University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 1

Influence of Age and Anxiety on Recognition of Facial Expressions of Emotion:
Exploring the Role of Attentional Processes

By

Sarah Rastegar

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

Bob G. Knight, Ph.D., Committee Chair
Margaret Gatz, Ph.D., Committee Member
Mara Mather, Ph.D., Committee Member

December 2012

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 2
Acknowledgements
I would like to thank the members of my committee, Bob Knight, PhD, Margaret Gatz,
PhD, and Mara Mather, PhD, for the guidance and support they provided throughout this project.
I would also like to thank the research assistants who helped to complete this study for their
effort and dedication.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 3
Table of Contents
Acknowledgements 2
List of Tables 4
List of Figures 5
Abstract 6
Introduction 7
Facial Emotion Processing and Aging 7
Facial Emotion Processing and Mood 13
The Current Study 22
Hypotheses 23

Methods 27
Participants 27
Measures 28
Stimuli 31
Design and Procedure 32
Analytic Plan 35

Results 36
Characteristics of the Sample 36
Hypothesis 1. Efficacy of the Mood Induction Procedures 39
Hypotheses 2 and 3. Accuracy and Intensity Ratings by Age and
Induction Group 45
Hypothesis 4. Accuracy and Intensity Ratings by Age, State and
Trait Anxiety Groups 48
Hypothesis 5. Accuracy and Intensity Ratings by Age, State Anxiety,
and Attentional Control Groups 52
Post-Hoc Analysis. The Relationship Between Trait Anxiety and
Attentional Control by Age Group 57

Discussion 59
Limitations 69
Theoretical and Practical Implications 71
Summary 75

References 77
Endnote 89

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 4
List of Tables

Table 1: Baseline Age, Health, and Mood (Trait and State) Characteristics
Across Age and Induction Groups, Whole Sample 90

Table 2: Distribution of Participants Across Age and Induction Groups by
Gender, Ethnicity, Education, Occupation, and Income Category,
Whole Sample 91

Table 3: Baseline Age, Health, and Mood (Trait and State) Characteristics
Across Age and Induction Groups, Among Responders to Induction 94

Table 4: Distribution of Participants Across Age and Induction Groups by
Gender, Ethnicity, Education, Occupation, and Income Category,
Among Responders to Induction 95

Table 5. Mean Accuracy (Percent Correct) for Identification of Facial
Expressions of Emotion by Age Group and Induction Group 98

Table 6. Mean Intensity Ratings for Correctly-Identified Facial Expressions
of Emotion by Age Group and Induction Group 99

Table 7. Mean Accuracy (Percent Correct) for Identification of Facial
Expressions of Emotion by Age Group and Trait Anxiety Group
Within (A) the Anxious and (B) the Calm Induction Groups 100

Table 8. Mean Intensity Ratings for Correctly-Identified Facial Expressions
of Emotion by Age Group and Trait Anxiety Group Within (A) the
Anxious and (B) Calm Induction Groups 101

Table 9. Mean Accuracy (Percent Correct) for Identification of Facial
Expressions of Emotion by Age Group and Attentional Control (Attn)
Group Within the (A) Anxious and (B) Calm Induction Groups 102

Table 10. Mean Intensity Ratings for Correctly-Identified Facial Expressions
of Emotion by Age Group and Attentional Control (Attn) Group
within the (A) Anxious and (B) Calm Induction Groups 103

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 5
List of Figures

Figure 1. Change in (A) Visual Analog Scale (VAS) scores and (B) State-Trait
Anxiety Inventory-Y2 State Short Form (STAI-S) scores over time by
induction group for whole sample (Time 1: baseline; Time 2:
post-mood induction) 104

Figure 2. Change in Visual Analog Scale (VAS) scores over time by induction
group in (A) younger and (B) older adults (Time 1: baseline; Time 2:
post-mood induction, Time 3: Following facial emotion identification
task, Time 4: post-induction 2) 105

Figure 3. Change in State-Trait Anxiety Inventory-Y2, State Short Form (STAI-S)
scores over time by induction group in (A) younger and (B) older adults.
(Time 1: baseline; Time 2: post-mood induction, Time 3: Following
facial emotion identification task, Time 4: post-induction 2) 106

Figure 4. Mean accuracy scores (proportion correct) for fear faces in the
(A) anxious and (B) calm induction groups by trait anxiety group 107

Figure 5. Mean intensity ratings for correctly-identified anger faces in the
(A) anxious and (B) calm induction groups by trait anxiety group 108

Figure 6. Mean intensity ratings for correctly-identified fear faces in the
(A) anxious and (B) calm induction groups by trait anxiety group 109

Figure 7. Mean accuracy scores (proportion correct) for fear faces by age group
and attentional control group. Error bars denote +/- 1 standard error 110

Figure 8. Mean intensity ratings for anger faces by induction group and
attentional control group. Error bars denote +/- 1 standard error 111

Figure 9. Mean intensity ratings for fear faces by age group and attentional
control group. Error bars denote +/- 1 standard error 112

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 6
Abstract
Prior laboratory research suggests that older adults are less accurate than younger groups
in identifying facial expressions of certain negative emotions. However, the ability to recognize
positive expressions appears to be preserved with age. The reasons for this variability are as yet
undetermined. Anxiety, via concurrent attentional and interpretive biases, may play an
important role in age differences in the recognition of facial expression by valence, in light of
research that suggests anxiety increases accuracy and intensity appraisal of emotional faces,
particularly those that are threat-relevant (i.e. anger and fear). The current project used an
experimental mood induction paradigm in order to examine the effects of anxious mood, trait
anxiety, and attentional control on accuracy and subjective intensity ratings in a facial expression
recognition task across younger and older individuals. Results indicated that older adults with
concurrent state and trait anxiety demonstrated mood-congruent facial processing, with higher
accuracy and intensity ratings for certain threat-relevant faces; high trait and state anxious
younger adults demonstrated the opposite pattern. Attentional control ability moderated threat-
relevant facial processing, with lower attentional control resulting in an increased ability to
accurately identify fear faces among older adults as well as in higher intensity ratings for threat-
relevant faces among those who were state anxious. Anxiety and attentional control appeared to
be more closely related in older than younger adults. Our results suggest that anxiety and
accompanying mood-congruent biases may help to explain past research that has found age
differences in the ability to identify certain negative facial expressions.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 7
Introduction
Facial Emotion Processing and Aging
Aging and the identification of facial expressions of emotion. The ability to accurately
perceive and identify other individuals’ expressions of emotions is important to social
functioning across the lifespan (Feldman, Philippot, & Custrini, 1991). However, a large body
of research has indicated that older adults often show deficits in the ability to correctly recognize
and label negative facial expressions of emotion relative to younger cohorts (Isaacowitz et al.,
2007; Ruffman, Henry, Livingstone, & Phillips, 2008). In reviews on age differences in
emotional accuracy, Isaacowitz et al. (2007) and Ruffman et al. (2008) found patterns across
studies which suggested that older adults were less accurate in identifying angry and sad facial
expressions of emotion than younger groups (Isaacowitz et al., 2007; Ruffman et al., 2008).
Adding to this, a cross-sectional study involving six cohorts (ages 14 to 84 years) determined the
proportion of individuals accurately identifying angry and sad expressions decreased linearly
with increasing age (Mill, Allik, Realo, & Valk, 2009).
Results on the accurate identification of fear expressions have been somewhat more
inconsistent, with some studies indicating a reduced ability for older adults to correctly label fear
relative to younger adults (Calder et al., 2003; Henry et al., 2008; Wong, Cronin-Golomb, &
Neargarder, 2005) and others suggesting no age-related differences (Phillips, MacLean, & Allen,
2002; Sullivan & Ruffman, 2004, Study 1). The identification of fear has not been found to
follow the same linear pattern with increasing age as anger or sadness. It appears that fear
identification may peak in early adulthood (during individuals’ 20-30s), and subsequently show
linear declines by decade (Mill et al., 2009; Williams et al., 2009). Across age groups, facial
expressions of fear are often shown to be the most difficult emotion to identify (e.g. Calder et al.,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 8
2003; Isaacowitz et al., 2007). Replicating differences in accuracy by age, older adults have been
found to rate negative faces (including anger, fear, and sadness) as less arousing (Mather et al.,
2004) and intense (Riediger, Voelke, Ebner, & Lindenberger, 2011) than younger adults.
Meanwhile, research has suggested that older adults’ ability to recognize positive
emotions is preserved. Across age groups, happy facial expressions of emotion have been shown
to be the most correctly-labeled, with accuracy levels in most studies at or near ceiling
(Isaacowitz et al., 2007; Ruffman et al., 2008). In addition, older adults have been found to
perceive happy and neutral faces as equally as arousing (Mather et al., 2004) and intense as
younger adults (Riediger et al., 2011). Taken together, data have indicated that older adults may
perform as well as younger adults in identifying and interpreting this positively-valenced
emotion.
What might explain older adults’ impairments on negative, but not positive, emotion
identification? A number of possible accounts have been put forward by researchers to explain
these observed differences, including changes in the functioning of brain areas associated with
emotion recognition, general age-related cognitive decline, and problematic gaze patterns (for a
discussion, see Isaacowitz & J.T. Stanley, 2011). However, research examining each of these
purported mechanisms has fallen short of convincingly accounting for the age-related gap in
negative emotion identification performance (Isaacowitz & J.T. Stanley, 2011). Another
hypothesis proposed for the age-related decline in the identification of negative, but
simultaneous preservation of positive, emotional expressions has offered a motivational account
(for a discussion, see Charles & Campos, 2011).
Age-related differences in motivation and attention for emotional information.
Looking broadly to the literature on aging, explanations offered for differences in emotion
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 9
processing with age have focused on motivational changes in attention. Older age is associated
with attention directed towards positive and away from negative stimuli (Isaacowitz, Wadlinger,
Goren, & Wilson, 2006; Mather & Carstensen, 2003). This attentional shift, explained by
Socioemotional Selectivity Theory (SST), is often referred to as the “positivity effect” in late-life
(Carstensen, Fung, & Charles, 2003). As specified by SST, older adults, due to advancing age,
are anticipated to experience a shortened time perspective (Carstensen et al., 2003). In turn,
perceived constraints on time and the increasing salience of mortality are thought to increase
older adults’ motivation for positive experience and emotion regulation (Carstensen et al., 2003).
These motivational goals, prioritizing affect optimization, are hypothesized to drive the positive
attentional biases (positive or neutral information favored over negative) often observed in
emotion research with older adults (Carstensen et al., 2003). According to Carstensen, attention
acts as a key pathway by which the age-related increase in the ratio of positive to negative
information accessed by older adults is enacted (Carstensen et al., 2003; Charles & Carstensen,
2009; Mather & Carstensen, 2005).
A significant amount of research has explored the role of attention to positive and
negative information across age groups. While older adults are often found to perform less well
than younger groups on non-emotional cognitive tasks involving inhibiting information, they
nonetheless are shown to be effective at selectively attending to emotional information
(Samanez-Larkin, Robertson, Mikels, Carstensen, & Gotlib, 2009). Older adults may overcome
general cognitive deficits by allocating a higher proportion of attentional resources to positive
information in service of emotion regulation goals (Mather & M. Knight, 2005).
Research has shown that older adults divert more attention to positive or neutral versus
negative information, including facial expressions of emotion. Older individuals have been
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 10
found to show attentional avoidance of negative (sad, angry) faces and facilitated attention to
positive (happy) and neutral faces, while younger adults demonstrated no attentional biases by
valence (Mather & Carstensen, 2003). In the same study, older adults were significantly more
accurate in their recognition of positive faces seen following the task than those with negative
expressions, while no such memory biases were observed for younger adults (Mather &
Carstensen, 2003). These results suggest that the attentional biases of older adults correspond to
more elaborate processing of positive (happy) than negative (sad or angry) facial expressions.
Similar attentional patterns towards positive and away from negative emotional
expressions with increasing age have been replicated in other studies examining reaction time
(Ebner & Johnson, 2011) and gaze (Isaacowitz et al., 2006). Of note, research indicates that
older adults may show more pronounced reductions in attentional engagement with certain
discrete negative emotions. Older adults have been found to look less at anger and fear faces,
but show no differences with younger adults in fixation time to happy or sad faces (Isaacowitz et
al., 2006). Such emotion-specific results suggest that anger and fear faces, in particular, may
show the most pronounced differences in attentional engagement by age group.
Limits to the “positivity effect”: The role of attentional functioning and alternate
goals in the processing of emotional information. While older adults selectively attend to
positive over negative information, it seems that this bias is the result of emotion regulation goals
initiated later on in processing and not impaired stimulus detection. Older adults have been
shown to initially orient to negative images (Rosler et al., 2005), including faces (Hahn, Carlson,
Singer, & Gronlund, 2006), in the same manner as younger adults. In addition, the ability to
quickly detect a threatening facial expression (anger) as different from an array of neutral faces
is preserved, and potentially even enhanced, in older age (Mather & M. Knight, 2006). These
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 11
findings suggest that the rapid, automatic processing of negative emotional expression stimuli,
particularly those indicative of threat, is intact with increasing age. However, once attention has
been captured, it appears that older adults typically engage less with negative information than
younger individuals, as demonstrated by decreased dwell time on (Rosler et al., 2005) and
increased attentional inhibition for negative stimuli (Hahn et al., 2006). Overall, this research
indicates that older adults’ decreased attention to negative information, including facial
expressions, may not simply be a product of deterioration in “bottom-up” processes (Mather &
M. Knight, 2006), but may occur as a result of “top-down” efforts initiated later in processing.
The positivity effect seems to be driven by controlled processes which older adults’ use to
disengage attention from negative or threatening stimuli in later phases of information-
processing (Hahn et al., 2006).
The role of attentional control in older adults’ discrepant processing of valenced-
information has been further elaborated in research examining the effects of attentional load and
alternate goals on memory for positive and negative stimuli. Older adults have been found to
demonstrate the positivity effect in situations in which their attentional resources are not
compromised. However, when demands on attention are high (e.g. in a divided-attention task),
older individuals appear to no longer demonstrate the positivity effect, instead viewing and
recalling information more similarly to younger individuals (more negative than positive; M.
Knight et al., 2007; Mather & M. Knight, 2005, Experiment 3).
Adding support to the role of higher-order processes in the positivity effect, older adults
who score highest on cognitive control measures are most likely to evidence increased positive
memory (Mather & M. Knight, 2005, Experiment 2), as well as gaze preferences favoring
positive stimuli (Isaacowitz, Toner, & Neupert, 2009). In addition, laboratory research has found
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 12
that when older adults’ attention is directed to alternate goals, for example, by explicitly
instructing participants to focus on accuracy as opposed to emotion-focused aims, the positivity
effect as well as age differences in emotional memory are eliminated, resulting in both older and
younger participants favoring negative over positive information (Kennedy, Mather &
Carstensen, 2004; Lockenhoff & Carstensen, 2007).
The results of these studies suggest that the positivity effect only emerges when “top-
down” modulation of attentional resources are not constrained or redirected to alternate goals.
The positivity effect may be the “default” motivational stance of older adults (Charles &
Carstensen, 2009). However, when demands are placed on older adults’ attention, either directly
through attentional load, or indirectly via the activation of alternate motivational goals, older
adults appear to process negative information in ways similar to younger adults (Charles &
Carstensen, 2009). Thus, it seems that when older adults’ attention is compromised or diverted,
positivity effects are no longer observed.
Implications of age-related differences in motivation and attentional functioning in
the identification of negative emotional faces. If older adults generally use attention to devote
more processing resources to positive and away from negative information, this might offer a
clue as to the reasons for the observed impairments of older cohorts in labeling negative facial
expressions. Older adults may typically have a higher motivation to preserve emotional well-
being and may use attentional biases toward positive and away from negative information in
order to promote these goals. As suggested by this motivational account, age differences in
emotion recognition may, in part, arise from older adults’ reduced engagement with negative
information. However, as detailed above, the positivity effect is not inevitable. In situations in
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 13
which older adults’ attentional control is reduced or redirected due to low attentional resources
and/or competing goals, older adults process valenced information similarly to younger groups.
Thus, it seems plausible that when facing conditions in which top-down positive
attentional biases are disrupted, due to either altered motivational goals or reduced attentional
capacity, older adults’ processing of, and potentially accuracy for, negative facial expressions
may be increased. In addition, as older adults have been found to look less at anger and fear
faces in particular (Isaacowitz et al., 2006), it is reasonable to suppose that processing of these
two discrete negative emotions may be most prone to change with shifts in older adults’ typical
patterns of attentional engagement. While motivational goals and attention can be redirected in a
number of ways, including distractors or explicit instructions (Mather & M. Knight, 2005; Shiota
& Levenson, 2009), mood state, in particular, may contribute to spontaneous shifts in emotion-
processing in daily life.

Facial Emotion Processing and Mood
To varying degrees, emotional states exert an ever-present influence on responses to
stimuli in normal human experience (Damasio, Everitt, & Bishop, 1996). The impact of mood
state on emotional information-processing is widely-recognized. Mood congruence refers to the
finding that current mood state biases the processing of stimuli, favoring processing and retrieval
of similarly-toned emotional material relative to neutral or emotionally-incongruent information
(Bower, 1981; Blaney, 1986). Much of the research on mood-congruence has focused on
depressed mood, in which appraisal is thought to be affected by a negative interpretation
matching the current mood state (A.T. Beck, Rush, Shaw, & Emery, 1979; Bower, 1981).
Similar to depression, anxiety is posited to cause an emotional information-processing bias, in
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 14
this case, directing attention and, subsequently, other cognitive resources, towards environmental
or internal cues suggesting threat or danger (A.T. Beck & Clark, 1997).
Anxiety-related differences in motivation and attention for emotional information.
Anxiety is conceptualized as an evolutionarily-adaptive cognitive-motivational state that
sensitizes individuals to detect and evaluate possible threat in the surrounding environment
(Mogg & Bradley, 1998). As in theories of aging and emotion, attention is a critical feature in
models of anxiety and its disorders. Many of the dominant perspectives in the field of anxiety
maintain that attentional biases both contribute to and maintain anxious symptoms (Barlow,
2000; A.T. Beck & Clark, 1997; Mathews & MacLeod, 2002). The feedback loop in these
theories involves interpretative biases, in which benign stimuli are judged as threatening, and
hypervigilance, resulting in attentional narrowing to sources of threat, and as a consequence,
enhanced threat recognition (Barlow, 2000). Summarizing the relationship between anxiety and
the attentional system, Eysenck, Derakshan, Santos, and Calvo (2007) have proposed attentional
control theory. This theory posits that anxiety shifts the balance between stimulus-driven
(“bottom-up”) and controlled (“top-down”) attention. Anxiety is thought to enhance attention
for threat-related stimuli, while simultaneously impairing controlled attention for non-threat
goals (Barlow, 2000; Eysenck et al., 2007).
Empirical work has demonstrated a positive relationship between anxiety and attention to
negative stimuli, and particularly emotional facial expressions thought to signal threat
(specifically, anger and fear faces; Davis et al., 2011). High state anxious individuals have
demonstrated increased attentional orienting and vigilance to (Bradley, Mogg, Millar, 2000), as
well as delayed attentional disengagement from (E. Fox, Russo, Bowles, & Dutton, 2001;
Georgiou et al., 2005) threat-relevant faces as compared to low state anxious individuals. These
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 15
findings support the idea that in currently anxious individuals, attention is caught and held by
negative, and more specifically threat-related, facial expressions of emotion.
While promoting threat detection and processing, anxious mood appears to come at a cost
to non-threat-related goals. Individuals with high state anxiety have been shown to have general
impairments in attentional control, exhibiting reductions in the efficiency of responses in non-
emotional switching tasks relative to those with lower levels of anxiety (for a review, see
Derakshan & Eysenck, 2009). Combined with the results of studies on attentional bias to threat
in anxiety, research supports the prediction that anxiety facilitates attention to threat while
interfering with other, non-threat goals.
The role of trait predisposition and attentional functioning on the effects of anxiety.
Literature on anxiety has often distinguished between state and trait anxiety. State anxiety
describes the current, transitory experience of anxiety, involving worry, arousal, and negative
emotionality, whereas trait anxiety is defined as the enduring tendency to experience heightened
anxious responding (Spielberger, 1985). Theory and research predict that, for those experiencing
current anxiety, individuals with high levels of trait anxiety will be more reactive (Eysenck,
1992; Mogg & Bradley, 1998; Williams, Watts, MacLeod, & Mathews, 1997). Thus, the
interaction between state and trait anxiety may most prominently produce changes in threat-
relevant processing (Eysenck, 1992; Williams et al., 1997). Multiple studies have found that
only those with the combined effects of high trait and state anxiety demonstrate pronounced
mood-congruent attentional biases, including increased attentional engagement and/or delayed
disengagement for threat-relevant stimuli (e.g. Egloff & Hock, 2001; Keogh & French, 1999; for
a discussion, see Mathews & MacLeod, 1994), and intensified negative judgments (Rusting,
1999). Taken together, it seems that the effects of state anxiety on attention may be moderated by
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 16
trait anxiety, with the effects of elevated state anxiety most apparent in those who report high
trait anxiety (Mathews & Mackintosh, 1998).
Another important moderator of the effects of anxiety on attention may be individual
differences in attentional control capacity. Attentional control reflects an individual’s ability to
select and/or inhibit information for further processing, and may relate to both repeated practice
(Kramer, Larish, & Strayer, 1995) as well as the functioning of brain areas such as the prefrontal
cortex (Ochsner & Gross, 2005). The effects of anxiety on attention may be most pronounced in
individuals with low baseline levels of attentional control due to their reduced ability to inhibit
threat-relevant information. Anxious individuals with low attentional control show increased
engagement with threat cues (including anger faces) relative to high anxious individuals with
high attentional functioning and those with low anxiety across attentional control abilities
(Derryberry & Reed, 2002; Reinholdt-Dunne, Mogg, & Bradley, 2009). These results suggest
that the combination of high anxiety and low attentional control predict increased processing of
threat-relevant stimuli.
Taken together, attentional bias to threatening stimuli in those with elevated levels of
anxiety relative to nonanxious controls has been demonstrated in a number of research studies,
using different experimental paradigms (for a meta-analysis, see Bar-Haim, Lamy, Pergamin,
Bakermans-Kranenburg, & van IJzendoorn, 2007). In addition, research suggests a negative
relationship between anxiety and performance on other, non-threat attentional goals, especially
those requiring attentional control (for a review, see Derakshan & Eysenck, 2009). Thus, anxiety
shifts individuals’ attention away from non-threat goals and towards threat detection. As
discussed, the effects of anxiety on such processes are expected to be particularly evident among
those with both a diathesis to anxious processing and those with low baseline attention
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 17
functioning. Since anxiety promotes attentional processing of threat, particularly for face
stimuli, this lends to the possibility that anxiety might increase attentional engagement with
threat-relevant expressions of emotion, increasing subsequent identification and intensity ratings
of such faces.
The effects of anxiety on the identification of facial expressions of emotion.
Clinically, trait-, and state anxious adults have been found to be more accurate in identifying
negative facial expressions of emotion than low anxious controls, with a number of studies
showing specificity for anger (Mohlman, Carmin, & Price, 2007) and/or fear (Sprengelmeyer et
al., 1997; Surcinelli, Codispoti, Montebarocci, Rossi, & Baldaro, 2006) recognition as compared
to other expression types (but see Cooper, Row, & Penton-Voak, 2008; Philippot & Douilliez,
2005). Analyzing the data by stimulus intensity and error types illuminates unique patterns in
those with clinically-significant anxiety. Individuals with social anxiety have displayed a hyper-
sensitivity to threat-related emotional expressions, detecting negative emotions, particularly
anger and fear, at lower levels of emotional intensity (Joormann & Gotlib, 2006; Frenkel, Lamy,
Algom, Bar-Haim, 2009), with greater sensitivity (Richards et al., 2002), and with faster
response times when placed in an anxiety-provoking condition (Leber, Heidenreich, Stangier, &
Hofmann, 2009; Mullins & Duke, 2004). High anxiety sensitive individuals have also been
found to report increased subjective ratings for threat-relevant faces (anger and fear) relative to
low anxiety sensitive individuals, and to rate fear faces as more intense when in a high anxious
as compared to a calm condition (Liebman & Allen, 1995). Taken together, there is evidence to
suggest that those with high anxiety may be more accurate in the identification of negative facial
expressions, particularly those signaling threat, as well as more likely to ascribe heightened
subjective threat value to such stimuli.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 18
The effects of age and anxiety on attention for emotional information. While the vast
majority of work on attentional and interpretive biases in anxiety, including that cited above, has
focused on younger adults, a small literature exists that has pointed to the role of anxiety in
enhancing attention to emotional information in aging populations. Studies with older
populations have demonstrated attentional biases towards threat stimuli in anxious older adults
relative to low anxious peers similar to that seen in younger individuals. Research has shown that
older individuals induced to feel anxiety displayed an attentional bias towards negative stimuli in
a dot-probe task of approximately the same magnitude as that seen in younger adult samples
(L.S. Fox & B.G. Knight, 2005). Older adults high in trait anxiety have also demonstrated
increased attentional orienting to sad faces (L.O. Lee & B.G. Knight, 2009). In addition, older
adults with high trait anxiety have been shown to perform more poorly than peers with low trait
anxiety (Beaudreau & O’Hara, 2009) and younger adults across trait anxiety level on tasks of
controlled attention for non-threat information (Hogan, 2003). These studies suggest that the
attentional effects of anxiety operate similarly in older adults as in younger groups, facilitating
threat-detection and engagement while reducing performance in other non-threat attentional
goals.
Implications of age-related differences in anxiety on the identification of negative
emotional faces. The possible implications of anxiety on the processing of threat-relevant
information, and particularly faces, across age groups are multiple. Anxiety taxes the overall
attentional system, reducing the resources allotted to processing of non-threat goals. However,
for emotional tasks, anxiety would be anticipated to increase detection and processing of
negative information across age groups, due to increases in attentional allocation, efficiency, and
depth of engagement with negative, and specifically threat-relevant, stimuli. While both anxious
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 19
older and younger adults might be expected to show increased processing biases to threat-
relevant stimuli, such as anger and fear facial expressions, the impact of anxious state might be
even more pronounced for older adults in both domains.
For one, the effects of anxiety on attentional biases have been shown to be more
pronounced for those with low attentional capacity, and, on average, attention capacity declines
with age (Cavanaugh & Blanchard-Fields, 2010). Secondly, the negative attentional biases
accompanying anxiety, which facilitate threat processing at the expense of other attentional
goals, stand in sharp contrast to the positive attentional biases that frequently occur with
increasing age. The positivity effect is driven by controlled processes that bias older adults to
selectively attend to positive, while avoiding negative, stimuli in the service of current emotion
regulation goals. In contrast, anxiety is an anticipatory future-oriented mood state which
prepares individuals to cope with upcoming negative events by focusing on and processing
threatening information at the expense of neutral and positive stimulus-processing (Barlow,
2000; Eysenck et al., 2007).
These opposing attentional goals raise the question of how anxiety might interact with
age and attentional control in the identification of negative emotions. Based on research
suggesting that when cognitive resources are low, the positivity effect is reversed in older adults
(Mather & M. Knight, 2005), anxious older adults would be expected to show unique reductions
in their abilities to selectively-attend to positive information due to impaired attentional control
in the service of threat-processing. Moreover, by removing the positivity bias typically present
for older adults, anxiety should increase processing of negative information in anxious older
adults relative to age-matched controls to a greater extent than between anxious and control
younger adults. The possible implications of anxiety on the processing of threat-relevant
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 20
information, such as anger and fear emotional faces, across age groups is evident but has only
rarely been considered in emotional aging research to this point.
While some studies on age-related differences in emotion identification have assessed for
the presence of depressive symptoms, rarely have measures of anxiety been included in these
protocols. The relative lack of assessment of state and trait anxiety in research on aging and
emotion-processing is of particular concern in light of the fact that younger adults are
consistently found to score higher on anxiety assessments than older adults (Wolitzky-Taylor,
Castriotta, Lenze, M.A. Stanley, & Craske, 2010). Thus, it remains possible that differences by
age group in the ability to recognize certain facial expressions may, in fact, be the result of
differences in anxiety levels, particularly in the identification of those facial expressions that are
mood-congruent, such as anger and fear.
Some recent research has examined the role of anxious mood on age differences in facial
emotion processing during post-hoc analysis. In one study which looked at accuracy for facial
expressions of emotion with older and younger individuals, differences in sadness recognition by
age group favoring younger adults were statistically eliminated when negative affect scores,
significantly higher in the younger participants, were controlled for in a regression analysis
(Suzuki, Hoshino, Shigemasu, & Kawamura, 2007). However, another study on emotion
recognition that controlled statistically for differences in both baseline depressive and anxious
symptomology (again, significantly higher in younger as compared to older adults), found
neither positive nor negative emotion recognition accuracy scores were correlated with pre-task
mood scores (N.A. Murphy & Isaacowitz, 2010). A third study, examining age differences in
ratings of emotional intensity attributed to facial expressions of emotion, revealed that while
younger adults perceived negative emotions as more intense than older adults, this age effect was
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 21
reduced to non-significance in happy and sad faces when differences in baseline anxiety and
depression scores (once more, higher in younger adults) were covaried in the model (Phillips &
Allen, 2004).
Taken together, these studies provide mixed evidence, but are suggestive that differences
in mood by age may, in part, explain differences in emotion identification and perception in
older and younger adults. While attempting to control for differences in anxiety levels by age
group on emotion identification post-hoc is an important step in understanding the potential
effects of mood on emotion expression recognition, the lack of primary attention to this variable
in study designs may obscure the magnitude of its influence. As a result, the possible effects of
anxiety on older adults’ processing and interpretation of threat-related facial expressions of
emotion remain largely unaddressed. Given that older adults consistently perform poorly in
accurately appraising certain negative facial expressions of emotion relative to younger cohorts,
and also report lower levels of anxiety, we believed the need to examine the interaction of age
and anxiety on emotion recognition as a primary objective remained an important endeavor.
A number of research protocols have used mood induction to understand the effects of
variation in affect state on emotion identification in younger participants (T.M.C. Lee, Ng, Tang,
& Chan, 2008; Ridout, Noreen, & Johal, 2009). Mood induction offers an analog to everyday
and clinical mood variation, as well as number of potential research design benefits, including
experimental control of affective state and fewer potential confounds. Mood induction methods,
including those that allow for personalized responding as in prompted emotional memory
recollection, are shown to be an effective way to influence current affect, and have been used
successfully in research involving adult participants across age groups (e.g. Levenson,
Carstensen, Friesen, & Ekman, 1991; L.S. Fox, B.G. Knight, & Zelinkski, 1998; Poon & B.G.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 22
Knight, 2009). Despite the strengths and efficacy of this approach, at the present time, there
appears to be only limited research on the effects of induced anxious mood on facial emotion
expression identification in non-clinical younger populations, and none in older groups.
The Current Study
Data and theory suggest that both advanced age and anxiety, independently, impact
attention to valenced-information in opposing ways. Without competing attentional goals, older
adults demonstrate a positivity bias, attending to, processing, and retrieving positive information
more readily than negative. However, the positivity bias is not inevitable; when controlled
attention is redirected or reduced, older adults behave much more similarly to younger groups in
accessing negative information. In contrast to the positivity bias, anxiety drives attention in the
reverse direction, promoting threat detection and negative evaluation while disrupting the pursuit
of other non-threat attentional goals (Barlow, 2000; Eysenck et al., 2007).
Based on this research, we hypothesized that anxiety in older adults would disrupt the
positivity bias by prioritizing and diverting attention resources towards an incompatible goal:
threat-detection. We expected the attenuation of the positivity effect in anxious older adults, as a
result of reduced controlled attention combined with stimulus-driven attention to threat, would
result in older adults performing similarly to younger adults in processing certain types of
negative information. To this end, the current project used an experimental mood induction in
order to examine the effects of anxiety and age on accuracy and perceived intensity in the
identification of facial expressions of emotions. A significant limitation in aging and emotion
recognition research includes the heavy reliance on one set of stimuli (Pictures of Facial Affect,
Ekman & Friesen, 1976). In an attempt to overcome this shortcoming, the emotional stimuli
selected for this research project involved the use of more naturalistic expressions of facial
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 23
emotions (Gur et al., 2002). Based on lessons learned from past research, consideration was paid
to accuracy as well as perceived intensity for the face stimuli. For each face, the response format
included both a forced-choice selection of the facial emotion expression presented, as well as an
assessment of participants’ subjective intensity rating of that emotion, allowing for a more
nuanced understanding of these judgments. In addition, a limited presentation time for each face
stimulus provided increased experimental control as well as eliciting rapid emotion evaluation as
in real world situations. Finally, by exploring the roles of trait anxiety and attentional control
ability on the relationship between mood state and emotion recognition by age group, we sought
to further elucidate the moderators of any observed effects, an understudied aspect of research on
age differences in emotion expression identification.

Hypotheses
Hypothesis 1. Based on the past efficacy of mood induction procedures across ages, we
predicted a main effect of mood induction group resulting in significant differences in state
anxiety ratings for those assigned to the anxious as compared to a calm condition across older
and younger adults.
Hypothesis 2. Based on research regarding emotion recognition accuracy and attentional
biases accompanying aging and in anxiety, we predicted accuracy for correct facial expression
identification would vary by age and mood state.
i. Replicating past findings on emotion recognition accuracy by age group, we
predicted a main effect of age such that younger adults would be more accurate
than older adults in the identification of all negative facial expressions (anger,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 24
fear, sad), while no differences by age group would be observed in the accurate
identification of the positive facial expression (happy).
ii. Based on past findings in emotion recognition accuracy in anxiety, we predicted a
main effect of induction such that younger and older adults in the anxious mood
condition would be more accurate in the identification of threat-relevant facial
expressions of emotion (anger, fear) than age-matched peers in the calm
condition.
iii. Because younger and older adults typically present with different levels of
baseline anxiety and different emotion-processing goals, we predicted an Age by
Induction group interaction such that younger adults in the anxious mood
induction condition would be more accurate in the identification of threat-relevant
emotions (anger, fear) than all other groups. We expected older adults in anxious
mood condition would be more accurate than older adults in the calm condition in
the identification of threat-relevant expressions of emotion.
Hypothesis 3. Based on research regarding perceived emotional intensity and attentional
biases accompanying aging and in anxiety, we predicted subjective intensity ratings for correct
facial expression identification would vary by age and mood state:
i. Replicating past findings on perceived emotional intensity by age group, we
predicted a main effect of age. We hypothesized that, when correctly-identified,
older adults would rate positive (happy) faces as subjectively more intense than
younger adults, but that younger adults would rate negatively-valenced emotions
(anger, fear, sad) as more intense than older adults.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 25
ii. Based on past findings in perceived emotional intensity in anxiety, we predicted a
main effect of induction such that younger and older adults in the anxious mood
condition would rate correctly-identified threat-relevant facial expressions of
emotions (anger, fear) as more intense than age-matched peers in the calm
condition.
iii. Because younger and older adults typically present with different levels of
baseline anxiety and different emotion-processing goals, we predicted an Age by
Induction group interaction such that younger adults in the anxious mood
induction condition would rate correctly-identified threat-relevant emotions
(anger, fear) as displaying higher intensity than all other groups. We expected
older adults in the anxious mood condition would rate correctly-identified threat-
relevant emotions as higher intensity than older adults in the calm condition.
Hypothesis 4. Because emotion-processing biases may be most prominent in individuals
experiencing current anxiety who also have a tendency to anxious responding (State by Trait
interaction), we anticipated that the effects of anxious mood induction predicted in Hypotheses 2
and 3 might be most strongly observed in those with increased self-reported trait anxiety.
Specifically, within the anxious induction group, we expected:
i. Younger adults with high trait anxiety would be more accurate in the
identification of threat-relevant emotions (anger, fear) than all other groups. We
expected older adults with high trait anxiety would be more accurate than older
adults with low trait anxiety in the identification of threat-relevant emotions
expressions of emotion.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 26
ii. Younger adults with high trait anxiety would rate correctly-identified threat-
relevant emotions (anger, fear) as more intense than all other groups. We
expected older adults with high trait anxiety would rate correctly-identified threat-
relevant emotions as more intense than older adults with low trait anxiety.
Hypothesis 5. Based on theory and research suggesting that attentional ability moderates
the effects of anxiety on emotion-processsing, we predicted that attentional control ability (as
demonstrated by lower scores on a task of non-emotional attention) would be associated with the
relationship between mood and accuracy in identifying threat-relevant emotions (anger and fear),
as well as with the relationship between mood and subjective intensity ratings for these
emotions. We predicted that the relationship between attentional control and mood induction
group would be seen in both older adults and younger adults, but, due to the unique effects of
attentional ability on older adults’ emotion-processing, we expected this effect would be stronger
in older than younger participants.
i. We predicted state anxious individuals with low attentional control would be
more accurate in the correct identification of threat-relevant emotions (anger,
fear) and that this effect would be larger for older relative to younger participants.
ii. We predicted that state anxious individuals with low attentional control would
rate correctly-identified threat-relevant emotions (anger, fear) as more intense
than low trait anxious groups and that effect would be larger for older relative to
younger participants.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 27
Method
Participants
Younger adult participants (N = 76) were recruited from undergraduate psychology
classes at the University of Southern California (USC). Younger adult participants ranged in age
from 18 to 23 years old (M=19.74, SD=1.182) and received course extra credit for participation
in the protocol. Older adult participants (N = 66) were recruited from the USC Healthy Minds
subject pool, USC Emeriti Center, and USC Andrus Gerontology Center volunteer group. Older
adult participants ranged in age from 60 to 95 years old (M=71.41, SD=6.881) and received a
stipend ($10) for their participation in the study.
Participants were recruited via telephone, flyer, or online posting, and those interested
were provided with additional information regarding the study. All subjects were community-
dwelling adults living independently and English language proficient based on participants’
ability to understand and respond to the initial invitation. Self-report was used to assess for
currently diagnosed psychiatric disorders, including mood and anxiety disorders (e.g. major
depressive disorder, generalized anxiety disorder). Those individuals reporting current mood
and anxiety diagnoses were excluded, as well as subjects currently taking selective serotonin
reuptake inhibitors (SSRIs) due to potential effects of these psychotropic medications on the
recognition of fear expressions (Merens, Willem Van der Does, & Spinhoven, 2007; S.E.
Murphy, Norbury, O’Sullivan, Cowen, & Harmer, 2009). A brief telephone-administered
cognitive assessment (TELE; Gatz et al., 2002) was completed with older adult participants and
used to exclude those with possible dementia based on a cut-off score of 16. Subjects with self-
reported significant hearing or vision problems were excluded due to the auditory and visual
nature of the task.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 28
Measures
Demographic information. All subjects provided information regarding age, gender,
self-reported race/ethnicity, current occupation (Student, Retired, or Other), years of formal
education (ordinal categories), and current annual income (ordinal categories). Additionally, a 5-
point Likert item assessing subjective health (Poor, Fair, Average, Good, or Excellent) was used
to examine any major discrepancies in perceived physical well-being between groups.
Measures of trait and state anxiety.
Beck Anxiety Inventory - Trait. (BAIT; Kohn, Kantor, Decicco, & A.T. Beck, 2008). The
BAIT was used to measure trait anxiety. The BAIT is a 21-item self-report measure assessing
the tendency to experience cognitive, emotional, and somatic symptoms of anxiety. The BAIT is
an adapted version of the Beck Anxiety Inventory (BAI; A.T. Beck, Epstein, Brown, & Steer,
1988) that maintains item content but has been modified to change the timeframe of the items
from an assessment of prolonged state anxiety (“In the last week...”) to a measure of trait
anxiety. For each item, respondents rate their experience of symptoms described by the statement
using a 4-point scale (ranging from rarely or never to almost always).
The original BAI has been found to have adequate psychometric properties with non-
clinical samples of younger and older adults, including high internal consistency, good
convergent validity and test-retest reliability (A.T. Beck et al., 1988; Dennis, Boddington, &
Funnel, 2007; Wetherell & Arean, 1997). The modified BAIT has shown similar results with a
college sample, including higher test-retest correlations of r = .83 as compared to r = .69 for the
BAI over a 3-week period (Kohn et al., 2008). Evidence for discriminant validity of the BAIT in
a college sample found that the BAIT correlated most strongly with measures of prolonged state
anxiety, and less so with measures of state depression (Kohn et al., 2008). However,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 29
discriminant validity of the BAIT in older samples has not been conducted. Discriminant
validity between the original BAI and measures assessing depression have been less well-
established with older adult samples, which may be due to the reduced accuracy of older adults
in identifying symptoms of both anxiety and depression or to the high comorbidity of these two
disorders (near 50%) within older populations (Fiske, Wetherell, & Gatz, 2009; Wetherell et al.,
2009). The internal consistency for the BAIT in the current study was α = .901 for younger
adults and α = .814 for older adults.
Visual Analog Scale (VAS) Anxiety. An 11-point visual rating scale was used to assess
state anxiety over the course of the session. Response options were presented along a horizontal
continuum. Participants placed an “X” to indicate their current level of symptoms on a line with
numeric markers for whole values at equally-spaced intervals. Response anchors at the ends of
the state anxiety VAS scale were 0 - no anxiety/worry and 10 - as much anxiety as you have ever
felt. Responses marked between integers were scored as intermediate (e.g. 8.5) if closer to the
interval midpoint than a whole number value. VAS scales have been used in the assessment of
anxiety (Mitchell, Baker-Glenn, Granger, & Symonds, 2010), though explicit validation of the
VAS technique has not been conducted.
State-Trait Anxiety Inventory-Y2, State Short Form (STAI-Y2; Spielberger, Gorsuch,
Lushene, Vagg, & Jacobs, 1983). Ten-item short-form versions of the STAI-Y2 State scale
(herein referred to as STAI-S) were used to assess state anxiety symptoms over the course of the
session. For each item, respondents rated the perceived accuracy of the statement provided using
a 4-point scale (ranging from not at all to very much so). As in previous mood induction
research involving older and younger adults, alternate versions of the 10-item short-form (ten in
total) were derived from the longer original 20-item STAI-Y2 State scale by randomly-selecting
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 30
five positively worded items and five negatively worded items for each form (L.S. Fox & B.G.
Knight, 2005). For each participant, five versions of the STAI-S were randomly selected and
ordered for use during the protocol. This method was used to help guard against fatigue and
response set biases often found with repeat administrations of the same form in a brief span.
The STAI-Y2 State scale has been found to have adequate psychometric properties with
younger adults (Spielberger et al., 1983) and older participants, including good internal
consistency (α = .85) and 2-4 week test-retest coefficients in the moderate range (r = 0.62; M.A.
Stanley, J.G. Beck, & Zebb, 1996), appropriate for an assessment of state anxiety which should
vary over occasions. The 10-item short-form version of the STAI-Y2 has been used in past
research with older and younger participants and is shown to have good reliability across forms
(α = 0.84) and to be sensitive enough to capture changes in mood state due to mood induction,
(L.S. Fox & B.G. Knight, 2005). In the current study, we computed the reliabilities each of the
short form versions across time points. Alpha’s for the alternate forms ranged from α = .867 to
α = .930, suggesting good reliability for these short-form measures.
Measure of attentional control. The Connections Test (Salthouse et al., 2000) was used
to measure attentional control in the current study. The Connections Test is a variant of the trails
making test, frequently used to measure visual attention and shifting (Lezak Howieson, &
Loring, 2004). In the Connections Test, participants are asked to connect 49 numbers, letters, or
an alternating sequence of the two, in order as rapidly as possible. Participants attempt to
complete as many correct connections as possible in 20 seconds. There are four types of
connections tasks (numbers only, letters only, numbers-letters alternating, letters-numbers
alternating; see Appendix A). Within this study, one of each type of task was selected, resulting
in a total of 4 trials, administered in the order listed above. For each form, the total number of
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 31
correct connections minus the total number of errors (combined omissions and commissions)
made was calculated. Scores obtained were averaged across the simple (A) condition (number-
and letter-only), as well as alternating condition (B; number-letter and letter-number). Higher
values correspond to more efficient attention control. The ratio of alternating conditions to
simple conditions scores (B/A), used here, provides a measure of processing efficiency that
shows only a very small relation to age (r = 0.07; Salthouse, 2011). The Connections Test has
been shown to have good reliability, with alpha’s greater than .80 for all variables in the task
(Salthouse, Atkinson, & Berbish, 2003) and has been used in research with both younger and
older adults (Salthouse et al., 2000; Salthouse, 2011).

Stimuli
Facial expressions of emotion. The Penn Emotion Recognition Task (ER-40; Gur et al.,
2002) was used to assess emotion recognition. The Penn Emotion Recognition Task is a
standardized set of photographs of evoked facial expressions of emotion (anger, fear, happy, sad,
no emotion [neutral]). The target in each color photograph is an actor or actress coached to re-
live a mood-congruent experience of either mild or extreme intensity. Autonomic and mood
ratings of these evoked responses were recorded for each actor to validate each emotion
generated. The set contains 40 items, 8 facial expressions stimuli for each of the 4 emotions
listed, as well as 8 neutral faces. Each actor is shown only once, and actors in the set are
balanced for gender. In addition, actors within the Penn Emotion task represent a number of
ethnic groups, including Caucasian, African-American, Hispanic, and Asian individuals.
The ER-40 has been used in schizophrenia research and has been demonstrated to have
acceptable test-retest reliability for both patient and normal control populations (r = 0.76-0.80,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 32
respectively), good construct validity established through convergence with other measures of
emotion recognition and with functional activity in corresponding neural systems, and adequate
sensitivity to capture changes in accuracy following behavioral intervention (for a brief review,
see Carter et al., 2009). In a large study of both healthy participants and those with
schizophrenia, the mean percent correct emotion identification for healthy individuals was
84.0%, versus 73.1% in the patient sample, suggesting that the ER-40 is not constrained by either
ceiling or floor effects and may be able to capture meaningful differences in accuracy between
groups (Carter et al., 2009). The ER-40 has also been used in at least two studies involving both
younger and older participants, demonstrating moderate effect sizes in emotion recognition
accuracy differences between age groups, most prominently for anger and fear faces (Williams et
al., 2008; Sasson et al., 2010).

Design and Procedure
Subjects who expressed interest in participating in the study and met eligibility criteria as
outlined in the screening procedures were scheduled for a study visit. Participants were tested in
groups ranging from 2 to 6 individuals in quiet rooms at USC. Groups were randomized to one
of two mood induction conditions: anxious mood (experimental) or calm (control) mood
condition. The randomization procedure involved the use of a block design by age group in
order to ensure roughly equal assignment of participants from each age group to each mood
condition.
Upon arrival for the study visit, participants were provided with a study information form
approved by the USC Institutional Review Board. After reviewing this information and agreeing
to participate in the protocol, individuals provided demographic information. Participants also
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 33
completed the trait anxiety questionnaire (BAIT) and provided baseline evaluations of their
current state anxiety level using the VAS and STAI-S.
Mood induction. For each condition, mood was induced using a self-generated,
autobiographical memory approach (Salovey & Birnbaum, 1989; Levenson et al., 1991) and
sustained with mood congruent music (Jeffries, Smilek, Eich, & Enns, 2008). The initial
induction (induction 1) lasted approximately 3 minutes; the booster induction (induction 2) lasted
approximately 2 minutes. These times were selected based on feedback from participants in
pilot-testing. During the induction, participants in both the anxious mood and calm control
conditions were prompted to vividly recall a recent situation of personal relevance using a script
modeled after other mood induction research protocols (Jeffries et al., 2008; Rusting, 1999;
Wright & Mischel, 1982). Participants in the anxious mood condition were instructed to reflect
on a situation that has been a source of worry or anxiety with as much detail as possible,
allowing thoughts, images, and other associated sensations to come to mind. Participants in the
calm control condition received similar instructions, though the target memory was a situation
that had recently made them feel relaxed or calm. During the autobiographical memory task,
mood congruent music was played, continuing during the facial expression recognition task.
Mood-congruent music has been shown to enhance emotional responding, as well as to
increase activity in a number of brain regions thought to relate to automatic processing of
emotional information (e.g. bilateral amygdala; Baumgartner, Lutz, Schmidt, & Jancke, 2006;
Dyck et al., 2011). The music selected for each induction group was based on pilot-testing with
older and younger adults and chosen based on its efficacy and acceptability across both age
groups. For the anxious experimental group, the piece selected was Shostakovich’s String
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 34
Quartet No. 8 in C Minor (Op.110), Allegro Molto movement; for the calm control condition, the
piece selected was Debussy’s Claire de Lune.
Emotion recognition and attention task. Following the initial mood induction
(induction 1), the VAS and STAI-S were administered again to assess the efficacy of the mood
induction. Participants then completed the facial expression recognition task, involving facial
expression targets displayed one at a time. Images were presented for 5 seconds and in
randomized order to prevent sequence effects. Images and instructions were administered using
the stimulus presentation software, SuperLab Version 4 (Cedrus Corporation), and projected on a
screen via an overhead projector connected to the experimenter’s laptop computer. For each face,
participants were asked to select the emotion (happy, angry, fear, sad, neutral) expressed in the
face they had just seen and, subsequently, to rate the intensity of that emotion on a 9-point scale
(“1, low intensity” to “9, high intensity”) via pen and paper response format (Kellough & B.G.
Knight, 2012). Based on differences in average response time by age group observed in pilot-
testing, older participants were allotted 20 seconds between image presentations to record their
responses, while younger participants received 15 seconds. Prior to each new face presentation,
the experimenter confirmed that all individuals had completed their response and were once
again attending to the overhead screen. The abbreviated stimulus presentation and response time
were chosen in order to evaluate more rapid processing of the stimuli.
After completion of the facial recognition task, participants again completed the VAS and
STAI-S measures of state anxiety in order to assess for the continuing presence of anxiety post-
task. A subsequent “booster” of the mood induction procedure (induction 2) was completed in
order to maintain mood state prior to the administration of the attention measure. Participants
were again asked to call to mind a mood-relevant autobiographical memory and the mood
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 35
congruent music was played during this time. Following the mood “booster” and subsequent
mood check (VAS and STAI-S), participants completed the Connections task.
At the end of the study procedures, the experimenter debriefed all participants on the
nature of the study and answered any questions regarding the experiment. In order to facilitate
participants’ return to a relaxed state, those randomized to the anxiety induction were then led
through a relaxation procedure, involving a guided breath-focus meditation and calm mood
music (that used for the calm control group). The experimenter checked that all participants had
returned to or below their baseline level of anxiety based on either their final VAS or STAI-S
mood check; for participants that showed discrepant scores, the experimenter confirmed with the
individual verbally that they felt as they did when they had arrived. A licensed clinical
psychologist serving as co-investigator on this protocol was available by phone in order to assist
in the event that any participants needed immediate further assessment, and flyers with the
contact information for USC counseling services were available for any interested participants,
though neither of these services were required in any instance.

Analytic Plan
The effectiveness of the mood induction (Hypothesis 1) was assessed using repeated
measures analysis of variance (ANOVA) to examine the effect of mood condition on Visual
Analog Scale and STAI-S ratings, with time as a within-subjects factor and induction group (two
levels: anxious and calm) and age group (two levels: younger and older) as between-subjects
factors. To test Hypotheses 2 and 3, univariate ANOVAs were used to examine the between-
group effects of age and induction group on mean accuracy and intensity ratings for each type of
emotional facial expression (anger, fear, happy, sad). Because we had no a priori hypotheses
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 36
about the effects of age or induction group on accuracy or intensity ratings for neutral faces, this
category is not analyzed below.
1
In order to control for the increased risk of Type I error due to
multiple tests, a Bonferonni correction was applied, resulting in an a priori alpha-level threshold
of p < .0125 for each test.
To test Hypothesis 4, exploring the potential moderating effects of trait anxiety (two
levels: high and low trait anxiety), univariate ANOVAs were employed to test the three-way
between-subjects interactions between age and induction groups and this variable, with mean
accuracy and intensity ratings for each type of emotional facial expression (angry, fear, happy,
and sad) as dependent variables. For Hypothesis 5, the same analytic method was used to test the
three-way between-subjects interactions between age and induction groups and attentional
control ability (two levels: high and low attentional control) for each type of emotional facial
expression (angry, fear, happy, sad) as dependent variables. Secondary analyses explored the
relationship between trait anxiety and attentional control level across age groups using a phi
coefficient to correlate these two binary variables.

Results
Characteristics of the Sample
Of the 76 younger adults and 66 older adults who participated in the protocol, 3 younger
adults and 2 older adults, all of whom were randomized to the calm induction group, were
excluded as a result of recent (within the last month) mood or anxiety disorder and/or use of
selective serotonin reuptake inhibitors (SSRIs) self-reported on the day of the testing session.
This resulted in a sample size of 73 younger adults and 64 older adults.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 37
Demographic variables. Baseline demographic measures are reported in Table 1 (age in
years, self-reported health) and Table 2 (gender, ethnicity, occupation, education, income).
As expected, younger participants had significantly lower age in years (M = 19.70 years,
SD = 1.14) than older participants (M = 71.58, SD = 6.90; t(66.01) = -59.44, p < .0001).
Interestingly, there were no significant differences by age group in self-rated health (younger: M
= 4.04, SD = 0.74, older: M = 4.05, SD = 0.85, t(133)=-0.04, p = .965). There were no
differences in age in years or self-rated health by induction group within each age group (all p’s
> .05).
Chi square tests indicated that there were significantly more women than men across age
groups (X
2
(1, N = 137) = 6.14, p = .013) and a higher proportion of men in the older group than
the younger group (X
2
(1, N = 137) = 6.14, p = .013). Within age groups, an equal proportion of
men and women were randomized to each induction group. An equal proportion of men were
randomized to the calm induction group by age group; the same was true for women randomized
to the calm induction group. However, the proportion of older men randomized to the anxious
group was higher than the younger group, and the proportion of younger women randomized to
the anxious group was higher than the older group (X
2
(1, N = 137) = 7.95, p = .008).
Chi square tests indicated that there were significantly more Asian and Latino
participants in the younger than the older age group, and significantly more Caucasian
individuals in the older than the younger age group (X
2
(4, N = 137) = 40.68, p < .0001). Within
age groups, an equal proportion of individuals from each ethnicity were randomized to each
induction group.
Regarding occupation, all younger participants categorized themselves as “students,”
whereas approximately three-quarters of older adults identified themselves as “retired,” and the
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 38
other quarter of older individuals endorsed the “other” category, presumably to indicate they
were still working in some capacity. Within each age group, there were no differences in the
proportion of individuals within occupation categories randomized to each induction category.
All younger participants were categorized as having completed some college (12-15
years education) while approximately three-quarters of older individuals reported higher levels
of education (16 years of education or more). As a result, there were a greater proportion of older
individuals in almost all education categories except “12-15 years education,” which had a
higher proportion of younger individuals (X
2
(4, N = 136) = 88.95, p < .0001). The highest
educated category (20 or more years of education) had significantly more older individuals
randomized to the anxious than the calm induction (p < .05). Beyond this exception, in each age
group, there were no differences in the proportion of individuals within education categories
randomized to each induction category.
More than three-quarters of younger participants reported their yearly income as
“<$10,000” while the majority of older adults indicated a higher yearly income. Chi-square tests
revealed significant differences in the proportion of individuals from each income category by
age group for all but the two highest yearly income brackets (X
2
(5, N = 129) = 98.00, p < .0001).
Across induction groups, a higher proportion of younger adults reported an income of
“<$10,000” than older adults, while a higher proportion of older adults reported incomes of
“$10,000-$49.999” and “$50,000-$99,999” than younger adults (p < .05). In addition, within the
anxious induction, a greater proportion of older adults rated their income as “$100,000-
$199,999” than younger adults (p < .05). Within age groups, an equal proportion of individuals
from each income category were randomized to each induction group, with the exception of the
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 39
“$100,000-$199,999” category, where a greater proportion of older adults receiving this amount
per year were randomized to the anxious versus calm induction group (p < .05).
Trait anxiety and baseline state anxiety. Following previous research (Hogan, 2003;
L.O. Lee & B.G. Knight, 2009), younger participants evidenced significantly higher levels of
trait anxiety (BAIT: t(126.02) = 3.01, p = .003) and baseline state anxiety than older adults prior
to mood induction procedures (VAS: t(129.26) = 5.58, p < .0001; STAI-S: t(121.26) = 6.45, p <
.0001; see Table 1).

Hypothesis 1. Efficacy of the Mood Induction Procedures
Efficacy of the mood induction procedures, whole sample. To test Hypothesis 1, the
efficacy of the mood induction procedures as measured by state anxiety measures (VAS and
STAI-S) were first examined by comparing change in scores from baseline [T1] to post-
induction [T2] in the whole sample using repeated-measures ANOVAs, with age (younger,
older) and induction group (anxious, calm) as between-subjects factors and time (T1, T2), as a
within-subjects factor. There was a significant between-subjects main effect of age group (VAS:
F(1, 133) = 37.96, p < .0001, η
p
2
= .222; STAI-S scores (F(1, 133) = 33.13, p < .0001, η
p
2
=
.199) as well as between-subjects main effects of induction group (VAS: F(1, 133) = 41.83, p <
.0001, η
p
2
= .239; STAI-S (F(1, 133) = 20.19, p < .0001, η
p
2
= .132) on state anxiety scores, but
no significant between-subjects age by induction group interactions (VAS: p = .607; STAI-SF10:
p = .921). Scores indicated that younger adults reported higher anxiety scores than older adults
across induction groups and time points and that those randomized to the anxious induction
group had higher scores on measures of state anxiety following the induction (T2) than those
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 40
randomized to the calm induction group; there was no difference between induction groups at
baseline (T1).
Critically, significant within-subjects induction group by time interactions on anxiety
scores were also observed (VAS: F(1, 133) = 99.74, p < .0001, η
p
2
= .429; STAI-S scores: F(1,
133) = 77.74, p < 00001, η
p
2
= .369). Across both measures of state anxiety, individuals
randomized to the calm induction group had lower anxiety scores and individuals randomized to
the anxious score had higher anxiety scores following the induction procedures (see Figure 1).
Thus, the results of these analyses support Hypothesis 1, suggesting that an anxious mood
induction procedure, involving mood-congruent autobiographical recall and music, was effective
in increasing state anxiety ratings in both older and younger adults assigned to the anxious mood
condition as compared to the calm control condition. Because no significant three-way age by
induction group by time interactions were observed (VAS: p = .134; STAI-S: p = .080), we
conclude that the mood induction procedures effects on state anxiety scores from baseline to
post-induction were similar across age groups.
Inclusion criteria based on change in state anxiety from baseline to post-induction.
Though change in mood in the predicted directions with mood induction procedures was shown
across the whole sample as above, a review of the data indicated that a number of individuals,
both older and younger, were either unresponsive or had unexpected responses (i.e. in the reverse
direction) to the mood inductions. Because of the potential for the responses of these individuals
to dilute any effects of mood state in subsequent analyses, we elected a priori to include only
those who were responsive to the induction in our final data analysis. This type of minimum
change threshold is similar that adopted in other mood induction research conducted with
younger and older adults (Larcom & Isaacowitz, 2009).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 41
Within the anxious induction group, individuals who became less anxious or did not
change in anxiety levels from baseline [T1] to post-induction [T2], based on either the VAS scale
or STAI-S, were considered to have been unresponsive to the induction and were removed from
further analysis. Within the calm control group, individuals who had had an increase in anxiety
scores following induction, either on the VAS scale or STAI-S, were removed from further
analysis as their mood trajectory was considered to mimic the anxious group and run counter to
the objective of the control condition. This resulted in the removal the results of 11 younger and
13 older adults in the anxious induction group, as well as the results of 5 younger and 4 older
adults in the calm induction group. Thus, our responsive to induction sample included a total of
57 younger adults (25 in the anxious and 32 in the calm induction group) and 47 older adults (20
in the anxious and 27 older adults in the calm induction group).
Analyses of the baseline characteristics of the remaining sample were repeated as above
(see Table 3 and Table 4). There were very few differences in the pattern of characteristics (age
in years, self-reported health, gender, ethnicity, occupation, education, income, trait and state
anxiety measures) by age or induction group within the responsive subset as compared to the
whole sample. Of note, younger adults randomized to the calm control group within this smaller
sample were found to have significantly higher baseline STAI-S scores than age-matched peers
randomized to the anxious condition (younger calm group: M = 19.44, younger anxious group:
M = 16.16, t(55) = 2.32, p = .024). In addition, within the responsive sample, chi-square tests
indicated there were significantly more African American participants in the older as compared
to the younger group (p < .05), a difference not found in the entire sample. Beyond these
dissimilarities, scores within the responsive group mimicked patterns seen in the whole sample.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 42
Efficacy of the mood induction procedures, responsive sample. The efficacy of the
mood induction as measured by state anxiety ratings was assessed using repeated-measures
ANOVAs, with age (younger, older) and induction group (anxious, calm) used as between-
subjects factors and time (baseline [T1], post-induction [T2], following facial emotion
identification task [T3], post-booster induction [T4]) as a within-subjects factor for those who
met our minimum threshold for being responsive to the induction procedures (Figures 2 and 3).
There was a significant between-subjects main effect of age group on both VAS (F(1,
100) = 20.67, p < .0001, η
p
2
= .171) and STAI-S scores (F(1, 100) = 30.53, p < .0001, η
p
2
=
.234), with post-hoc t-tests revealing that younger adults scored significantly higher than older
adults on VAS scores at all time points except post-booster induction (p = .066) and across all
time points for the STAI-S (all p’s < .05). Results of the repeated-measures ANOVAs for VAS
and for STAI-S scores indicated that the induction procedures were effective in producing
significant differences in state anxiety by induction group in the predicted directions (VAS: F(1,
100)=59.96, p < .0001, η
p
2
= .375; STAI-S (F(1, 100) = 39.26, p < .0001, η
p
2
= .282; see Figures
2 and 3). For both measures of state anxiety, there was no significant age by induction group
interaction (VAS: p = .542; STAI-S: p = .911).
Within-subjects effects of time, age group, and induction group were also observed. Due
to the violation of sphericity in the repeated-measures ANOVA for VAS scores [Mauchly’s test:
p < .05, ε > .75], the Hyunh-Feldt correction was applied to the within-subjects effect analyses
for this measure. This revealed a significant three-way interaction of time by age group by
induction group for VAS scores (F(5.80, 284.93) = 2.73, p = .047, η
p
2
= .027). For STAI-S
scores, the sphericity assumption for repeated measures ANOVA was met (Mauchly’s test: p >
.05). A within-subjects three-way interaction (time by age group by induction group) effect was
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 43
also found for STAI-S scores (F(3, 300) = 6.82, p < .0001, η
p
2
= .064), indicating that the
trajectories of anxiety scores were different by both age and induction groups across time.
Post-hoc within-subjects one-way ANOVAs using simple contrasts was used to compare
scores at each timepoint (post-induction [T2], following facial emotion identification task [T3],
post-booster induction [T4], and post-debriefing/end of study procedures [T5]) to baseline (T1).
Results revealed that both younger and older adults randomized to the anxious induction group
had significantly higher VAS and STAI-SF scores relative to baseline [T1] across all study
procedure timepoints (post-induction [T2], following facial emotion identification task [T3], and
post-booster induction [T4]; all p’s < .02). This suggests that the trajectory of the anxious
induction over time was roughly similar by age group.
With the calm control condition, younger adults had significantly lower VAS and STAI-S
scores as compared to baseline [T1] across all subsequent time points (post-induction [T2],
following facial emotion identification task [T3], and post-booster induction [T4]; all p’s < .005).
In contrast, older adults appeared to demonstrate a shortened duration of response to the calm
condition over time, returning to baseline levels of state anxiety as measured by both the VAS
and the STAI-S scale at T3 (following the facial emotion recognition task: F(1, 26) = 0.53, p =
.475 and F(1, 26) = 2.65, p = .116, respectively) as well as at T4 for STAI-S scores (post-booster
induction F(1, 26) = 3.10, p = .090). Because of increasing anxiety scores in the calm induction
group, older adults in the calm group no longer differed significantly in VAS anxiety scores
compared to those in the anxious induction group at T3 (calm group: M = 1.72, SD = 2.34;
anxious group: M = 2.70, SD = 2.22, t(45) = -1.45, p = .154). There was a trend for anxious
older adults’ STAI-S scores (M = 16.5, SD = 4.15) to remain higher than calm older adults’
scores (M = 14.17, SD = 3.84) at T3, though this did not reach significance (t(45) = -1.97, p =
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 44
.055). Thus, while the calm induction appeared effective in reducing the state anxiety level of
older adults randomized to this condition based on scores immediately following induction, older
adults in this group appeared not to have maintained this change in mood state over time as
successfully as younger adults, as demonstrated by their return to baseline levels of state anxiety
by the end of the facial emotion recognition task (T3).
According to within-subjects one-way ANOVAs using simple contrasts comparing
change in VAS Anxiety and STAI-S scores from baseline (T1) to post-debriefing/end of study
procedures (T5; following a relaxation exercise for those in the anxious group), older and
younger participants anxiety scores in both induction groups had returned to their baseline levels,
or lower, by the end of the study procedures.
Magnitude of the effect of mood induction procedures on state anxiety by age group. The
effect size for change in anxiety scores from baseline [T1] to post-induction [T2] between
induction groups from was calculated for each age group. Within the younger adult subgroup, the
effect size for mean change in VAS scores between the anxious and calm groups was found to be
Cohen’s d = 2.49 (95% CI: 1.88 - 3.11), and for STAI-S was found to be Cohen’s d = 2.53 (95%
CI: 1.93 - 3.13), representing large effect sizes (J. Cohen, 1992). A similar magnitude effect size
for mean change in VAS from baseline to post-induction 1 was found within the older adult
subgroup (d = 2.63 [95% CI: 1.85 - 3.40]). However for STAI-S scores, the mean change effect
size between groups was slightly lower for older adults as compared to the younger age group (d
= 2.14 [95% CI: 1.53 – 2.75]), likely driven by older adults’ reduced response to the calm control
procedures on this measure.
To summarize, for those who met our a priori criteria for responsiveness to the initial
induction, as with the whole sample, results indicated that the study procedures were effective in
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 45
increasing state anxiety ratings in both older and younger adults assigned to the anxious as
compared to the calm mood condition. Within the subsample responsive to mood induction
procedures, both older and younger adults demonstrated mood score changes to the initial
induction of approximately the same magnitude despite differences in baseline state anxiety.
However, based on significant within-subjects three-way effects (age group by induction group
by time) and post-hoc analyses, it appears that younger adults were slightly more reactive to
induction procedures, and particularly the calm control condition, throughout the protocol as
compared to older adults.

Hypotheses 2 and 3. Accuracy and Intensity Ratings by Age and Induction Group
Data preparation. Accuracy for each type of facial expression of emotion (anger, fear,
happy, sad) was calculated by dividing the number of correctly identified facial expression
within an emotion category by the total number of targets in that category, resulting in a
proportion of total correct identifications for each facial expression group. Subjective intensity
ratings (range [low to high]: 1-9) for each type of facial expression of emotion (anger, fear,
happy, sad) were calculated by averaging intensity ratings scores across those instances when an
emotion was correctly-identified for each participant.
We had elected to use a threshold of 25% or more mean accuracy within each facial
expression type for inclusion of a participant’s score in a given analysis. This was done in order
to prevent those with extremely low accuracy scores (at or below chance) from biasing results.
Across fear, happy, and sad faces, all participants, older and younger, were at least 25% accurate
in their identification of the emotion; thus, no participant’s data were excluded in these analyses.
Within the anger facial expression category, however, some individuals’ accuracy scores did not
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 46
satisfy the 25% threshold. In total, 3 older adults’ scores (one older adult in the anxious group,
two older adults in the calm group) were not included in subsequent analyses related to anger
faces as a result of very low accuracy scores.
Accuracy in facial emotion identification by age and induction groups. Between-
subjects ANOVAs were used to examine the effects of age and induction group on accuracy for
each type of facial expression overall (anger, fear, happy, sad). In order to control for the
increased risk of Type I error due to multiple tests, a Bonferonni correction was applied, with
four individual tests resulting in an a priori alpha-level threshold of p < .0125. Results are
presented in Table 5.
Replicating past research (Isaacowitz et al., 2007; Ruffman et al., 2008), there were
significant effects of age group in the identification of negatively-valenced faces, in all cases
favoring younger adults. Older adults in our sample were less accurate in identifying anger
(older: M = .625, SD = .193, younger: M = .746, SD = .158, F(1, 97) = 10.30, p = .002, η
p
2
=
.096), fear (older: M = .785, SD = .189, younger: M = .915, SD = .111, F(1, 100) = 18.75, p <
.0001, η
p
2
= .158), and sad (older: M = .753, SD = .164, younger: M = .890, SD = .134, F(1, 100)
= 25.90, p < .0001, η
p
2
= .206) emotional faces overall than younger adults. As expected, there
were no differences by age group in the accurate identification of happy faces (older: M = .950,
SD = .103, younger: M = .978, SD = .060, F(1, 100) = 2.58, p = .111). Overall, these results
largely support Hypothesis 2(i).
Induction group main effects were observed for accuracy in the identification of sad
faces. Surprisingly, calm individuals were more likely to correctly identify sad faces than
anxious individuals (F(1, 100) = 13.51, p < .0001, η
p
2
= .119). However, there was no main
effect of induction group on accuracy for anger, fear, or happy faces, and no significant age
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 47
group by induction group interaction effects observed on accuracy across facial expression type
(all p’s > .05). Contrary to our hypotheses, older and younger adults within the anxious induction
group were not significantly more accurate than age-matched peers in the calm condition for any
facial emotion expression type, including anger and fear faces. Thus, we did not find support for
either Hypothesis 2(ii) or Hypothesis 2(iii).
Intensity Ratings for Facial Emotion Expression by Age and Induction Groups.
Between-subjects ANOVAs were used to examine the effects of age and induction group on
average subjective intensity ratings for each type of correctly-identified facial expression (anger,
fear, happy, sad). Again, a Bonferonni correction was applied and an a priori alpha-level set at p
< .0125 for all tests. Results are presented in Table 6.
When age differences were observed, they suggested that older adults perceived stronger
emotional intensity across several emotion categories. Older adults rated happy faces as more
intense than younger adults (older: M = 6.96, SD = 1.02, younger: M = 6.47, SD = .90, F(1, 100)
= 7.07, p = .009, η
p
2
= .066). The same pattern was seen for sad faces, with older adults rating
sad faces as more intense than younger adults (older: M = 6.51, SD = 1.11, younger: M = 5.83,
SD = .89, F(1, 100) = 11.71, p = .001, η
p
2
= .105) as well as for anger faces (older: M = 6.83, SD
= 1.07, younger: M = 6.27, SD = .97, F(1, 97) = 7.31, p = .008, η
p
2
= .070), but not fear faces
(older: M = 5.81, SD = 1.27, younger: M = 5.98, SD = .91, F(1, 100) = .67, p = .417). Thus,
Hypothesis 3(i) received partial support, with older adults rating happy faces as more intense
than younger adults. However, the finding that older adults perceived sad faces and anger faces
as more intense than younger adults, and that the two age groups did not differ in mean intensity
ratings for fear faces, was not predicted by our hypothesis.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 48
A significant induction group main effect on intensity ratings was observed for one of the
two threat-relevant facial expression categories. Anxious individuals rated fear faces as more
intense (M = 6.21, SD = 1.16) than those in the calm control group, as predicted by Hypothesis
3(ii) (M = 5.67, SD = 0.98, F(1, 100) = 6.56, p = .012, η
p
2
= .062). However, differences in
intensity ratings by induction group for anger faces did not reach statistical significance (p =
.063). Mood induction group did not significantly influence intensity ratings of happy or sad
faces (p = .485 and p = .365, respectively). Thus, Hypothesis 3(ii), that an anxious mood
induction would result in higher intensity ratings for threat-relevant facial expressions of
emotions was only supported by results for fear, but not anger, faces. In addition, there were no
significant age by induction group interaction effects observed on subjective intensity scores
across facial expression type, indicating that older and younger adults judgment of intensity were
affected similarly by induction procedures, contrary to our hypothesis. These results provide
partial support for Hypothesis 3(ii) and fail to support Hypothesis 3(iii).

Hypothesis 4. Accuracy and Intensity Ratings by Age, State and Trait Anxiety Groups
Data preparation. As discussed previously, theory and research indicate that mood-
congruent biases to threat in state anxiety may be most pronounced in individuals who have a
self-reported trait tendency to anxiety. Based on this, we predicted that differences in mood-
congruent biases in emotion identification and judgment by induction and age group might be
most evident when comparing those with low and high trait anxiety levels. Therefore, we
examined potential differences in accuracy and intensity ratings by state anxiety (induction
group) and age group as a function of baseline trait anxiety, measured by the Beck Anxiety
Inventory-Trait (BAIT).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 49
BAIT scores were stratified into two groups using the median score by age group (older
adults’ BAIT median: 4; younger adults’ BAIT median: 7). Those scoring above the median for
their age group were considered to endorse “high trait anxiety” and those below, “low trait
anxiety.” Within the younger adults group, 26 individuals were categorized as “high trait
anxiety” (13 in the anxious and 13 in the calm induction group) and 28 individuals as “low trait
anxiety” (12 in the anxious and 16 in the calm induction group).
Within the older adults group, 24 individuals were categorized as “high trait anxiety” (9
in the anxious and 15 in the calm induction group) and 22 individuals as “low trait anxiety” (10
in the anxious and 12 in the calm induction group). Those with median scores were not included
in the analysis (four individuals: three younger adults in the calm group, one older adult in the
anxious group). In addition, one older adult in the calm induction group did not complete the
BAIT, and was therefore excluded from these analyses. Chi square tests indicated that there
were no significant differences in the proportion of participants with high or low attentional
control assigned to each induction condition within each age group (all p’s < .75). Because of
the relatively small sample size per cell and the exploratory nature of this hypothesis, we elected
to maintain an a priori alpha-level of p < .05 as our threshold for statistical significance in the
analyses below.
Accuracy and intensity ratings in facial emotion identification by age, state anxiety,
and trait anxiety groups. To begin, three-way, between-subjects ANOVAs [2 (age: younger,
older) x 2 (state anxiety: anxious, calm) x 2 (trait anxiety: high, low)] were conducted on
accuracy and intensity scores for each type of facial expression (anger, fear, happy, sad). As
main effects of age group and induction group, as well as their interaction, are reported above,
these are not repeated here. There were no significant main effects of trait anxiety group, nor 2-
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 50
way interactions of trait anxiety by induction group, on either accuracy or intensity ratings for
any facial expression type (all p’s > .05). However, significant three-way interactions (age by
induction by trait anxiety group) for accuracy and intensity ratings were found with respect to
threat-relevant faces, as described below.
Accuracy by age, state anxiety, and trait anxiety. Regarding accuracy, a three-way
interaction was found on accuracy for fear faces (F(1, 92) = 4.71, p = .033, η
p
2
= .049; Figures 4a
and 4b), though not anger faces (p = .836), happy (p = .439), or sad faces (p = .681). Post-hoc
2x2 ANOVAs (age group x trait anxiety group) were used to examine the significant three-way
interaction on accuracy for fear faces at each level of state anxiety (anxious, calm conditions)
separately. A significant age by trait anxiety interaction in the identification of fear faces (F(1,
40) = 7.82, p = .008, η
p
2
= .163) was observed for individuals in the anxious induction (see
Figure 4a).
Within the anxious induction condition, scores suggested that high trait anxiety younger
adults were less accurate (M = 0.904, SD = 0.075) than low trait anxiety younger adults (M =
.979, SD = .049) in identifying fear faces, while high trait anxiety older adults (M = .889, SD =
.098) were more accurate than low trait anxiety older adults (M = .725, SD = .262) at identifying
these faces (see Table 7a). No effects of trait anxiety group, or age by trait anxiety group
interactions were seen in the calm (non-state anxious) group for accuracy for fear faces (p = .331
and p = .786, respectively; see Figure 4b and Table 7b). A main effect of age was found
suggesting that older adults in the calm condition were less accurate (M = .776, SD = .177) in
identifying fear faces within our stimuli set than younger adults in the calm condition (M = .901,
SD = .127, F(1, 52) = 8.82, p = .005, η
p
2
= .145; see Figure 4b and Table 7b).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 51
These results partially support Hypothesis 4(i), suggesting that trait anxiety moderates the
effects state anxiety, and operates differently across age groups, with high trait and state anxious
older adults performing better than low trait and state anxious older in the identification of at
least one type of threat-relevant face, fear. However, that low trait and state anxious younger
adults were observed to be more accurate in fear face identification than high trait and state
anxious younger adults within the anxious induction group runs contrary to our predictions in
Hypothesis 4(i).
Intensity ratings by age, state anxiety, and trait anxiety. A three-way interaction was
observed for intensity ratings for both anger faces (F(1, 89) = 4.36, p = .040, η
p
2
= .047; see
Figures 5a and 5b) and fear faces (F(1, 92) = 5.17, p = .025, η
p
2
= .053; see Figures 6a and 6b),
but not for happy (p = .454), or sad faces (p = .230). As above, post-hoc 2x2 ANOVAs (age
group x trait anxiety group) were used to examine these significant three-way interactions for
intensity ratings at each level of state anxiety (anxious, calm). A similar pattern as seen with
accuracy was observed, such that age by trait anxiety group interactions were found in ratings for
threat-related faces for those in the state anxious, but not the calm, condition.
Within the anxious induction condition, significant age by trait anxiety interactions in
subjective ratings of anger (F(1, 39) = 8.45, p = .006, η
p
2
= .178; see Figure 5a) and fear faces
(F(1, 40) = 5.88, p = .020, η
p
2
= .128; see Figure 6a) were found. For those randomized to the
anxious mood condition, scores suggested that high trait anxiety younger adults (M = 6.23, SD =
1.00) perceived anger faces as less intense than low trait anxiety younger adults (M = 6.94, SD =
0.76); in contrast, high trait anxiety older adults (M = 7.37, SD = 0.80) rated anger faces and as
more intense than low trait anxiety peers (M = 6.41, SD = 1.11; see Table 8a). The same pattern
was seen for fear faces, with high trait anxiety younger adults randomized to the anxious
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 52
condition rating fear faces as less intense (M = 5.86, SD = 1.19) than low trait anxiety peers (M =
6.77, SD = 0.64), and high trait anxiety older adults (M = 6.37, SD = 0.920) rating anger faces
and as more intense than low trait anxiety peers (M = 5.68, SD = 1.46; see Table 8a). No effects
of trait anxiety group, or age by trait anxiety group interactions were seen on intensity ratings for
those in calm induction group for either anger (p = . 567 and p = .958, respectively) or fear faces
(p = .481 and p = .560, respectively; See Figures 5b and 6b and Table 7b).
Together, these findings also partially support Hypothesis 4, suggesting that trait anxiety
moderates the effects of age on intensity ratings for those who are state anxious, with high trait
anxiety older adults perceiving more intensity in threat-relevant emotions (anger and fear) than
older adults with low trait anxiety. Like accuracy, findings for younger adults subjective
intensity ratings in the anxious induction group ran contrary to our hypothesis, with high trait
anxiety younger adults appearing to rate these threat-relevant emotions as less intense than low
trait anxiety younger adults, and not the reverse, as predicted.

Hypothesis 5. Accuracy and Intensity Ratings by Age, State Anxiety, and Attentional
Control Groups
Data preparation. Attentional functioning has been found to be associated with
emotional processing biases in older adults (Isaacowitz et al., 2009). Attentional control was
assessed in this study using the Connections Test (Salthouse et. al, 2000). As described
previously, scores on the simple (A) and alternating (B) versions of the task were computed for
each participant by calculating the number of correct connections made minus the number of
errors (omissions or commissions) for each test and then averaging across version types
(Salthouse et al., 2000).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 53
In order to help to control for age differences in speed, we divided the alternating average
score (Connections Task B) for each participant by their simple condition average score
(Connections Task B) in order to compute a ratio (B/A). The use of this ratio has been shown to
lessen the direct relationship of this measure to speed, as well as to reduce its relationship with
age to near zero (Salthouse, 2011). Individuals’ scores on the simple (A) conditions of this task
are typically higher than on the more difficult alternating (B) version, resulting in a B/A ratio
less than 1; this was the case for all but two participants in our sample. Values closer to 1
represent higher efficiency on a more complicated attention task involving switching (alternating
(B) version) relative to a more simple attention task (simple (A) condition).
Using ANOVA to test for the effects of age and induction groups on the B/A ratio score,
we found no significant main effects of age (F(1, 100) = 2.39, p = .125) or induction group (F(1,
100) = 1.41, p = .238) and no significant age by induction group interaction (F(1, 100) = 0.04, p
= .837), suggesting that neither age nor induction group had a significant influence on this
measure of attention. Errors in either condition, simple (A) or alternating (B) Connections Tasks,
were relatively infrequent considering that, unlike classic Trails Making Tests, commissions and
omissions are not corrected by the test administrator during the task. There were no significant
differences by age group in the total number of errors (omissions plus commissions) made in
either the simple (A) condition (younger: M = 0.54 SD = 1.36, older: M = 0.40, SD = 1.23, F(1,
100) = 0.28, p = .600) or the alternating (B) condition (younger: M = 2.11, SD = 4.30, older: M =
2.09, SD = 3.12, F(1, 100) = 0.001, p = .981). In addition, no effects of induction group nor age
by induction interaction reached significance for either the simple (A) condition or alternating
condition (B) (all p’s > .05). Because there were no induction effects on Connections Task B/A
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 54
ratio scores or error frequency, B/A ratio scores were used as an individual-differences variable
in post-hoc analyses across induction groups as below.
In order to assess the effects of attentional control on accuracy and subjective intensity
ratings for emotional faces by age and induction group, we first analyzed the distribution of ratio
scores in each age group in order to stratify participants’ attentional performance. The median
score of each age group was used to categorize individuals as either above or below the 50
th

percentile in that subgroup (younger median: 0.7600, older median: 0.7895). Individuals above
the median for their age group were categorized as having “high attentional control” and those
below the median as having “low attentional control.”
Within the younger adults group, 28 individuals were categorized as “high attentional
control” (16 in the anxious and 12 in the calm induction group) and 28 individuals as “low
attentional control” (9 in the anxious and 19 the in the calm induction group). Within the older
adults group, 23 individuals were categorized as “high attentional control” (9 in the anxious and
14 in the calm induction group) and 23 individuals as “low attentional control” (10 in the anxious
and 13 in the calm induction group). Those with median scores were not included in the analysis
(1 older adult in the anxious induction group).
Chi square tests indicated that there were no significant differences in the proportion of
participants with high or low attentional control assigned to each induction condition within each
age group (all p’s < 0.1). While at first glance it appears that more “low attentional control”
younger individuals were randomized to the calm versus the anxious induction, chi-square tests
indicate that the proportion of “ low attentional control” younger individuals assigned to each
induction condition was not significantly different than of “low attentional control” older
individuals (X
2
(1, N = 51) = 0.694, p = .405). Because of the relatively small sample size per
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 55
cell and the exploratory nature of this hypothesis, we had elected to maintain an alpha-level of p
< .05 as our threshold for statistical significance in the analyses below.
Accuracy and intensity ratings in facial emotion identification by age, state anxiety,
and attentional control groups.
Accuracy by age, state anxiety, and attentional control. A three-way, between-subjects
ANOVA [2 (age: younger, older) x 2 (state anxiety: anxious, not anxious) x 2 (attentional
control: high, low)] was conducted to test for higher-order interaction effects on accuracy scores
for each emotion type (anger, fear, happy, sad). As differences by age and state anxiety have
been discussed earlier, the results of age and induction group main effects on accuracy will not
be repeated here. There was a main effect of attentional control group on fear face accuracy
scores (F(1, 94) = 4.83, p = .030, η
p
2
= .049), which was subsumed by a significant two-way
interaction of age by attentional control group on fear faces accuracy (F(1, 100) = 6.62, p = .012,
η
p
2
= .066; see Figure 7). While younger state individuals did not differ in their accuracy for fear
faces by attentional control ability (low attentional control: M = .902, SD = .115, high attentional
control: M = .924, SD = .109, t(54) = -0.745, p = .459), older individuals did (see Figure 7).
Within the older age group, those with higher attentional control were less accurate in
identification of fear faces than older adults with lower attentional control as hypothesized (low
attentional control: M = .853, SD = .149, high attentional control: M =.723, SD = .206; see Table
9). There was no significant effects of attentional control group, no two-way interactions of
attentional control with age or induction group, and no three-way interactions of attentional
control by age by induction group on accuracy scores for anger, happy, or sad faces (all p’s >
.10).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 56
These results provide mixed support for Hypothesis 5(i), with attentional control
correlating negatively with accuracy for fear faces in older adults, such that low attentional
control older adults were more accurate than age-matched high attentional control peers.
However, contrary to our prediction, this relationship was not moderated significantly by mood
induction group. As hypothesized, the effect of attentional control appeared to be stronger in our
older sample, as attentional control level did not significantly affect the accuracy of younger
anxious adults’ performance in fear identification.
Intensity ratings by age, state anxiety, and attentional control. Similar to that done for
accuracy, a 3-way between-subjects ANOVA (with age, induction, and attentional control as
between-subjects factors) was completed to examine higher-order effects on intensity ratings
across emotion types (anger, fear, happy, sad).
There was a main effect of attentional control group on intensity ratings for anger faces
(F(1, 91) = 5.04, p = .027, η
p
2
= .052), which was subsumed by a significant two-way interaction
of induction by attentional control group on anger faces accuracy (F(1, 91) = 5.50, p = .021, η
p
2

