., Mayr, U., & Kliegl, R. (1993). Speed and intelligence in old age.
Psychology and Aging, 8, 207-220.

Magai, C., Consedine, N. S., Krivoshekova, Y. S., Kudadjie-Gyamfi, E., & McPherson,
R. (2006). Emotion experience and expression across the adult life span: Insights from a
Multimodal Assessment Study. Psychology and Aging, 21, 303-317.

Mather, M., & Carstensen, L. L. (2003). Aging and attentional biases for emotional faces.
Psychological Science, 14, 409-415.

Mather, M., & Knight, M. (2005). Goal directed memory: The role of cognitive control in
older adults’ emotional memory. Psychology and Aging, 20, 554-570.

Mather, M., & Knight, M. (2006). Angry faces get noticed quickly: Threat detection is
not impaired among older adults. Journal of Gerontology: Psychological Sciences, 61B,
P54-P57.

Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D.
Sluyter (Eds.), Emotional development and emotional intelligence: Implications for
educators (pp. 3-31). New York: Basic Books.

McDowell, C. L., Harrison, D. W. & Demaree, H. A. (1994). Is right hemisphere decline
in the perception of emotion a function of aging. International Journal of Neuroscience,
79, 1-11.

Meyer, B., Pilkonis, P. A., & Beevers, C. G. (2004). What’s in a (neutral) face?
Personality disorders, attachment styles, and the appraisal of ambiguous social cues.
Journal of Personality Disorders, 18, 320-336.

Moreno, C., Borod, J., Welkowitz, J. & Alpert, M. (1993). The perception of facial
emotion across the adult life span. Developmental Neuropsychology, 9, 305-314.

62

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.

Ong, A. D., & Bergeman, C. S. (2004). The complexity of emotions in later life. Journal
of Gerontology: Psychological Sciences, 59B, 117-122.

Phillips, L. H., MacLean, R. D. J. & Allen, R. (2002). Aging and the perception and
understanding of emotions. Journals of Gerontology, 57, 526-530.

Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the
general population. Applied Psychological Measurement, 1, 385-401.

Radloff, L. S., & Teri, L. (1986). Use of the Center for Epidemiological Studies-
Depression Scale with older adults. Clinical Gerontologist, 5, 119-135.

Rösler, A., Ulrich, C., Billino, P., Sterzer, S., Weidauer, T., Bernhardt, H., Steinmetz, L.,
Frölich, A., & Kleinschmidt, A. (2005). Effects of arousing emotional scenes on the
distribution of visuospatial attention: Changes with aging and early subcortical vascular
dementia. Journal of the Neurological Sciences, 229-230, 109-116.

Sullivan, S. & Ruffman, T. (2004). Emotion recognition deficits in the elderly.
International Journal of Neuroscience, 114, 403-432.

Yeung, D. Y., Fung, H. H., & Lang, F. R. (2007). Gender differences in social network
characteristics and psychological well-being among Hong Kong Chinese: The role of
future time perspective and adherence to Renqing. Aging and Mental Health, 11, 45-56.

Yoon, K. L., & Zinbarg, R. E. (2007). Threat is in the eye of the beholder: Social anxiety
and the interpretation of ambiguous facial expressions. Behaviour Research and
Therapy, 45, 839-847.

Abstract (if available)

Abstract Prior research has established differences between older and younger adults’ (1) attention for emotional stimuli, (2) retrieval of emotional memories, and (3) appraisal of their own emotional experiences. These age differences are seen as evidence of emotion regulation, which is thought to service shifting emotional goals in later life. The current study sought to explore emotion regulation further by assessing age differences in the interpretation of others’ emotions. Specifically, we examined age differences in the judgment of emotion from facial expressions. Our judgment paradigm allowed for the evaluation of two prominent forms of emotion regulation described in the literature, affect optimization and affect complexity. Results showed that older adults did exhibit affect optimization and affect complexity in their judgments of the facial expressions. Specifically, older adults perceived more positive affect and more mixed affect in the expressions compared to younger adults. Additionally, when future time perspective was experimentally manipulated for each age group, age differences in positive affect were reduced and perception of negative affect increased across age groups. Findings are discussed within the framework of developmental lifespan theories of emotion.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Influence of age and anxiety on recognition of facial expressions of emotion: exploring the role of attentional processes

PDF

Dedifferentiation of emotion regulation strategies in the aging brain: an MVPA investigation

PDF

Emotional reactivity and memory biases in older and younger adults

PDF

Optimism when imagining the future in old age influences the positivity effect in memory

PDF

Age differences in the effects of anxiety on selective attention: a test of the vigilance-avoidance hypothesis

PDF

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

PDF

How emotional arousal influences memory and learning in younger and older adults

PDF

Emotion, attention and cognitive aging: the effects of emotional arousal on subsequent visual processing

PDF

Role of age and comorbidities in the quality of life of newly diagnosed prostate cancer patients

PDF

Personality and cognitive aging: are positive traits protective?

PDF

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

PDF

Emotions in engineering: methods for the interpretation of ambiguous emotional content

PDF

Influence of sad mood and old age schema on older adults' physical symptoms processing

PDF

The effects of familism and cultural justification on the mental and physical health of family caregivers

PDF

Emotional arousal amplifies the selectivity of visual selective attention

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

Psychosocial and behavioral ractors associated with emotional eating in adolescents

PDF

Building and validating computational models of emotional expressivity in a natural social task

PDF

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

PDF

Value-based decision-making in complex choice: brain regions involved and implications of age

Asset Metadata

Creator Kellough, Jennifer L. (author)

Core Title Aging and emotion regulation in the judgment of facial emotion

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 11/11/2009

Defense Date 09/08/2009

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

Tag affect complexity,affect optimization,aging,emotion,emotion regulation,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Knight, Bob G. (committee chair), Gatz, Margaret (committee member), Mather, Mara (committee member)

Creator Email jennkello@yahoo.com,kellough@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2726

Unique identifier UC1229251

Identifier etd-Kellough-3386 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-271256 (legacy record id),usctheses-m2726 (legacy record id)

Legacy Identifier etd-Kellough-3386.pdf

Dmrecord 271256

Document Type Thesis

Rights Kellough, Jennifer L.

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu