Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Emotion, attention and cognitive aging: the effects of emotional arousal on subsequent visual processing
(USC Thesis Other)
Emotion, attention and cognitive aging: the effects of emotional arousal on subsequent visual processing
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Emotion, Attention and Cognitive Aging: The Effects of Emotional Arousal on Subsequent
Visual Processing
By
Matthew Ryan Sutherland
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)
August 2014
© Copyright 2014 Matthew Ryan Sutherland
i
Dedication
This dissertation is dedicated to my hero and matriarch, Laura Marie Sutherland. It’s
been a crazy road, but I made it. I will continue to make you proud.
ii
Acknowledgements
My sincerest and deepest appreciations go to my graduate advisor and mentor, Mara
Mather. Her constant professionalism, enthusiasm for science and relentless work ethic have left
an indelible mark on me that I can only hope to emulate throughout my career. Some debts can
never be repaid, but I will do everything I can to give back to my colleagues, future students and
to the field as a whole. As for the rest of my committee: Bosco Tjan, Michael Dawson, Laurent
Itti and Rand Wilcox; thank you for your support, astute criticisms and belief in my project. It
was an honor to defend my work to such an elite group of scientists. And of course many thanks
to the past and present members of the Emotion and Cognition Lab at the University of Southern
California. You are an exclusive bunch—cheers to the future!
iii
Table of Contents
List Of Figures ............................................................................................................................... vi
Abstract ...........................................................................................................................................1
Chapter 1 ..........................................................................................................................................2
1.1 Introduction ................................................................................................................................2
1.2 Emotion And Visual Processing ................................................................................................5
1.3 Emotion And Subsequent Visual Processing .............................................................................8
1.4 Arousal-Biased Competition (ABC) Theory ...........................................................................11
1.5 Emotion And Visual Processing In Older Adults ....................................................................14
1.6 Overview ..................................................................................................................................18
Chapter 2 ........................................................................................................................................23
2.1 Experiment 1 ............................................................................................................................23
2.2 Methods....................................................................................................................................23
2.2.1 Stimuli ...................................................................................................................................23
2.2.2 Participants ............................................................................................................................24
2.2.3 Procedure ..............................................................................................................................24
2.2.4 Data Analysis ........................................................................................................................25
2.3 Results And Discussion ...........................................................................................................26
Chapter 3 ........................................................................................................................................28
iv
3.1 Experiment 2 ............................................................................................................................28
3.2 Methods....................................................................................................................................28
3.2.1 Stimuli ...................................................................................................................................28
3.2.2 Participants ............................................................................................................................28
3.2.3 Procedure ..............................................................................................................................29
3.2.4 Data Analysis ........................................................................................................................29
3.3 Results And Discussion ...........................................................................................................29
Chapter 4 ........................................................................................................................................32
4.1 Experiment 3 ............................................................................................................................32
4.2 Methods....................................................................................................................................33
4.2.1 Stimuli ...................................................................................................................................33
4.2.2 Participants ............................................................................................................................33
4.2.3 Procedure ..............................................................................................................................33
4.2.4 Data Analysis ........................................................................................................................34
4.3 Results And Discussion ...........................................................................................................35
4.4 Results Summary .....................................................................................................................39
Chapter 5 ........................................................................................................................................41
5.1 General Discussion ..................................................................................................................41
5.2 Concluding Remarks ................................................................................................................55
v
References ......................................................................................................................................58
Appendix A: Tables .......................................................................................................................68
Appendix B: Figures ......................................................................................................................82
vi
List
of
Figures
Table 1: IADS Stimuli Used In Experiment 1 ...............................................................................68
Table 2: HLM Analysis Of Experiment 1 (Free Recall) ................................................................69
Table 3: HLM Analysis Of Experiment 1 (Incorrect Responses) ..................................................71
Table 4: IADS Stimuli Used In Experiment 2 ...............................................................................72
Table 5: HLM Analysis Of Older Adults And Young Adults Of Experiment 2 (Free Recall) .....73
Table 6: HLM Analysis Of Young Adults Of Experiment 2 (Free Recall) ...................................75
Table 7: HLM Analysis Of Older Adults And Young Adults Of Experiment 2 (Incorrect
Responses) .....................................................................................................................................76
Table 8: HLM Analysis Of Young Adults Of Experiment 2 (Incorrect Responses) .....................77
Table 9: IAPS Stimuli Used In Experiment 3 ................................................................................78
Table 10: HLM Analysis Of Experiment 3 (Recognition Memory) ..............................................79
Table 11: HLM Analysis Of Experiment 3 (Fixation Count Biases) ............................................80
Table 12: Additional HLM Analysis Of Experiment 3 (Recognition Memory) ...........................81
Figure 1: Experimental Procedure Used In Experiments 1 And 2 .................................................82
Figure 2: Experimental Procedure Used In Experiment 3 .............................................................83
Figure 3: Mean Pupil Sizes of Experiment 3 .................................................................................84
1
Abstract
Emotional arousal changes the way visual information is selectively processed.