= .057; see Figure 8). This pattern of results was also observed for fear faces, with a main effect
of attentional control group on intensity ratings (F(1, 94) = 5.88, p = .017, η
p
2
= .059), also
subsumbed by a significant two-way interaction of induction by attentional control group (F(1,
94) = 4.78, p = .031, η
p
2
= .048; see Figure 9). There was no significant effects of attentional
control group, nor two-way interactions of attentional control with age or induction group on
intensity ratings for happy or sad faces (all p’s > .10) and no three-way interactions of attentional
control by age by induction group on intensity ratings for fear, happy, or sad faces (all p’s > .10).
The three-way interaction of attentional control by age by induction group on intensity ratings
for anger faces approached significance (F(1, 91) = 3.41, p = .068, η
p
2
= .036).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 57
Individuals randomized to the calm induction did not differ in intensity ratings by
attentional control group for either anger (low attentional control: M = 6.32, SD = 1.08, high
attentional control: M = 6.41, SD = 1.04; see Figure 8) or fear faces (low attentional control: M =
5.70, SD = 0.97. high attentional control: M = 5.63, SD = 1.02; see Figure 9). However,
individuals randomized to the anxious induction demonstrated significant differences in intensity
ratings for anger and fear faces by attentional control group. Those with high attentional control
in the anxious induction group rated anger faces as less intense (M = 6.36, SD = 0.98) than those
with low attentional control (M = 7.28, SD = 0.80; see Table 10). This same pattern was seen for
fear faces, with those scoring high on attentional control perceiving less intensity (M = 5.86, SD
= 1.19) than low attentional control individuals (M = 6.79, SD = 0.78; see Table 10).
These results provide mixed support for Hypothesis 5(ii), suggesting that, among those in
the anxious induction group, individuals with low attentional control rate correctly-identified
threat-relevant emotions (anger, fear) as more intense than those with high attentional control.
As predicted, the influence of attentional control did not appear in calm group. However,
contrary to our prediction, the relationship between attentional control and intensity ratings for
threat-relevant faces was not significantly different between the two age groups, as demonstrated
by the lack of three-way interactions of age with induction and attentional control group.

Post-Hoc Analysis. The Relationship Between Trait Anxiety and Attentional Control by
Age Group
Our findings thus far demonstrate that trait anxiety influences accuracy and intensity
ratings for threat-relevant emotions, with high trait anxiety seeming to increase older adults’ and
reduce younger adults’ accuracy for fear faces and intensity ratings for fear and anger faces when
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 58
in an anxious mood. Similarly, attentional control ability also related to accuracy and intensity
ratings for threat-relevant emotions, with older adults with low attentional control demonstrating
greater accuracy for fear faces, and adults with low attentional control across age groups rating
anger and fear faces as more intense when in an anxious mood state. Thus, our findings
regarding the effects of trait anxiety and attentional control within the older adult subsample
suggested a negative relationship between these two variables, such that the overall pattern for
the high trait anxiety group was similar to that of low attentional control groups (i.e. increased
accuracy and perceived intensity for threat-relevant facial expressions), while a clear pattern
between trait anxiety and attentional control was not apparent in our younger adult subsample.
Past research suggests a negative relationship between trait anxiety and attentional control
(Derryberry & Rothbart, 1988), particularly in older adults (D. Cohen, Eisdorfer, Vitaliano, &
Bloom, 1980; Deptula, Singh, & Pomara, 1993), prompting our investigation of the link between
trait anxiety and attentional control by age group within our sample.
A phi coefficient was calculated in order to examine the relationship between trait
anxiety and attentional control group status, two binary variables. Within the whole sample, trait
anxiety level was not significantly correlated with attentional control ability (p = .069).
However, dividing the sample by age group revealed a pronounced relationship between trait
anxiety level and attentional control in the older adult group. Trait anxiety and attentional
control group within older adults in this sample was found to show a strong negative correlation
(phi = -.422, p = .005). This suggests that older adults with high trait anxiety within our sample
also tended to have lower performance on our measure of attention, while those with low trait
anxiety obtained higher attentional control scores. The connection between trait anxiety and
attentional functioning group was not seen in our younger adult sample (p = .909). These
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 59
findings support past research suggesting that cognitive control and trait anxiety level are
negatively associated in older individuals, but may not be related in younger individuals.

Discussion
To the best of our knowledge, this study offered a unique exploration into the relationship
between anxiety, age, and attention in the interpretation of facial expressions of emotion. Using
a laboratory-based anxious mood induction procedure, we examined the effects of state anxiety
on both younger and older adults’ accuracy for, and perceived intensity ratings of, emotional
faces. In addition, we sought to examine theoretically-important potential moderators of the
effects of age and anxious mood state on accuracy and intensity ratings, including trait anxiety
level and attentional control ability. Specifically, we hypothesized that increased state anxiety
level would result in facilitated stimulus-driven processing and threat-relevant biases for certain
facial expressions of emotion (anger, fear) favoring accurate identification and increased
perceived intensity across age groups. In line with theory and research, we predicted that the
effects of anxious mood state would be most pronounced among those with a dispositional
diathesis towards anxiety. In addition, due to anxiety’s effects on attentional functioning, we
predicted an interaction between attentional control ability and mood state, such that individuals
with low attentional control might be disproportionately affected by state anxiety, resulting in
heightened accuracy and intensity ratings for threat-relevant emotions. Assuming anxiety would
interrupt “default” attentional goals typically allocated towards positive information in older
individuals, we predicted the hypothesized interaction between attentional control and anxious
mood state might be most evident in older adults in our sample.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 60
Our first hypothesis, which predicted that the use of mood induction procedures
involving mood-congruent autobiographical memory recall and music would result in significant
differences in state anxiety for those assigned to the anxious mood condition as compared to a
calm mood control condition in both older and younger adults, was confirmed. Within the whole
sample, our mood procedures were effective, producing significant differences in anxious mood
state by assignment post-induction across age groups: younger and older individuals randomized
to the anxious condition showed increased state anxiety relative to baseline as well as to those
assigned to the calm condition. The efficacy of the mood induction across age groups is in
keeping with a number of research studies that have found autobiographical mood induction
procedures to be successful in eliciting specific moods (e.g. sad, happy) in younger and older
adults (Levenson et al., 1991; Poon & B.G. Knight, 2009).
Our second hypothesis, based on facilitated threat-identification in state anxious mood,
predicted that accuracy for facial expression identification would vary by age and mood state. In
particular, we hypothesized that state anxious mood would benefit both younger and older adults
in the accurate identification of threat-relevant facial expressions of emotion, such as anger and
fear. This hypothesis was not supported. We replicated age differences in facial expression
identification (Isaacowitz et al., 2007; Ruffman et al., 2008), with younger adults outperforming
older individuals in the identification of negative, but not happy, faces. However, there was no
main effect of induction on accuracy for threat-relevant faces (anger, fear), nor did induction
group moderate emotion recognition accuracy across age groups. Based on research and theory
suggesting that anxious individuals perceive heightened threat in stimuli, our third hypothesis
predicted that perceived intensity of facial expressions would also vary by age and anxious state.
We expected a main effect of age for ratings of emotional faces, with younger adults rating
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 61
negative emotions as more intense and positive emotions as less intense than older individuals.
In addition, we predicted that those who were state anxious would rate correctly-identified
threat-relevant facial expressions of emotion (anger, fear) as more intense than individuals in the
calm condition.
Hypothesis 3 received mixed support. While older adults did perceive positive emotions as more
intense than younger adults, as anticipated, older individuals also perceived negative emotions
(anger, sad) as more intense than younger adults. This suggests that older individuals in our
sample showed a non-specific elevation in perceived intensity for emotional faces relative to
younger participants, similar to that seen by other researchers (Gruhn & Scheibe, 2008).
Following our prediction, though, anxious induction was associated with increased perceived
intensity in correctly-identified fear faces across age groups relative to calm induction
participants. To our knowledge, this is the first study to demonstrate that current anxious mood
state in older adults might contribute to increases in the perception of threat in fear faces, a
finding that has been demonstrated in younger adults (Liebman & Allen, 1995). However,
induction group membership did not affect perception of anger, as had been expected. That the
effects of induction on intensity ratings for facial expressions were not moderated by age group
within our study suggests that the tendency for state anxiety to elicit mood-congruent appraisals
of intensity for fear facial expressions may operate similarly across age groups.
Looking next at the role of individual differences in reactivity to anxiety in Hypothesis 4,
we found that trait anxiety level was a moderator of the effects of state anxiety and age on
accuracy and intensity ratings for threat-relevant faces. In line with a diathesis-stress model in
which the effects of anxiety are more pronounced in those with an underlying propensity, we had
hypothesized that younger and older adults with the dual load of state and trait anxiety would be
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 62
most likely to demonstrate to threat-relevant cognitive biases, such as higher accuracy and
intensity ratings for threat-relevant emotional faces, than low trait anxious individuals within the
anxious induction group.
It appears that while trait anxiety was a significant moderator of the effects of state
anxiety and age group, the effects of this trait on emotion perception operated in opposing
directions by age group. The hypothesized mood-congruent pattern was generally confirmed in
our older adult subsample, with high trait and state anxious older adults identifying fear faces
more accurately, as well as rating correctly-identified anger and fear faces as more intense than
low trait, high state anxious peers. However, the opposite was observed in younger adults.
Contrary to our hypothesis, younger high trait and state anxious individuals demonstrated a
tendency to misidentify fear faces and to rate accurately-identified fear and anger faces as less
intense relative to low trait, high state anxious peers. Thus, when facing state anxiety, high trait
anxious older adults appeared to demonstrate mood-congruent biases, while high trait anxious
younger adults seemed to display mood-incongruent process of threat-relevant emotions, relative
to low trait anxious peers.
Previous research explicitly examining the effects of state or trait anxiety in emotion
processing across age groups is limited. However, in one of the only studies directly comparing
the attentional biases of older and younger adults selected for anxiety levels, results have also
demonstrated mood-congruent anxious biases in older, but not younger, adults. In their study of
the effects of age and trait anxiety on mood congruent-attentional biases, L.O. Lee and B.G.
Knight (2009) found that trait anxiety moderated attentional biases for older adults in the
processing of certain negative faces (e.g. sad faces), but that younger individuals did not display
attentional biases to emotional faces across any level of trait anxiety (high, moderate, or low).
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 63
While more research is clearly needed, the results of this study and that of L.O. Lee and B.G.
Knight (2009) suggest that anxious mood-congruent attentional biases in anxiety might operate
differently in younger and older adults, with older adults demonstrating more pronounced mood-
congruent biases.
Examining our fifth hypothesis, our findings indicated that individual differences in
attentional control ability moderated the effects of state anxiety across age groups. Based on
theory and research, we hypothesized that the effects of state anxious mood would be most
evident in those with poor attentional functioning. Thus, we expected that younger and older
adults with state anxiety and low attentional control would be most likely to show increased
threat-relevant stimulus processing, as demonstrated by higher accuracy and intensity ratings for
threat-relevant emotional faces those with high attentional control. We predicted this pattern to
be more pronounced in older adults, due to reduced attentional capacity coupled with changes in
attentional goals as a result of anxious mood state. This pattern was only partially confirmed.
As expected, older adults with low attentional control identified fear faces more
accurately than high attentional control peers, though this effect was observed across induction
groups. Younger adults did not appear to differ by attentional control ability in accuracy for fear
faces. Differences in the effects of attentional control by age group, irrespective of current mood
state, may be placed in context of literature on age and memory attention for emotional
information. Past research has established that individual differences in attentional control
ability play an important role in older adults’ attention to positive information. Only older adults
with high attentional functioning appear to deploy visual attention towards positive and away
from negative faces, while older adults with low attentional control do not (Isaacowitz et al.,
2009; Noh, Lohani, & Isaacowitz, 2011; for a review, see Isaacowitz, 2012). In contrast,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 64
research suggests that attentional control ability and attentional load may not moderate younger
adults’ patterns of visual attention for emotional information (Noh et al., 2011; Mather & M.
Knight, 2005, Experiment 3; but see Isaacowitz et al., 2009). Taken together, research suggests
that general attentional abilities may relate to patterns of attention processing for emotional
stimuli in older, more so than younger, individuals, as in our study. Thus, our findings extend
the results of past research on the interaction of cognitive control and age in emotional attention
and memory to emotion identification for fear faces.
Regarding intensity ratings, state mood, but not age, interacted with attentional control to
influence evaluations of correctly-identified threat-relevant emotions. Both older and younger
adults who were state anxious and had low attentional control rated threat-relevant stimuli as
more intense than those who had high attentional control or were not state anxious (i.e. in the
calm induction group). This suggests that, across age groups, once state anxious individuals have
accurately-identified a threat-relevant emotion, those with low attentional control show
interpretative biases, either amplifying the intensity of this threat or being unable to down-
regulate the perceived magnitude of the threat, relative to peers with high attentional control.
These findings extend work on the relationship between anxiety and attentional control
(Derryberry & Reed, 2002; Reinholdt-Dunne et al., 2009) to the evaluation of threat, suggesting
the effects of low attentional control on engagement with negative information in those
experiencing anxiety might operate similarly in older and younger individuals.
That the effect of attentional control ability on fear face identification was not moderated
by state anxiety may relate to the possibility that fear faces are a powerful enough image to
capture attention in older adults with low attentional control across mood groups, and so do not
involve the increased processing thought to be associated with anxious mood. Future research
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 65
should attempt to clarify this finding and the inconsistencies observed in our pattern of results for
accuracy and perceived intensity in order to determine the types of emotional tasks in which
attentional control is most relevant (e.g. memory, judgment), and the role of age and anxious
mood state in these effects.
Finally, because of literature relating trait anxiety to attentional control ability
(Derryberry & Rothbart, 1988) and our observed effects, we examined the link between trait
anxiety and attentional control by age group within our sample. We found that categorical ratings
of trait anxiety level (high/low) and attentional control (high/low) demonstrated a strong negative
correlation in older adults, but were unrelated in younger participants. This finding is in line with
past research that has suggested that older individuals may be more susceptible to the effects of
experienced anxious mood on cognition, while younger adults may not be adversely affected (D.
Cohen et al., 1980; Deptula et al., 1993). The correlational nature of the relationship observed in
our study cautions against making any causal interpretations from these results. However, that
the mood-cognition relationship appears to have been stronger in older as compared to younger
adults in our sample is an important finding, which may explain why, within the limited research
literature on negative mood and emotion-processing across age groups, effects are seen in older
but not younger adults (e.g. L.O. Lee & B.G. Knight, 2009).
We suggest that the inverse relationship between high levels of state and trait anxiety on
accuracy as well as intensity ratings for threat-relevant emotional stimuli in younger versus older
adults observed in this study may be explained by differences in the ability of high trait anxious
younger and older individuals to regulate their emotions when they experience anxious mood
state. High trait anxiety represents an enduring tendency, characterized by longstanding patterns
of cognition and behavior. Coupled with our post-hoc findings that trait anxiety and attentional
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 66
control level were negatively-associated in older but not younger individuals, it seems possible
that one’s ability and/or resources to manage this tendency in the face of current anxious mood
may change with age.
The lack of correlation between trait anxiety and attentional group in younger individuals
in our study suggests that both low and high trait anxious individuals may have similar
attentional resources available to them when facing an anxious situation. However, we speculate
that anxious younger adults with high trait anxiety, who more frequently experience anxious
responding, may be better practiced in mobilizing cognitive resources, such as avoidance and
reappraisal, in order to limit their engagement with negative information and manage mood,
when facing an anxious mood relative to low trait anxious peers. This might explain why, when
anxious, high trait anxious younger adults in our study appeared to demonstrate reduced
engagement with threat-stimuli relative to low trait peers.
In contrast, among anxious older adults, those with high trait anxiety experienced mood-
congruent responding. Trait anxiety and attentional group demonstrated a strong negative
correlation in older adults in our sample, with those with high trait anxiety more likely to score
below the median in their attentional functioning ability. Thus, although older adults high in trait
anxiety would also be expected to have had more experience with anxiety than low trait anxious
peers, it seems that they may also have a reduced ability to modulate attention in order to
override the automatic biases that accompany state anxious mood. Based on the results of the
study, we suggest that older adults high in trait anxiety, possibly as a result of reduced attentional
control, may not be as able as peers with low trait anxiety to mitigate their processing of threat
information when anxious.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 67
Taken together, the results of our study suggest that trait anxiety might differentially
influence emotion-processing in those who are anxious across age group, though the reasons for
these differences (e.g. attentional control) remain speculative. As a result, we posit that a
tendency towards anxious responding (as in trait anxiety) might be protective in the short-term,
preparing younger individuals to more effectively mood-repair when anxious, but that a lifetime
of hyper-reactivity may have erosive affects on attentional control and older adults’ ability to
attenuate engagement with threat-stimuli when anxious. Such a developmental hypothesis would
explain the age differences in responding to threat-relevant information among high trait anxious
individuals. Future research should explicitly test if trait anxiety influences attentional processes
and engagement with emotional information over time.
In our study, trait anxiety and attentional control each interacted with state anxiety and
age group to affect processing of fear and anger faces, but did not show a general negativity
effect as no influences were seen on sad faces. The relatively circumscribed moderating effects
of trait anxiety and attentional control by emotion type can be understood by dimensional
conceptualizations of affect as well as discrete emotions theory. Anger and fear are thought to
involve key evolutionary reactions to threat (activation of the “fight or flight” response to
confront or avoid aversive stimuli) and to share valence and arousal dimensions, both negative
and involving heightened levels of activation (Russell & Feldman-Barrett, 1999). Thus, anger
and fear faces may be particularly relevant to anxious mood states, both because they serve as
threat signals to which anxious individuals are motivated to attend, and because they are mood-
congruent with anxiety. However, from a discrete emotions perspective, fear is more closely
associated with general threat and the experience of anxiety than is anger (for discussions, see
Levine & Pizarro, 2004 and Richards et al., 2002). Models of mood-congruence (e.g. Bower,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 68
1981) would predict that because fear is more similar to anxiety, the associative networks
connected with fear expressions may be more activated than those associated with anger faces in
those who are anxious. This would lead to strengthened memory for and judgment of fear
relative to anger faces, as observed in our study.
Adding to this line of reasoning, while processing of both fear and anger faces has been
found to be associated with increased neural activity in the amygdala, fear faces have been more
strongly linked to amygdala activation than anger expressions (for a review, see Adolphs, 2002).
As high levels of anxiety have been also been shown to be associated with increased amygdala
reactivity relative to normal controls (Stein, Simmons, Feinstein, & Paulus, 2007), this suggests
that the increased amygdala activity associated with anxiety may relate more strongly to
interpretation of fear faces than anger. As a result, it is not surprising that mood-congruent
processing biases emerged more consistently in the evaluation of fear faces within our study.
The relative specificity of the neurological activity profile associated with anxiety would also
explain the lack of effects of anxious mood state on sad face recognition and appraisal in the
current study. Though a negative mood, sadness has been found across a number of studies to
most robustly involve other brain regions, such as the subcallosal cingulate cortex (SCC; Phan,
Wager, Taylor, & Liberzon, 2002), over the amygdala. And, though not examined in our study,
differences in neural activation by emotion type may well explain past research that has
demonstrated preservation in the recognition of the emotion, disgust, with age. Disgust, like
happiness, appears to predominantly involve brain areas besides the amygdala (e.g., basal
ganglia; Phan et al., 2002). Thus, the ability to recognize discrete facial expressions of emotion
may depend on distinct combinations of experienced mood state and neural activation, which
may change or remain the same with advancing age depending on the specific mood type, not
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 69
just its valence. Our results support the assertion that anxiety has a relatively distinct profile, and
not just a negativity bias, in its effects on emotion processing.