Emotional stimuli dominate attention and are better remembered at the expense of competing
stimuli. But exposure to an emotionally arousing stimulus also influences attention biases to
subsequently presented neutral stimuli. Evidence from cognitive aging literature suggests that
emotional influences on attention are mostly preserved. Arousal-biased competition (ABC)
theory (Mather & Sutherland, 2011) hypothesizes that emotional arousal modulates the
selectivity of attention by increasing the effects of bottom-up attention biases. Yet no studies
have examined whether emotional arousal increases or decreases subsequent attention biases
when bottom-up and top-down attention are simultaneously manipulated. Moreover, no studies
have examined whether cognitive aging modulates the effects of emotional influences on
subsequent attention. In three experiments the effects of emotional arousal on subsequent
bottom-up and top-down attention biases are examined in both older and young adults. The
results indicate that negative arousal enhances attention biases that are driven by perceptual
salience in older and young adults, but positive arousal only increases bottom-up attention biases
in young adults. However, both positive and negative arousal weakened subsequent top-down
attention biases in both age groups, and this effect did not interact with bottom-up attention.
2
CHAPTER 1
Emotion, Attention and Cognitive Aging: The Effects of Emotional Arousal on Subsequent
Visual Processing
1.1 Introduction
Natural settings overwhelm the visual system with more stimuli than the brain can
process. We do not “see” everything in view, but instead selectively process a subset of what the
eyes detect. Attention organizes visual input into single objects that are experienced as separate
from a larger background, thus a large part of attention is filtering out sensory signals so that a
subset of them are processed at a higher level, where they ultimately reach awareness and can
influence behavior. Attention is studied behaviorally through measures of latency and accuracy
in identifying visual targets (Wolfe, 1994) and by measuring eye movements towards competing
stimuli (Duchowski, 2007). It is also studied physiologically by correlating behavioral responses
with single cell recordings in the primate brain (Chelazzi, Duncan, Miller, & Desimone, 1998;
Chelazzi, Miller, Duncan, & Desimone, 2001), and hemodynamic responses in the human brain
(Beck & Kastner, 2005; Kastner, De Weerd, Desimone, & Ungerleider, 1998). Classic theories
of attention focus on the stage of visual processing that selective attention occurs (Broadbent,
1958; Deutsch & Deutsch, 1963), however more recent theories have emphasized how stimuli
come to dominate attention when they compete spatially or temporally for limited resources
(Beck & Kastner, 2009; Bundesen, 1990; Desimone & Duncan, 1995).
As a stimulus enters the visual field, it forms a representation in visual cortex, and with
multiple stimuli in view, each representation is forced to compete with other representations to
attract attention. This competition in visual cortex manifests as mutual suppression, as each
representation inhibits competitors (Beck & Kastner, 2009; Desimone & Duncan, 1995; Kastner
3
& Ungerleider, 2001). However, if a given stimulus is more perceptually salient, meaning its
perceptual features contrast highly with other stimuli in view, its representation will dominate
(see Itti & Koch, 2001). Dominant representations are then selectively enhanced by high-level
cortical regions, which lift the suppressive influences of the competing representations (Chelazzi,
et al., 1998; Kastner, et al., 1998). This is the common view of attention, and is described as
biased competition, for attention biases the activity in visual cortex by strengthening certain
representations and weakening others.