Limitations
While this study offered an initial examination of the effects of age, anxiety, and attention
in emotion processing, there are notable limitations to our approach. The lack of enduring effects
of the calm induction in older adult participants, resulting in non-significant differences in
anxiety by induction group following the face recognition task, may have obscured main effects
of state anxiety, as well as any interactions with age, in accuracy for threat faces and/or
perceived intensity for anger faces as predicted by Hypotheses 2 and 3. In addition, as a
considerable amount of data (approximately 25%) was not included in the current analysis due to
those participants’ lack of response to the initial mood induction procedures, our statistical power
to detect differences by anxiety and age group in our outcome measures may have been reduced.
Future research examining the effects of anxious mood state on facial emotion processing across
age groups would do well to find a control procedure equally effective in maintaining calm mood
in both younger and older individuals in order to determine if state anxiety influences emotion
processing outcomes across age groups.
Another criticism related to the interpretation of results in this study is our use of a
between-subjects design. As a result of this approach, we are limited to interpreting differences
found between mood conditions relative to the other group (i.e. anxious individuals are “higher”
or “lower” in scores relative to calm individuals). In order to more clearly determine if transient
anxiety “improves” or “impairs” identification, it would be useful to employ a within-subjects
design, contrasting the effects of both anxious and calm mood state on emotion processing in
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 70
each participant. However, we note that a within-subjects design also has limitations, including
practice and order effects associated with repeated testing, and the possibility of more
pronounced demand characteristics in situations where the two conditions being compared are
highly salient to participants. Thus, we assert that our between-groups design provides a
reasonable first step in examining the potential effects of anxious mood state on emotion
processing across age groups, which should be further scrutinized in crossover designs.
In our study, we observed between-group differences in the effects of age, anxiety, and
attentional control on accuracy and intensity ratings for specific, threat-relevant faces (anger,
fear). While many of the findings obtained largely support past research and theory (e.g.
attentional bias to threat), we therefore speculate on the causal mechanisms of differences
observed based on relevant literature. However, in this study, we did not directly observe the
specific processes through which the proposed attentional biases may have operated to affect
emotion identification and perception, which past research indicates might include different
fixational patterns to certain areas of target faces (e.g. Wong et al., 2005) or facial muscle
mimicry in the rater (e.g. Bailey, Henry, & Nangle, 2009). The use of a group-administered
protocol in our study precluded the measurement of such variables as reaction time, eye-tracking,
or electromyography. As there are many possible pathways by which attentional biases to threat
might operate, a number of which are measurable, examining the processes that appear to have
linked age, anxiety, and cognitive abilities to identification and interpretive biases for threat
would be an important next step in understanding the causes of these biases.
As mentioned, due to the cross-sectional nature of the data collected, our findings linking
trait anxiety and attentional control in older, but not younger, adults cannot tell us about the
relative contribution of one factor in the development of the other. The connection between self-
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 71
reported trait anxiety and performance on a task of attentional control in older adults invites the
question of whether reductions in attentional control bring about increases in anxious responding
(because of reduced ability to inhibit negative thoughts and information), or if a tendency
towards anxious reactivity, over time, taxes the attentional control system. Longitudinal research
of these two variables in aging individuals would be a valuable next step in determining the
causal directionality of anxiety and attentional control over time.
Finally, the sample used in this study may not be representative of younger and older
adults in the general population, particularly in regards to educational achievement. Both
younger and older adults in our sample were, on average, very highly educated, with the
overwhelming majority having obtained more than a high school degree, and more than half of
older participants endorsing having received some level of post-graduate education. That
participants in our study had higher educational levels may have particular implications for our
examination of differences by attentional control in light of research that suggests that attentional
control is positively related to academic achievement (Rueda, Checa, & Rothbart, 2010). Thus,
our sample could be predicted to have higher and more homogeneous attentional abilities than a
normative sample. It is possible that the effects of attentional control on emotion processing
across age and mood groups might be more pronounced in a representative sample.

Theoretical and Practical Implications
Despite these limitations, the results of the current protocol provide new information
about the effects of anxiety and attentional functioning on facial emotion perception across the
lifespan. Our results have both theoretical and applied implications regarding aging, mood, and
the processing of emotional information. The effects of anxiety in older adults have historically
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 72
been over-looked in aging research, particularly by those studying normal aging. However, the
prevalence of sub-syndromal anxiety symptoms in non-clinical older adult populations appears
to be relatively common, and, by some reports, the “norm” (in one review, anxiety symptoms in
community-dwelling older samples ranged from 15% to 52.3%; Bryant, Jackson, & Ames,
2008). That the effects of anxious mood in those with trait anxiety led to mood-congruent biases
in accuracy and perception in older individuals suggests that anxiety may lead to the attenuation
of the positivity affect in older individuals, potentially as a result of changes in motivated
attention. Thus, our study points the possibility that anxious temperament and current anxiety
may limit the positive motivational biases predicted by SST, possibly as a result of reduced
attentional control. Our results add support to the Strength and Vulnerability Integration (SAVI)
theoretical model proposed by Charles (2010), suggesting that while older adults may typically
demonstrate more effective emotion regulation than younger adults, when older adults
“experience high levels of sustained emotional arousal…age-related advantages in emotional
well-being are attenuated, and older adults are hypothesized to have greater difficulties returning
to homeostasis” (p. 1068).
In addition, our results replicate past findings that lower attentional control ability
increases older adults’ processing of negative facial expressions of emotion. While this idea may
initially seem counterintuitive, these findings align with the work of other researchers who have
examined the limits of the positivity effect based on the availability of cognitive resources
(Mather & M. Knight, 2005; Isaacowitz et al., 2009). Older adults with high attentional control
may benefit in their ability to attend to positive information at the expense of the negative,
whereas older adults with low cognitive control, due to a reduced ability to filter out the
negative, may actually demonstrate better processing of negative information.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 73
From a practical perspective, limiting one’s attention to positive information may be
advantageous when the goal is mood regulation. However, there are a number of situations in
which failing to engage with negative information may pose a danger. The recognition of
emotion states in others is important to evaluating the context of social interaction, helping the
viewer to determine if another person, or the surrounding environment in which both find
themselves, suggests safety or is a threat to well-being. The ability to recognize emotional faces,
specifically fear and shame, is associated with important interpersonal decision-making
processes such as deceit detection, with older adults’ impairments in the identification of such
faces shown to, in part, explain their increased vulnerability to deception (J.T. Stanley &
Blanchard-Fields, 2008). Thus, elucidating the factors that may facilitate the identification of
fear faces in older adults’, such as anxiety and attentional ability, could have important
implications for understanding and promoting older adults’ ability to protect themselves from
fraud when in face-to-face interactions. Our results suggest that anxiety and reduced attentional
control, at face value undesirable attributes, may sometimes serve as protective factors for older
adults in situations in which the goal is to detect danger, boosting threat detection and promoting
appropriate responding.
Finally, the results of this research have implications for therapeutic intervention with
younger and older individuals. One of the most effective psychosocial treatment approaches for
late-life anxiety is cognitive behavioral therapy (CBT; Ayers, Sorrell, Thorp, & Wetherell, 2007;
Wolitzky-Taylor et al., 2010). CBT, and other cognitive therapies, are based on the premise that
mood and cognition are linked: negative mood precipitates negative information-processing
biases and vice versa. Cognitive therapies are thought to work by “undoing” these links,
teaching individuals to attend to and interpret information in a less negatively-biased and more
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 74
emotionally-neutral or realistic manner (Beck et al., 1979). Based on this proposed mechanism,
cognitive therapies should only be effective in those who demonstrate negative mood-cognition
coupling. That older adults with high trait anxiety in the current study appeared to show more
pronounced mood-congruent biases when anxious than those with low trait anxiety suggests that
this group might logically benefit from cognitive interventions.
Our study was cross-sectional and cannot account for intra-individual change over time.
However, a tendency towards anxiety has been shown to be moderately stable over the lifespan
(Conley, 1988; Roberts & DelVecchio, 2000). Thus, it seems likely that many younger, trait
anxious individuals will be among those with the highest levels of this attribute as older adults.
If mood-congruent attentional and interpretive biases for those with trait anxiety are more
pronounced in older individuals, as the results of this study suggest, then intervening to modify
trait anxiety early in the lifespan may have long-term benefits.
Lastly, the results of this study demonstrated that both younger and older individuals with
low attentional control may be more prone to negative mood-congruent interpretive biases than
those with high attentional control. In recent years, therapeutic interventions have been
developed that attempt to re-train emotional attention. In what is often referred to as attention
training (Hallion & Ruscio, 2011), the same tasks used to assess mood-congruent cognitive
biases (e.g. dot probe tasks) have been used to alter these biases (for a review, see MacLeod &
Mathews, 2012). A number of studies have found that by having participants engage in a
modified dot-probe task in which the target more frequently replaces positive-stimuli, individuals
experiencing high levels of anxiety develop a positive attentional bias, a change that is associated
with symptom improvement (Amir, Beard, Burns, & Bomyea, 2009; Hazen, Vasey, & Schmidt,
2009) which may last over months (Schmidt, Richey, Buckner, & Timpano, 2009). There is
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 75
preliminary evidence to suggest that attention modification training is as effective in changing
healthy older adults’ visual attention to emotional information and mood as it is younger adults’
(Isaacowitz & Choi, 2011). While more research is needed, the current results provide indirect
support that applying such techniques therapeutically to aid younger and older adults with trait
anxiety in controlling their attention, and consequently, their mood might offer a potential
clinical intervention.

Summary
To our knowledge, this is the first study to use a mood induction procedure to
demonstrate the effects of anxiety in the perception of facial emotion expressions across age
groups. In addition, by exploring the role that attentional control has in the relationship between
mood, age, and emotion perception, this study increases knowledge about the interplay between
cognition and mood in social decision-making across the lifespan, and further delineates the
situations in which the positivity effect does, and does not, operate. Taken together, the results of
this study suggest that anxiety has distinct effects on older and younger individuals’ processing
of threat-relevant faces of emotions.
Our findings demonstrate that older individuals predicted to be most inclined to
attentional biases to threat, such as those with concomitant trait and state anxiety and those with
low attentional control, perform better in accurately identifying threat-relevant facial
expressions, such as fear faces. Since older adults typically report less anxiety than younger
adults in general, this implies that age differences in accuracy for negative, and particularly
threat-relevent, emotions found in past research may, in part, be artifacts of differences in mood
state, general propensity to anxiety, and/or attentional ability, between age groups. In addition,
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 76
our findings regarding perceived intensity ratings for anger and fear faces suggest that older and
younger individuals display biases in their evaluation of threat-relevant facial stimuli when in an
anxious mood state, but the nature and direction of these biases may differ by age, trait anxiety
level, and attentional functioning ability.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 77
References
Adolphs, R. (2002). Recognizing emotion from facial expressions: Psychological and
neurological mechanisms. Behavioral and Cognitive Neuroscience Reviews, 1(1), 21-62.
doi: 10.1177/1534582302001001003

Amir, N., Beard, C., Burns, M., & Bomyea, J. (2009). Attention modification program in
individuals with Generalized Anxiety Disorder. Journal of Abnormal Psychology,
118(1): 28-33. doi: 10.1037/a0012589

Ayers, C.R., Sorrell, J.T., Thorp, S.R., & Wetherell, J.L. (2007). Evidence-based psychological
treatments in late-life anxiety. Psychology and Aging, 22(1), 8-17. doi: 10.1037/0882-
7974.22.1.8

Bailey, P.E., Henry, J.D., & Nangle, M.R. (2009). Electromyographic evidence of age-related
differences in the mimicry of anger. Psychology and Aging, 24(1), 224-229. doi:
10.1037/a0014112

Bar-Haim, Y.,Lamy, D., Pergamin, L., Bakermans-Kranenburg, M.J., & van IJzendoorn, M.H.
(2007). Threat-related attentional bias in anxious and nonanxious individuals: a meta-
analysis. Psychological Bulletin, 133(1), 1-24. doi: 10.1037/0033-2909.133.1.1

Barlow, D.H. (2000). Unraveling the mysteries of anxiety and its disorders from the perspective
of emotion theory. American Psychologist, 55(11), 1247-1263. doi: 10.1037/0003-
066X.55.11.1247

Baumgartner, T., Lutz, K., Schmidt, C.F., & Jancke, L. (2006). The emotional power of music:
how much enhances feeling of affective pictures. Brain Research, 1075, 151-164. doi:
10.1016/j.brainres.2005.12.065

Beaudreau, S.A., & O’Hara, R. (2009). The association of anxiety and depressive symptoms with
cognitive performance in community-dwelling older adults. Psychology and Aging,
24(2), 507-512. doi: 10.1037/a0016035

Beck, A.T., Rush, A.J., Shaw, B.F., & Emery, G. (1979). Cognitive therapy of depression. New
York: Guilford Press.

Beck, A.T, & Clark, D.A. (1997). An information processing model of anxiety: automatic and
strategic processes. Behavior Research Therapy, 35(1), 49-58. doi: 10.1016/S0005-
7967(96)00069-1

Beck, A.T., Epstein, M., Brown, G., & Steer, R.A. (1988). An inventory for measuring clinical
anxiety: psychometric properties. Journal of Consulting and Clinical Psychology, 56(6),
893-897. doi: 10.1037/0022-006X.56.6.893

Blaney, P.H. (1986). Affect and Memory: A review. Psychological Bulletin, 99(2): 229-246.
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 78

Bower, G.H. (1981). Mood and memory. American Psychologist, 36(2), 129-148. doi:
10.1037/0003-066X.36.2.129

Bradley, B.P., Mogg, K., & Millar, N.H. (2000). Covert and overt orienting of attention to
emotional faces in anxiety. Cognition and Emotion, 14(6), 789-808. doi:
10.1080/02699930050156636
Bryant, C., Jackson, H., & Ames, D. (2008). The prevalence of anxiety in older adults:
Methodological issues and a review of the literature. Journal of Affective Disorders, 109,
233-250. doi: 10.1016/j.jad.2007.11.008

Calder, A.J., Keane, J., Manly, T., Sprengelmeyer, R., Scott, S., Nimmo-Smith, I., & Young,
A.W. (2003). Facial expression recognition across the adult life span. Neuropsychologia,
41, 195–202. doi: 10.1016/S0028-3932(02)00149-5

Carstensen, L.L., Fung, H.H., & Charles, S.T. (2003). Socioemotional Selectivity Theory and the
regulation of emotion in the second half of life. Motivation and Emotion, 27(2), 103-123.
doi: 10.1023/A:1024569803230

Carter, C. S., Barch, D. M., Gur, R., Gur, R., Pinkham, A., & Ochsner, K. (2009). CNTRICS
final task selection: Social cognitive and affective neuroscience-based measures.
Schizophrenia Bulletin, 35(1), 153–162. doi: 10.1093/schbul/sbn157

Cavanaugh, JC, & Blanchard-Fields, F. (2010). Chapter 6: Attention and memory. In J. Perkins
(Ed.), Adult Development and Aging, (6th ed., 184-232). Belmont CA: Wadsworth,
Cengage Learning.

Charles, S.T., & Campos, B. (2011). Age-related changes in emotion recognition: How, why,
and how much of a problem? Journal of Nonverbal Behavior, 35, 287-295. doi:
10.1007/s10919-011-0117-2

Charles, S.T., & Carstensen, L.L. (2009). Social and emotional aging. Annual Review of
Psychology, 61, 383-409. doi: 10.1146/annurev.psych.093008.100448

Charles, S.T. (2010). Strength and Vulnerability Integration: A model of emotional well-being
across adulthood. Psychological Bulletin, 136(6), 1068-1091. doi: 10.1037/a0021232

Cohen, D., Eisendorfer, C., Vitaliano, P.P., & Bloom, V. (1990). The relationship of age,
anxiety, and serum immunoglobulins with crystallized and fluid intelligence. Biological
Psychiatry, 15(5), 699-709.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155-159. doi: 10.1037/0033-
2909.112.1.155

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 79
Conley, J.J. (1984). Longitudinal consistency of adult personality: Self-reported psychological
characteristics across 45 years. Journal of Personality and Social Psychology, 47(6),
1325-1333. doi: 10.1037/0022-3514.47.6.1325

Cooper, R.M., Rowe, A.C., Penton-Voak, I.S. (2008). The role of trait anxiety in the recognition
of emotional facial expressions. Journal of Anxiety Disorders, 22, 1120-1127. doi:
10.1016/j.janxdis.2007.11.010

Damasio, A.R., Everitt, B.J., & Bishop, D. (1996). The Somatic Marker Hypothesis and the
possible functions of the prefrontal cortex. Philosophical Transactions: Biological
Sciences, 351(1346), 1414-1420. doi: 10.1098/rstb.1996.0125

Davis, F.C., Somerville, L.H., Ruberry, E.J., Berry, A.B.L., Shin, L.M., Whalen, P.J. (2011). A
tale of two negatives: Differential memory modulation by threat-related facial
expressions. Emotion, 11(3), 647-655. doi: 10.1037/a0021625

Dennis, R.E., Boddington, S.J.A., & Funnel, N.J. (2007). Self-report measures of anxiety: are
they suitable for older adults? Aging & Mental Health, 11(6), 668-677. doi:
10.1080/13607860701529916

Deptula, D., Singh, R., & Pomara, N. (1993). Aging, emotional states, and memory. The
American Journal of Psychiatry, 150(3), 429-434. Retrieved from
http://ajp.psychiatryonline.org/journal.aspx?journalid=13

Derakshan, N., & Eysenck, M.W. (2009). Anxiety, processing efficiency, and cognitive
performance: New developments from attentional control theory. European Psychologist,
14(2), 168-176. doi: 10.1027/1016-9040.14.2.168

Derryberry, D., & Reed, M.A. (2002). Anxiety-related attentional biases and their regulation by
attentional control. Journal of Abnormal Psychology, 111(2), 225-236. doi:
10.1037/0021-843X.111.2.225

Derryberry, D., & Rothbart., M.K. (1988). Arousal, affect, and attention as components of
termperament. Journal of Personality and Social Psychology, 55(6), 958-966. doi:
10.1037/0022-3514.55.6.958

Dyck, M., Loughead, J., Kellermann, T., Boers, F., Gur, R.C., & Mathiak, K. (2011). Cognitive
versus automatic mechanisms of mood induction differentially activate left and right
amygdala. Neuroimage, 54, 2503-2513. doi: 10.1016/j.neuroimage.2010.10.013

Ebner, N.C., & Johnson, M.K. (2010). Age-group differences in interference from younger and
older emotional faces. Cognition and Emotion, 24(7): 1095-1116. doi:
10.1080/02699930903128395

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 80
Egloff, B., & Hock, M. (2001). Interactive effects of state and trait anxiety on emotional Stroop
interference. Personality and Individual Differences, 31, 875-882. doi: 10.1016/S0191-
8869(00)00188-4

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

Eysenck, M.W., Derakshan, N., Santos, R., & Calvo, M.G. (2007). Anxiety and cognitive
performance: attentional control theory. Emotion, 7(2), 336-353. doi: 10.1037/1528-
3542.7.2.336

Eysenck, M.W. (1992). Anxiety: The cognitive perspective. East Sussex, UK: Lawrence
Erlbaum Associates Ltd., Publishers.

Feldman, R.S., Philippot, P., & Custrini, R.J. (1991). Chapter 9: Social competence and
nonverbal behavior. In R.S. Feldman & B. Rime (Eds.), Fundamentals of Nonverbal
Behavior, (329-350). New York, NY: Press Syndicate of the University of Cambridge.