Two factors primarily bias attention—perceptual salience and goal-relevance. Perceptual
salience refers to biases in attention that are described as bottom-up, for the source of the bias
arises externally (Nothdurft, 2000; Proulx & Egeth, 2008). But perceptual salience interacts with
top-down attention, which is driven by the relevance of a particular stimulus (Beck & Kastner,
2009; Desimone & Duncan, 1995). Bottom-up and top-down attention together determine
stimulus priority in the competition for selective attention (Fecteau & Munoz, 2006). However,
the motivational relevance of a stimulus can also bias attention and influence the effects of
stimulus priority (Dolan, 2002).
Emotional stimuli can dominate attention in the presence of stimuli that are perceptually
salient (Niu, Todd, & Anderson, 2012) or in the presence of stimuli that are goal-relevant
(Hodsoll, Viding, & Lavie, 2011; Muller, Andersen, & Keil, 2008). Thus in addition to bottom-
up attention and top-down attention, emotional attention influences whether a particular sensory
signal is selected for further processing (Todd, Cunninghan, Anderson, & Thompson, 2012;
Vuilleumier, 2005). A number of different experimental procedures have been used to
demonstrate emotional biases in attention, and although much of this evidence comes from
studies that use negative stimuli (Armony & Dolan, 2002; Carlson, Reinke, & Habib, 2009;
4
Ohman, Flykt, & Esteves, 2001), there is also evidence that positive stimuli likewise dominate
attention (Alpers & Pauli, 2006; Brosch, Sander, Pourtois, & Scherer, 2008; Sheth & Pham,
2008).
Furthermore, stimuli compete for attention not only across space, but also across time. If
stimuli are presented individually in the same location and in rapid succession, not all of them
will attract attention (Chun & Potter, 1995; Raymond, Shapiro, & Arnell, 1992). When a rapid
stream of stimuli is presented, the stimuli that do attract attention will inhibit attention to stimuli
that are presented immediately after, and when an emotional stimulus is presented in the stream,
subsequent stimuli are even more likely to be missed (Most, Chun, Widders, & Zald, 2005; Most,
Chun, Johnson, & Kiehl, 2006; Most, Smith, Cooter, Levy, & Zald, 2007), which is referred to
as emotion-induced blindness. On the other hand, if enough time elapses the opposite occurs, as
subsequently presented neutral stimuli are more likely to be identified when preceded by an
emotional stimulus (Bocanegra & Zeelenberg, 2009a; Phelps, Ling, & Carrasco, 2006). This
effect has been dubbed emotion-induced hypervision. Thus exposure to an emotional stimulus
results in attention being directed to the emotion-eliciting stimulus, but once the emotional
stimulus is removed and its representation subsides, attention to subsequently presented neutral
stimuli is enhanced.
Moreover, very few studies have examined the influence of cognitive aging on emotion-
induced changes in subsequent attention. Nevertheless a number of studies have examined the
effects of cognitive aging on emotional attention, and this evidence is rather intriguing. Age-
related decline in attention has been well documented in the cognitive aging literature (Braver &
Barch, 2002; Hedden & Gabrieli, 2004), yet there is evidence that emotional influences on
attention are relatively preserved (Hahn, Carlson, Singer, & Gronlund, 2006; Mather & Knight,
5
2006; Ruffman, Ng, & Jenkin, 2009). In other words, compared to other forms of attention,
emotional attention is less susceptible to the effects of cognitive aging. Yet these findings have
only been demonstrated in studies examining attention biases towards emotional stimuli. It is
unclear whether emotional influences on subsequent attention are maintained in older age.
1.2 Emotion and Visual Processing
A wide range of experimental evidence suggests that emotionally arousing stimuli
dominate attention to a greater extent than neutral stimuli, which has been demonstrated with a
number of different experimental procedures. Continuous flash suppression occurs when one eye
is exposed to moving visual noise, and the second eye is exposed to a static image (Tsuchiya &
Koch, 2005), forcing the two stimuli to compete for attention. During their presentation the
visual noise dominates awareness. But as the contrast level of the static image is increased, it
eventually pops into awareness. However, if the static image is emotionally arousing, such as a
face expressing fear, it reaches awareness at lower contrast levels, suggesting that visual
processing is particularly sensitive to stimuli that elicit emotion (Yang, Zald, & Blake, 2007).