Fiske, A., Wetherell, J.L., & Gatz, M. (2009). Depression in older adults. Annual Review of
Clinical Psychology, 5, 363-389. doi: 10.1146/annurev.clinpsy.032408.153621

Fox, E., Russo, R., Bowles, R., & Dutton, K. (2001). Do threatening stimuli draw or hold visual
attention in subclinical anxiety? Journal of Experimental Psychology: General, 130(4):
681-700. doi: 10.1037/0096-3445.130.4.681

Fox, L.S., & Knight, B.G. (2005). The effects of anxiety on attentional processes in older adults.
Aging and Mental Health, 9(6), 585-593. doi: 10.1080/13607860500294282

Fox, L.S., Knight, B.G., & Zelinski, E.M. (1998). Mood induction with older adults: A tool for
investigating effects of depressed mood. Psychology and Aging, 13(3), 519-523. doi:
10.1037/0882-7974.13.3.519

Frenkel, T.I., Lamy, D., Algom, D., & Bar-Haim, Y. (2009). Individual differences in perceptual
sensitivity and response bias in anxiety: Evidence from emotional faces. Cognition and
Emotion, 23(4): 688-700. doi: 10.1080/02699930802076893

Gatz, M., Reynolds, C.A., John, R., Johansson, B., Mortimer, J.A., & Pedersen, N.L. (2002).
Telephone screening to identify potential dementia cases in a population-based sample of
older adults. International Psychogeriatrics, 14(3), 273-289. doi:
10.1017/S1041610202008475

Georgiou, G., Bleakley, C., Hayward, J., Russo, R., Dutton, K., Eltiti, S., & Fox, E. (2005).
Focusing on fear: Attentional disengagement from emotional faces. Visual Cognition,
12(1), 145-158. doi: 10.1080/13506280444000076

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 81
Gruhn, D., & Schiebe, S. (2008). Age-related differences in valence and arousal ratings of
pictures from the International Affective Picture System (IAPS): Do ratings become more
extreme with age? Behavior Research Methods, 40(2), 512-521. doi:
10.3758/BRM.40.2.512

Gur, R. C., Sara, R., Hagendoorn, M., Marom, O., Hughett, P., Macy, L., …, & Gur, R.E.
(2002). A method for obtaining 3-dimensional facial expressions and its standardization
for use in neurocognitive studies. Journal of Neuroscience Methods, 115(2), 137–143.
doi: 10.1016/S0165-0270(02)00006-7

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. doi:
10.1016/j.actpsy.2006.01.008

Hallion, L.S., & Ruscio, A.M. (2011). A meta-analysis of the effect of cognitive bias
modification on anxiety and depression. Psychological Bulletin, 137(6): 940-958.

Hazen, R.A., Vasey, M.W., & Schmidt, N.B. (2009). Attentional retraining: A randomized
clinical trial for pathological worry. Journal of Psychiatric Research, 43: 627-633. doi:
10.1016/j.jpsychires.2008.07.004

Henry, J.D., Ruffman, T., McDonald, S., Peek-O’Leary, M., Phillips, L.H., Brodaty, H., &
.Rendall, P.G. (2008). Recognition of disgust is selectively preserved in Alzheimer’s
disease. Neuropsychologia, 46, 1363-1370. doi: 10.1016/j.neuropsychologia.2007.12.012

Hogan, M.J. (2003). Divided attention in older but not younger adults is impaired by anxiety.
Experimental Aging Research, 29, 111-136. doi: 10.1080/03610730303712

Isaacowitz, D.M., Lockenhoff, C.E., Lane, R.D., Wright, R., Sechrest, L., Riedel, R., & Costa,
P.T. (2007). Age differences in recognition of emotions in lexical stimuli and facial
expressions. Psychology and Aging, 22(1), 147-159. doi: 10.1037/0882-7974.22.1.147

Isaacowitz, D.M., & Stanley, J.T. (2011). Bringing an ecological perspective to the study of
aging recognition of emotional facial expressions: Past, present, and future methods.
Journal of Nonverbal Behavior, 35, 261-278. doi: 10.1007/s10919-011-0113-6

Isaacowitz, D.M., Wadlinger, H.A., Goren, D., & Wilson, H.R. (2006). Selective preference in
visual fixation away from negative images in old age? An eye-tracking study. Psychology
and Aging, 21(1), 40-48. doi: 10.1037/0882-7974.21.1.40

Isaacowitz, D.M., Toner, K., & Neupert, S.D. (2009). Use of gaze for real-time mood regulation:
effects of age and attentional functioning. Psychology and Aging, 24(4), 989-994. doi:
10.1037/a0017706

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 82
Isaacowitz, D.M. (2012). Mood regulation in real time: Age differences in the role of looking.
Current Directions in Psychological Science, 21(4), 237-242. doi:
10.1177/0963721412448651

Isaacowitz, D. M., & Choi, Y. (2011). The malleability of age-related positive gaze preferences:
Training to change gaze and mood. Emotion, 11(1): 90-100. doi: 10.1037/a0021551

Jeffries, L.N., Smilek, D., Eich, E., & Enns, J.T. (2008). Emotional valence and arousal interact
in attentional control. Psychological Science, 19(3), 290-295. doi: 10.1111/j.1467-
9280.2008.02082.x

Jennings, J.R., Brock, K., & Nebes, R. (1989). Aging but not arousal influences the effect of
environmental noise on the span of attention. Experimental Aging Research, 15(1-2), 61-
71. doi: 10.1080/03610738908259759

Joormann, J., & Gotlib, I.H. (2006). Is this happiness I see? Biases in the identification of
emotional facial expressions in depression and social phobia. Journal of Abnormal
Psychology, 115(4), 705-714. doi: 10.1037/0021-843X.115.4.705

Keogh, E., & French, C.C. (1999). The effect of trait anxiety and mood manipulation on the
breadth of attention. European Journal of Personality, 13, 209-223. doi:
10.1002/(SICI)1099-0984(199905/06)13:3<209::AID-PER346>3.0.CO;2-N

Kellough, J.L., & Knight, B.G. (2012). Positivity effects in older adults’ perception of emotion:
The role of future time perspective. Journals of Gerontology, Series B: Psychological
Sciences and Social Sciences, 67(2), 150– 158.

Kennedy, Q., Mather, M., & Carstensen, L.L. (2004). The role of motivation in the age-related
positivity effect in autobiographical memory. Psychological Science, 15(3), 208-214. doi:
10.1111/j.0956-7976.2004.01503011.x

Knight, M., Seymour, T.L., Grant, J.T., Baker, C., Nesmith, K., & Mather, M. (2007). Aging and
goal-directed emotional attention: distraction reverses emotional biases. Emotion, 7(4),
705-714. doi: 10.1037/1528-3542.7.4.705

Kohn, P.M., Kantor, L., DeCicco, T.L., & Beck, A.T. (2008). The Beck Anxiety Inventory-Trait
(BAIT): a measure of dispositional anxiety not contaminated by dispositional depression.
Journal of Personality Assessment, 90(5), 499-506. doi: 10.1080/00223890802248844

Kramer, A.F., Larish, J.F., & Strayer, D.L. (1995). Training for attentional control in dual task
settings: A comparison of young and old adults. Journal of Experimental Psychology:
Applied, 1(1), 50-76. doi: 10.1037/1076-898X.1.1.50

Lam, R.W., Michalaak, E.E., & Swinson, R.P. (2005). Assessment scales in depression and
anxiety. Boca Raton, FL: Taylor & Francis.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 83
Larcom, M.J., & Isaacowitz, D.M. (2009). Rapid emotion regulation after mood induction: Age
and individual differences. Journal of Geronotology: Psychological Sciences, 64B(6),
733-741. doi: 10.1093/geronb/gbp077

Leber, S., Heidenreich, T., Stangier, U., & Hofmann, S.G. (2009). Processing of facial affect
under social threat in socially anxious adults: mood matters. Depression and Anxiety, 26,
196-206. doi: 10.1002/da.20525

Lee, L.O., & Knight, B.G. (2009). Attentional bias for threat in older adults: moderations of the
positivity bias by trait anxiety and stimulus modality. Psychology and Aging, 24(3), 741-
747. doi: 10.1037/a0016409

Lee, T.M.C., Ng, E.H.H., Tang, S.W., & Chan, C.C.H. (2008). Effects of sad mood on facial
emotion recognition in Chinese people. Psychiatry Research, 159, 37-43. doi:
10.1016/j.psychres.2007.04.022

Levenson, R.W., Carstensen, L.L., Friesen, W.V., & Ekman, P. (1991). Emotion, physiology,
and expression in old age. Psychology and Aging, 6(1), 28-35. doi: 10.1037/0882-
7974.6.1.28

Levine, L.J., & Pizarro, D.A. (2004). Emotion and memory research: A grumpy overview. Social
Cognition, 22(5), 530-544. doi: 10.1521/soco.22.5.530.50767

Lezak, MD, Howieson, D.B., & Loring, D.W. (2004). Neuropsychological Assessment (4th ed.).
New York: Oxford University Press.

Liebman, S.E., & Allen, G.J. (1995). Anxiety sensitivity, state anxiety and perceptions of facial
emotions. Journal of Anxiety Disorders, 9(4), 257-267. doi: 10.1016/0887-
6185(95)00007-B

Lockenhoff, C.E., & Carstensen, L.L. (2007). Aging, emotion, and health-related decision
strategies: Motivational manipulations can reduce age differences. Psychology and
Aging, 22(1), 134-146. doi: 10.1037/0882-7974.22.1.134

Mather, M., Canli, T., English. T., Whitfield, S., Wais, P., Ochsner, K., …Carstensen, L.L.
(2004). Amygdala responses to emotionally valenced stimuli in older and younger adults.
Psychological Science, 15(4), 259-263.doi: 10.1111/j.0956-7976.2004.00662.x

Mather, M.. & Carstensen, L.L. (2003). Aging and attentional biases for emotional faces.
Psychological Science, 14(5), 409-415. doi: 10.1111/1467-9280.01455

Mather, M., & Carstensen, L.L. (2005). Aging and motivated cognition: the positivity effect in
attention and memory. Trends in Cognitive Sciences, 9(10), 496-502. doi:
10.1016/j.tics.2005.08.005

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 84
Mather, M., & Knight, M. (2005). Goal-directed memory: the role of cognitive control in older
adults’ emotional memory. Psychology and Aging, 20(4), 554-570. doi: 10.1037/0882-
7974.20.4.554

Mather, M., & Knight, M. (2006). Angry faces get noticed quickly: threat detection is not
impaired among older adults. Journal of Geronotology: Psychological Sciences, 61B(1),
P54-P57. Retrieved from http://psychsocgerontology.oxfordjournals.org/

Mathews, A., & MacLeod, C. (2002). Induced processing biases have causal effects on anxiety.
Cognition and Emotion, 16(3), 331-354. doi: 10.1080/02699930143000518

Mathews, A., & MacLeod, C. (1994). Cognitive approaches to emotion and emotional disorders.
Annual Review of Psychology, 45, 25-50. doi: 10.1146/annurev.ps.45.020194.000325

Mathews, A., & Mackintosh, B. (1998). A cognitive model of selective processing in anxiety.
Cognitive Therapy and Research, 22(6), 539-560. doi: 10.1023/A:1018738019346

Merens, W., Willem Van der Does, A.J., & Spinhoven, P. (2007). The effects of serotonin
manipulations on emotional information processing and mood. Journal of Affective
Disorders, 103, 43-62. doi: 10.1016/j.jad.2007.01.032

Mill, A., Allik, J., Realo, A., & Valk, R. (2009). Age-related difference in emotion recognition
ability: a cross-sectional study. Emotion, 9(5), 619-630. doi: 10.1037/a0016562

Mitchell, A.J., Baker-Glenn, E.A., Granger, L., & Symonds, P. (2010). Can the distress
thermometer be improved by additional mood domains? Part I. Initial validation of the
Emotion Thermometers tool. Psycho-Oncology, 19, 125-133. doi: 10.1002/pon.1523

Mogg, K., & Bradley, B.P. (1998). A cognitive-motivational analysis of anxiety. Behaviour
Research and Therapy, 36, 809-848. doi: 10.1016/S0005-7967(98)00063-1

Mogg, K., Garner, M., & Bradley, B.P. (2007). Anxiety and orienting of gaze to angry and
fearful faces. Biological Psychology, 76, 163-169. doi: 10.1016/j.biopsycho.2007.07.005

Mohlman, J., Carmin, C.N., & Price, R.B. (2007). Jumping to interpretations: social anxiety
disorder and the identification of emotional facial expression. Behaviour Research and
Therapy, 45, 591-599. doi: 10.1016/j.brat.2006.03.007

Mullins, D.T., & Duke, M.P. (2004). Effects of social anxiety on nonverbal accuracy and
response time I: Facial expressions. Journal of Nonverbal Behavior, 28(1), 3-33. doi:
10.1023/B:JONB.0000017865.24656.98

Murphy, N.A., & Isaacowitz, D.M. (2010). Age effects and gaze patterns in recognising
emotional expressions: an in-depth look at gaze measures and covariates. Cognition and
Emotion, 24(3), 436-452. doi: 10.1080/02699930802664623

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 85
Murphy, S.E., Norbury, R., O’Sullivan, U., Cowen, P.J., & Harmer C.J. (2009). Effect of a
single dose of citalopram on amygdale response to emotional faces. British Journal of
Psychiatry, 194, 535-540. doi: 10.1192/bjp.bp.108.056093

Noh, S.R., Lohani, M., & Isaacowitz, D.M. (2011). Deliberate real-time mood regulation in
adulthood: The importance of age, fixation, and attentional functioning. Cognition and
Emotion, 25(6), 998-1013. doi: 10.1080/02699931.2010.541668

Ohman, A., Lundqvist, D., & Esteves, F. (2001). The face in the crowd revisited: a threat
advantage with schematic stimuli. Journal of Personality and Social Psychology, 80(3),
381-396.

Ochsner, K.N., & Gross, J.J. (2005). The cognitive control of emotion. TRENDS in Cognitive
Science, 9(5), 242-249. doi: 10.1016/j.tics.2005.03.010

Phan, K.L., Wager, T., Taylor, S.F., & Liberzon, I. (2002). Functional neuroanatomy of emotion:
A meta-analysis of emotion activation studies in PET and fMRI. NeuroImage, 16, 331-
348. doi: 10.1006/nimg.2002.1087

Philippot, P., & Douilliez, C. (2005). Social phobics do not misinterpret facial expression of
emotion. Behaviour Research and Therapy, 43, 639-652. doi: 10.1016/j.brat.2004.05.005

Phillips, L. H., MacLean, R. D. J., & Allen, R. (2002). Age and the understanding of emotions:
neuropsychologial and sociocognitive perspectives. Journals of Gerontology, Series B:
Psychological Sciences and Social Sciences, 57, P526– P530. doi:
10.1093/geronb/57.6.P526

Phillips, L.H., & Allen, R. (2004). Adult aging and the perceived intensity of emotions in faces
and stories. Aging Clinical Experimental Research, 16(3), 190-199. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/15462461

Poon, C.Y.M., & Knight, B.G. (2009). Influence of sad mood and old age schema on older
adults’ attention to physical symptoms. Journal of Geronotology: Psychological
Sciences, 64B(1), 41-44. doi: 10.1093/geronb/gbn025

Riediger, M., Voelkle, M.C., Ebner, N.C., & Lindenberger, U. (2011). Beyond “happy, angry, or
sad?”: Age-of-poser and age-of-rater effects on multi-dimensional emotion perception.
Cognition & Emotion, 25(6), 968-982. doi:10.1080/02699931.2010.540812

Reinholdt-Dunne, M.L., Mogg, K., & Bradley, B.P. (2009). Effects of anxiety and attention
control on processing pictorial and linguistic information. Behaviour Research and
Therapy, 47, 410-417. doi: 10.1016/j.brat.2009.01.012

Richards, A., French, C.C., Calder, A.J., Webb, B., Fox, R. & Young, A.W. (2002). Anxiety-
related bias in the classification of emotionally ambiguous facial expressions. Emotion,
2(3), 273-287. doi: 10.1037/1528-3542.2.3.273
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 86

Ridout, N., Noreen, A., & Johal, J. (2009). Memory for emotional faces in naturally occurring
dysphoria and induced sadness. Behavior Research and Therapy, 47, 851-860. doi:
10.1016/j.brat.2009.06.013
Roberts, B.W., & DelVecchio, W.F. (2000). The rank-order consistency of personality traits
from childhood to old age: A quantitative review of longitudinal studies. Psychological
Bulletin, 126(1), 3-25. doi: 10.1037//0033-2909.126.1.3.

Rosler, A., Ulrich, C., Billino, J., Sterzer, P., Weidauer, S., Bernhardt, T., …, Kleinschmidt, A.
(2005). Effects of arousing emotional scenes on the distribution of visuospatial attention:
Changes with aging and early subcortical vascular dementia. Journal of Neurological
Sciences, 229-230, 109-116. doi: 10.1016/j.jns.2004.11.007

Rueda, M.R., Checa, P., & Rothbart, M.K. (2010). Contributions of attentional control to
socioemotional and academic development. Early Education and Development, 21(5),
744-764. doi: 10.1080/10409289.2010.510055

Ruffman, T., Henry, J.D., Livingstone, V., & Phillips, L.H. (2008). A meta-analytic review of
emotion recognition and aging: implications for neuropsychological models of aging.
Neuroscience and Biobehavioral Reviews, 32, 863-881. doi:
10.1016/j.neubiorev.2008.01.001

Russell, J.A., & Feldman Barrett, L. (1999). Core affect, prototypical emotional episodes, and
other things called emotion: Dissecting the elephant. Journal of Personality and Social
Psychology, 76(5), 805-819. doi: 10.1037/0022-3514.76.5.805

Rusting, C.L. (1999). Interactive effects of personality and mood on emotion-congruent memory
and judgment. Journal of Personality and Social Psychology, 77(5), 1073-1086. doi:
10.1037/0022-3514.77.5.1073

Salovey, P., & Binrnbaum, D. (1989). Influence of mood on health-relevant cognitions. Journal
of Personality and Social Psychology, 57(3), 539-551. doi: 10.1037/0022-3514.57.3.539

Salthouse, T.A., Toth, J. Daniels, K., Parks, C., Pak, R., Wolbrette, M., & Hocking, K.J. (2000).
Effects of aging on efficiency of task switching in a variant of the Trail Making Test.
Neuropsychology, 14(1), 102-111. DOI: 10.1037//0894-4105.14.1.102

Salthouse, T.A. (2011). What cognitive abilities are involved in trail-making performance?
Intelligence, 39(4), 222-232. doi: 10.1016/j.intell.2011.03.001

Salthouse, T.A., Atkinson, T.M., & Berish, D.E. (2003). Executive functioning as a potential
mediator of age-related cognitive decline in normal adults. Journal of Experimental
Psychology: General, 132(4), 566-594. doi: 10.1037/0096-3445.132.4.566

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 87
Samanez-Larkin, G.R., Robertson, E.R., Mikels, J.A., Carstensen, L.L., & Gotlib, I.H. (2009).
Selective attention to emotion in the aging brain. Psychology and Aging, 24(3), 519-529.
doi: 10.1037/a0016952

Sasson, N.J., Pinkham, A.E., Richard, J., Hughett, P., Gur, R.E., & Gur R.C. (2010). Controlling
for response biases clarifies sex and age differences in facial affect recognition. Journal
of Nonverbal Behavior, 34, 207-221. doi: 10.1007/s10919-010-0092-z

Schmidt, N.B., Richey, J.A., Buckner, J.D., Timpano, K.R. (2009). Attention training for
Generalized Social Anxiety Disorder. Journal of Abnormal Psychology, 118(1): 5-14.
doi: 10.1037/a0013643

Shiota, M.N., & Levenson, R.W. (2009). Effects of aging on experimentally instructed detached
reappraisal, positive reappraisal, and emotional behavior suppression. Psychology and
Aging, 24(4): 890-900. doi: 10.1037/a0017896

Spielberger, C.D. (1985). Anxiety, cognition and affect: A state-trait perspective. Hillsdale, NJ:
Lawrence Erlbaum Associates.

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for
the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists Press.

Sprengelmeyer, R., Young, A.W., Pundt, I., Sprengelmeyer, A., Calder, A.J., Berrios, G.,
…Przuntek, H. (1997). Disgust implicate in obsessive-compulsive disorder. Proceedings
of the Royal Society B, 265, 1767-1773. Retrieved from
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1688750/

Stanley, J.T., & Blanchard-Fields, F. (2008). Challenges older adults face in detecting deceit:
The role of emotion recognition. Psychology and Aging, 23(1), 24-32. doi: 10.1037/0882-
7974.23.1.24

Stein, M.B., Simmons, A.N., Feinsten, J.S., & Paulus, M.P. (2007). Increased amygdala and
insula activity during emotion processing in anxiety-prone subjects. American Journal of
Psychiatry, 164(2), 318-327. doi: 10.1176/appi.ajp.164.2.318

Sullivan, S., & Ruffman, T. (2004). Emotion recognition deficits in the elderly. International
Journal of Neuroscience, 114, 403–432. doi: 10.1080/00207450490270901

Surcinelli, P., Codispoti, M., Montebarocci, O., Rossi, N., & Baldaro, B. (2006). Facial emotion
recognition in trait anxiety. Journal of Anxiety Disorders, 20, 110-117. doi:
10.1016/j.janxdis.2004.11.010

Suzuki, A., Hoshino, T., Shigemasu, K., & Kawamura, M. (2007). Decline or improvement?
Age-related differences in facial expression recognition. Biological Psychology, 74, 75-
84. doi: 10.1016/j.biopsycho.2006.07.003

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 88
Wenzel, A., & Lystad, C. (2005). Interpretation biases in angry and anxious individuals.
Behaviour Research Therapy, 43, 1045-1054. doi: 10.1016/j.brat.2004.02.009

Wetherell, J.L, & Arean, P.A. (1997). Psychometric evaluation of the Beck Anxiety Inventory
with older medical patients. Psychological Assessment, 9(2), 136-144. doi: 7/1040-
3590.9.2.136

Wetherell, J.L., Petkus, A.J., McChesney, K., Stein, M.B., Judd, P.H., Rockwell, E., Sewell,
D.D., & Patterson, T.L. (2009). Older adults are less accurate than younger adults at
identifying symptoms of anxiety and depression. Journal of Nervous and Mental
Disorders, 197, 623-626. doi: 10.1097/NMD.0b013e3181b0c081

Williams, J. M. G., Watts, F. N., MacLeod, C. & Mathews, A. (1997). Cognitive psychology and
emotional disorders (2nd ed.). Chichester: Wiley.

Williams, L.M., Mathersul, D., Palmer, D.M., Gur, R.C., Gur, R.E., & Gordon, E. (2009).
Explicit identification and implicit recognition of facial emotions: I. Age effects in males
and females across 10 decades. Journal of Clinical and Experimental Neuropsychology,
31(3), 257-277. doi: 10.1080/13803390802255635

Wolitzky-Taylor, K.B., Castriotta, N., Lenze, E.J., Stanley, M.A., & Craske, M.G. (2010).
Anxiety disorders in older adults: A comprehensive review. Depression and Anxiety, 27,
190-211. doi: 10.1002/da.20653

Wong, B., Cronin-Golomb, A., & Neargarder, S. (2005). Patterns of visual scanning as
predictors of emotion identification in normal aging. Neuropsychology, 19, 739–749. doi:
10.1037/0894-4105.19.6.739

Wright, J., & Mischel, W. (1982). Influence of affect on cognitive social learning person
variables. Journal of Personality and Social Psychology, 43(5), 901-914. doi:
10.1037/0022-3514.43.5.901

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 89
Endnote
1
Post-hoc analyses of accuracy and intensity ratings for neutral facial expressions were
conducted using the same tests as described for emotional facial expressions (Hypotheses 2
through 5). Among these analyses, only one test reached significance, indicating that age and
attentional control ability group had interactive effects on accuracy for neutral facial expressions
(F(1, 94) = 4.561, p = 0.035, η
p
2
= .046), with younger adults high in attentional control (M =
.737, SD = 0.184) appearing to identify neutral faces more accurately than low attentional control
peers (M = .688, SD = 0.172), and older adults low in attentional control (M = .777, SD = 0.184)
appearing to identify neutral faces more accurately than high peers (M = .674, SD = 0.176). This
pattern is notable inasmuch as it mimics our findings for differences by age and attentional
control group in the accurate identification of fear faces (see Figure 6). Error analyses revealed
that older adults with high attentional control were more likely to label neutral faces as happy
than low attentional control older adults (p = 0.031), suggestive of a positivity bias in those with
high control ability, whereas in younger adults, there was a trend for those with low attentional
control to label neutral faces as sad, though this did not reach significance (p = 0.082).

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 90

Table 1. Baseline Age, Health, and Mood (Trait and State) Characteristics Across Age and Induction Groups, Whole Sample

BAIT = Beck Anxiety Inventory-Trait; VAS Anxiety (1) = Visual Analog Scale Anxiety, Baseline, STAI-S = State-Trait Anxiety
Inventory-Y2, State Short Form

* Older and younger adult groups differ at p < 0.05

Age Group

Younger
Older
Induction Calm Anxious

Calm

Anxious

Demographic N M SD N M SD N M SD N M SD
Age (years) * 37 19.59 1.092 36 19.81 1.191 31 71.29 8.351 33 71.85 5.310
Self-rated health 36 3.94 0.674 36 4.14 0.798 31 4.00 0.775 32 4.09 0.928
BAIT * 37 8.08 5.790 36 8.94 8.763 30 5.07 4.226 33 5.55 5.466
VAS Anxiety (1) * 37 3.89 2.703 36 3.97 2.970 31 1.47 1.532 33 1.79 2.342
STAI-S (1) * 37 18.65 5.884 36 17.64 6.072 31 12.55 3.889 33 13.00 3.437
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 91
Table 2. Distribution of Participants Across Age and Induction Groups by Gender, Ethnicity, Education, Occupation, and Income
Category, Whole Sample

Age Group
Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Gender
Male 7 9.6 3 4.1 13.7 8 12.5 12 18.8 31.2
Female 30 41.1 33 45.2 86.3 23 35.9 21 32.8 68.8
Ethnicity

Afr. Amer./Black 4 5.5 3 4.1 9.6 5 7.8 3 4.7 12.5
Asian, Pac. Island. 16 21.9 14 19.2 41.1 0 0 2 3.1 3.1
Caucasian/White 11 15.1 12 16.4 31.5 25 39.1 25 39.1 78.1
Latino/Hispanic 3 4.1 6 8.2 12.3 0 0 1 1.6 1.6
Other 3 4.1 1 1.4 5.5 1 1.6 2 3.1 4.7
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 92
Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Occupation
Student 36 49.3 36 49.3 98.6 0 0 0 0 0
Retired 0 0 0 0 0 24 37.5 24 37.5 75.0
Other 0 0 0 0 0 7 10.9 8 12.5 23.4
Missing 1 1.4 0 0 1.4 0 0 1 1.6 1.6
Education
<12 years 0 0 0 0 0 0 0 0 0 0
12 years 0 0 0 0 0 2 3.1 1 1.6 4.7
12-15 years 37 50.7 35 47.9 100 7 10.9 7 10.9 21.8
16 years 0 0 0 0 0 5 7.8 6 9.4 17.2
17-19 years 0 0 0 0 0 14 21.9 10 15.6 37.5
20+ years 0 0 0 0 0 3 4.7 9 14.1 18.8
Missing 0 0 1 1.4 1.4 0 0 0 0 0
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 93

Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Income
<$10,000 30 41.1 29 39.7 80.8 0 0 1 1.6 1.6
$10,000-$49,999 2 2.7 3 4.1 6.8 16 25.0 10 15.6 40.6
$50,000-$99,999 0 0 1 1.4 1.4 11 17.2 6 9.4 26.6
$100,000-
$199,999
0 0 1 1.4 1.4 3 4.6 12 18.8 23.4
$200,000-
$499,999
1 1.4 0 0 1.4 0 0 2 3.1 3.1
>$500,000 1 1.4 0 0 1.4 0 0 0 0 0
Missing 3 4.1 2 2.7 6.8 1 1.6 2 3.1 4.7
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 94

Table 3. Baseline Age, Health, and Mood (Trait and State) Characteristics Across Age and Induction Groups, Among Responders to
Induction

BAIT = Beck Anxiety Inventory - Trait; VAS Anxiety (1) = Visual Analog Scale Anxiety at baseline; STAI-S = State-Trait Anxiety
Inventory-Y2, State - Short Form at baseline

* Older and younger adult groups differ at p < 0.05

^ Age group by induction group interaction at p < 0.05

Age Group

Younger
Older
Induction Calm Anxious

Calm

Anxious

Demographic N M SD N M SD N M SD N M SD
Age (years) * 32 19.59 1.073 25 19.72 0.980 27 70.30 7.065 20 71.60 6.227
Self-rated health 31 3.97 0.706 25 4.28 0.678 27 4.07 0.675 19 4.00 0.882
BAIT * 32 8.00 5.547 25 8.16 7.324 26 4.96 4.313 20 5.61 5.097
VAS Anxiety (1) * 32 4.06 2.523 25 3.24 2.678 27 1.44 1.540 20 1.28 1.650
STAI-S (1) *^ 32 19.44 5.913 25 16.16 4.365 27 12.82 4.095 20 13.30 3.729
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 95
Table 4. Distribution of Participants Across Age and Induction Groups by Gender, Ethnicity, Education, Occupation, and Income
Category, Among Responders to Induction

Age Group
Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Gender
Male 7 12.3 2 3.5 15.8 7 14.9 9 19.1 34.0
Female 25 43.9 23 40.4 84.2 20 42.6 11 23.4 66.0
Ethnicity

Afr. Amer./Black 1 1.8 1 1.8 3.5 5 10.6 3 6.4 17.0
Asian, Pac. Island. 14 24.6 9 15.8 40.4 0 0 1 2.1 2.1
Caucasian/White 11 19.3 9 15.8 35.1 21 44.8 15 31.9 76.7
Latino/Hispanic 3 5.3 5 8.8 14.0 0 0 1 2.1 2.1
Other 3 5.3 1 1.8 7.0 1 2.1 0 0 2.1
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 96
Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Occupation
Student 31 54.4 25 43.9 98.2 0 0 0 0 0
Retired 0 0 0 0 0 20 42.6 15 31.9 74.5
Other 0 0 0 0 0 7 14.9 4 8.5 23.4
Missing 1 1.8 0 0 1.8 0 0 1 2.1 2.1
Education
<12 years 0 0 0 0 0 0 0 0 0 0
12 years 0 0 0 0 0 1 2.1 1 2.1 4.2
12-15 years 32 56.1 25 43.9 100 6 12.8 5 10.6 23.4
16 years 0 0 0 0 0 5 10.6 1 2.1 12.8
17-19 years 0 0 0 0 0 13 27.7 6 12.8 40.4
20+ years 0 0 0 0 0 2 4.3 7 14.9 19.2
Missing 0 0 0 0 0 0 0 0 0 0

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 97
Younger Older
Induction
Calm Anxious Calm Anxious
Demographic N % N % Total %
within
age
group
N % N % Total %
within
age
group
Income
<$10,000 25 43.8 19 33.3 77.1 0 0 1 2.1 2.1
$10,000-$49,999 2 3.5 3 5.2 8.7 15 31.9 5 10.6 42.6
$50,000-$99,999 0 0 1 1.8 1.8 8 17.0 4 8.5 25.5
$100,000-
$199,999
0 0 1 1.8 1.8 3 6.4 7 14.9 21.3
$200,000-
$499,999
1 1.8 0 0 1.8 0 0 1 2.1 2.1
>$500,000 1 1.8 0 0 1.8 0 0 0 0 0
Missing 3 5.3 1 1.8 7.0 1 2.1 2 4.3 6.4

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 98
Table 5. Mean Accuracy (Percent Correct) for Identification of Facial Expressions
of Emotion by Age Group and Induction Group

Mean Accuracy (%)
Younger Adults Older Adults†

Emotion
Calm
N=32
Anxious
N=25
Calm
N=27
Anxious
N=20
Anger * 76.6 72.0 59.0 67.1
Fear * 89.5 94.0 77.3 80.0
Happy 98.0 97.0 95.4 94.4
Sad *# 92.6 84.5 80.6 68.1

* Significant age group main effect at p < 0.0125
# Significant induction main effect at p < 0.0125
† Older adult sample size for “anger” emotion category: calm - N = 25; anxious - N = 19

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 99
Table 6. Mean Intensity Ratings for Correctly-Identified Facial Expressions of
Emotion by Age Group and Induction Group

Mean Intensity (1-9)
Younger Adults Older Adults†
Emotion
Calm
N=32
Anxious
N=25
Calm
N=27
Anxious
N=20
Anger * 6.04 6.57 6.74 6.96
Fear # 5.73 6.30 5.59 6.10
Happy * 6.45 6.49 6.87 7.10
Sad * 5.75 5.94 6.44 6.61

* Significant age group main effect at p < 0.0125
# Significant induction main effect at p < 0.0125
† Older adult sample size for “anger” emotion category: calm - N = 25; anxious - N = 19

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 100

(B) Calm Induction
Younger Adults Older Adults††

Emotion
Low Trait
N=16
High Trait
N=13
Low Trait
N=12
High Trait
N=15
Anger * 82.0 72.1 63.8 55.8
Fear * 91.4 88.5 80.2 75.0
Happy 98.4 98.1 97.9 93.3
Sad * 91.4 94.2 79.2 81.7

* Significant age group main effect at p < 0.05
^ Significant age group by trait anxiety group (BAIT) interaction at p < 0.05
† Older adult sample size for “anger” emotion category: low trait- N= 9; high trait- N= 9
†† Older adult sample size for “anger” emotion category: low trait- N= 10; high trait- N=15

Table 7. Mean Accuracy (Percent Correct) for Identification of Facial Expressions
of Emotion by Age Group and Trait Anxiety Group Within (A) the Anxious and (B)
the Calm Induction Groups

(A) Anxious Induction
Younger Adults Older Adults†

Emotion
Low Trait
N=12
High Trait
N=13
Low Trait
N=10
High Trait
N=9
Anger 76.0 68.3 68.1 65.3
Fear *^ 97.9 90.4 72.5 88.9
Happy 95.8 98.1 92.5 95.8
Sad * 86.5 82.7 68.8 69.4
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 101

(B) Calm Induction
Younger Adults Older Adults††

Emotion
Low Trait
N=16
High Trait
N=13
Low Trait
N=12
High Trait
N=15
Anger * 6.08 5.93 6.84 6.67
Fear 5.68 5.65 5.78 5.43
Happy 6.54 6.35 7.04 6.72
Sad * 5.72 5.73 6.50 6.39
* Significant age group main effect at p < 0.05
^ Significant age group* trait anxiety group (BAIT) interaction at p < 0.05
† Older adult sample size for “anger” emotion category: low trait- N= 9; high trait- N= 9
†† Older adult sample size for “anger” emotion category: low trait- N= 10; high trait- N=15

Table 8. Mean Intensity Ratings for Correctly-Identified Facial Expressions of
Emotion by Age Group and Trait Anxiety Group Within (A) the Anxious and (B)
Calm Induction Groups

(A) Anxious Induction
Younger Adults Older Adults†

Emotion
Low Trait
N=12
High Trait
N=13
Low Trait
N=10
High Trait
N=9
Anger ^ 6.94 6.23 6.42 7.38
Fear ^ 6.78 5.86 5.68 6.37
Happy * 6.58 6.41 6.87 7.17
Sad * 6.08 5.81 6.23 6.83
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 102
Table 9. Mean Accuracy (Percent Correct) for Identification of Facial Expressions of
Emotion by Age Group and Attentional Control (Attn) Group Within the (A) Anxious and
(B) Calm Induction Groups

(A) Anxious Induction Group
Younger Adults Older Adults†
Emotion
Low Attn
N=9
High Attn
N=16
Low Attn
N=10
High Attn
N=9
Anger 68.1 74.2 65.0 71.9
Fear *#^ 93.1 94.5 91.3 69.4
Happy 94.4 98.4 95.0 94.4
Sad * 84.7 84.4 73.8 62.5
(B) Calm Induction Group

Younger Adults Older Adults††
Emotion
Low Attn
N=19
High Attn
N=12
Low Attn
N=12
High Attn
N=14
Anger * 77.0 75.0 58.3 59.6
Fear * 88.8 89.6 80.8 74.1
Happy 98.7 96.9 96.2 94.6
Sad * 92.8 91.7 76.9 83.9
* Significant age group main effect at p < 0.05
# Significant attentional control group main effect at p < 0.05
^ Significant age group by attentional control group interaction at p < 0.05
† Older adult sample size for “anger” emotion category: low attn- N= 10; high attn- N=8
†† Older adult sample size for “anger” emotion category: low attn- N= 11; high attn- N= 13

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 103
Table 10. Mean Intensity Ratings for Correctly-Identified Facial Expressions of
Emotion by Age Group and Attentional Control (Attn) Group within the (A) Anxious
and (B) Calm Induction Groups

(A) Anxious Induction Group
Younger Adults Older Adults†
Emotion
Low Attn
N=9
High Attn
N=16
Low Attn
N=10
High Attn
N=9
Anger # 6.89 6.39 7.63 6.29
Fear # 6.91 5.96 6.68 5.70
Happy * 6.61 6.42 7.50 6.69
Sad * 6.11 5.84 6.95 6.40

(B) Calm Induction Group
Younger Adults Older Adults††
Emotion
Low Attn
N=19
High Attn
N=12
Low Attn
N=12
High Attn
N=14
Anger * 6.17 5.87 6.56 6.90
Fear 5.75 5.71 5.62 5.56
Happy 6.45 6.44 6.96 6.78
Sad * 5.90 5.59 6.12 6.73
* Significant age group main effect at p < 0.05
# Significant executive control group main effect at p < 0.05
† Older adult sample size for “anger” emotion category: low attn- N= 10; high attn- N= 8
†† Older adult sample size for “anger” emotion category: low attn- N= 11; high attn- N= 13

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 104
(A) Visual Analog Scale (VAS) scores

(B) State-Trait Anxiety Inventory-Y2, State Short Form (STAI-S) scores

Figure 1. Change in (A) Visual Analog Scale (VAS) scores and (B) State-Trait Anxiety
Inventory-Y2 State Short Form (STAI-S) scores over time by induction group for whole sample.
(Time 1: baseline; Time 2: post-mood induction)
INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 105
(A) Younger adults (B) Older adults

Figure 2. Change in Visual Analog Scale (VAS) scores over time by induction group in (A) younger and (B) older adults. (Time 1:
baseline; Time 2: post-mood induction, Time 3: Following facial emotion identification task, Time 4: post-induction 2)

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 106
(A) Younger adults (B) Older adults

Figure 3. Change in State-Trait Anxiety Inventory-Y2, State Short Form (STAI-S) scores scores over time by induction group in (A)
younger and (B) older adults. (Time 1: baseline; Time 2: post-mood induction, Time 3: Following facial emotion identification task,
Time 4: post-induction 2)

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 107
(A) Anxious Induction (B) Calm Induction

Figure 4. Mean accuracy scores (proportion correct) for fear faces in the (A) anxious and (B) calm induction groups by trait anxiety
group. Error bars denote +/- 1 standard error.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 108
(A) Anxious Induction (B) Calm Induction

Figure 5. Mean intensity ratings for correctly-identified anger faces in the (A) anxious and (B) calm induction groups by trait anxiety
group. Error bars denote +/- 1 standard error.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 109
(A) Anxious Induction (B) Calm Induction

Figure 6. Mean intensity ratings for correctly-identified fear faces in the (A) anxious and (B) calm induction groups by trait anxiety
group. Error bars denote +/- 1 standard error.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 110

Figure 7. Mean accuracy scores (proportion correct) for fear faces by age group and attentional
control group. Error bars denote +/- 1 standard error.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 111

Figure 8. Mean intensity ratings for anger faces by induction group and attentional control
group. Error bars denote +/- 1 standard error.

INFLUENCE OF AGE AND ANXIETY ON EMOTION PROCESSING 112

Figure 9. Mean intensity ratings for fear faces by age group and attentional control group. Error
bars denote +/- 1 standard error.

Abstract (if available)

Abstract Prior laboratory research suggests that older adults are less accurate than younger groups in identifying facial expressions of certain negative emotions. However, the ability to recognize positive expressions appears to be preserved with age. The reasons for this variability are as yet undetermined. Anxiety, via concurrent attentional and interpretive biases, may play an important role in age differences in the recognition of facial expression by valence, in light of research that suggests anxiety increases accuracy and intensity appraisal of emotional faces, particularly those that are threat-relevant (i.e. anger and fear). The current project used an experimental mood induction paradigm in order to examine the effects of anxious mood, trait anxiety, and attentional control on accuracy and subjective intensity ratings in a facial expression recognition task across younger and older individuals. Results indicated that older adults with concurrent state and trait anxiety demonstrated mood-congruent facial processing, with higher accuracy and intensity ratings for certain threat-relevant faces

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Aging and emotion regulation in the judgment of facial emotion

PDF

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

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

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

PDF

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

PDF

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

PDF

Emotional reactivity and memory biases in older and younger adults

PDF

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

PDF

Emotional arousal amplifies the selectivity of visual selective attention

PDF

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

PDF

How age-related changes in the locus coeruleus affect selective attention and memory

PDF

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

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

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

PDF

Sources of stability and change in the trajectory of openness to experience across the lifespan

PDF

Dementia, co-morbid depressive symptoms, and mortality in older adults: a twin study

PDF

The link between the expression of negative emotion and future well-being: the role of disclosure modality in an East Asian American population

PDF

Untreated hyperglycemia associated with tau pathology and worse cognitive performance in older adults

PDF

Sensitive, specific, and generative face recognition in a newborn visual system

Asset Metadata

Creator Rastegar, Sarah (author)

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

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 11/28/2012

Defense Date 10/30/2012

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

Tag aging and emotion recognition,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 sarah.rastegar@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-122259

Unique identifier UC11291144

Identifier usctheses-c3-122259 (legacy record id)

Legacy Identifier etd-RastegarSa-1358.pdf

Dmrecord 122259

Document Type Thesis

Rights Rastegar, Sarah

Type texts

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

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

Repository Name University of Southern California Digital Library

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