And similar effects are observed in binocular rivalry (Blake & Logothetis, 2002). In this
procedure both eyes are shown static images, and each image takes turns rotating in and out of
awareness. Attention is measured by the duration each image occupies awareness and by which
image is the first to do so. When an emotional image competes directly with a neutral image, the
emotional image will more often be the first to be reported, and will dominate awareness more
often throughout the trial (Alpers & Pauli, 2006; Sheth & Pham, 2008). And these effects are not
limited to complex images. Similar findings have been observed in experiments using schematic
faces with neutral or emotional expressions (Alpers & Gerdes, 2007), as well as with line
gratings known as ‘gabor’ stimuli that are either superimposed on emotional faces (Bannerman,
6
Milders, De Gelder, & Sahraie, 2008), or associated with electric shock (Alpers, Ruhleder, Walz,
Muhlberger, & Pauli, 2005). Thus studies using continuous flash suppression and binocular
rivalry to measure attention indicate that when an emotional stimulus is in direct competition
with a neutral stimulus, the emotional stimulus dominates attention.
Another way of measuring attention is to embed a target stimulus within an array of
distracters, and to measure how quickly and accurately the presence or absence of the target is
reported. In visual search tasks emotionally arousing targets are more quickly identified than
neutral targets. For example, participants are faster and more accurate at identifying snakes or
spiders compared to neutral items like mushrooms and flowers (Ohman, et al., 2001). And
similar effects have been demonstrated using schematic faces expressing anger (Hahn, et al.,
2006). Moreover, when participants simply have to report the presence or absence of an oddball
stimulus, emotional oddballs are more quickly identified (Mather & Knight, 2006). Thus when
the presence of a target must be detected among task-irrelevant distracters, performance is
enhanced when the target is emotional.
Other studies using spatial cues to bias attention reveal similar effects. When a visual cue
is briefly displayed to direct attention to a particular location, stimuli that appear in that region
are more quickly identified. And similar effects have been demonstrated with fear-conditioned
spatial cues (Armony & Dolan, 2002), indicating that attention rapidly orients to locations
occupied by emotionally arousing stimuli. Similarly, in dot-probe studies stimuli are briefly
presented and then removed, with a dot-probe appearing behind a given stimulus. How quickly
the probe is identified indicates how much attention was directed to the stimulus in that location.
When probes replace emotionally arousing objects, compared to neutral objects, probes are more
7
quickly identified (Carlson, et al., 2009; Pourtois, Grandjean, Sander, & Vuilleumier, 2004),
again suggesting that emotional stimuli attract more attention than neutral stimuli.
Yet another way of measuring attention is to track eye movements, which can identify
attention biases over longer durations of time. Eye tracking studies reveal that individuals are
more likely to first fixate on a face or a scene if it is emotional, and they will continue to fixate
more on an emotional stimulus throughout its presentation (Knight, et al., 2007; LaBar, Mesulam,
Gitelman, & Weintraub, 2000). Similarly, when pictures of visual scenes are placed in one’s
periphery, initial and subsequent eye fixations are greater for emotional scenes compared to
neutral scenes (Calvo & Lang, 2005; Rösler, et al., 2005), even under direct instructions to
ignore the emotional stimuli (Nummenmaa, Hyona, & Calvo, 2006). Thus, eye tracking studies
also illustrate that emotional stimuli dominate attention, even in situations where there is enough
attention to temporarily prioritize each competing stimulus.
Taken together, a number of studies indicate that when multiple stimuli compete for
attention, if an emotional stimulus is present, it will dominate. However, the evidence so far
discussed in this section has used procedures that measured competition across space. Another
way of measuring competition is across time. Rapid-serial-visual-presentation (RSVP) involves
quickly displaying a series of images individually in the same location, and can demonstrate how
stimuli compete for attention across time (Chun & Potter, 1995; Raymond, et al., 1992). When
attempting to identify two targets in a sequence, if they are too close in the sequence, the second
target is missed because the first target is still being processed. This effect is called ‘attentional
blink,’ as attention is temporarily unavailable and seemingly blinks when a new stimulus appears.
And attentional blink effects are stronger when the first stimulus to occupy attention is emotional.
For example, if an emotional stimulus precedes a neutral target, the neutral target is more often
8
missed, as emotional stimuli dominate attention to a greater extent, thus suppressing attention to
subsequently presented targets (Most, et al., 2005; Most, et al., 2006; Most, et al., 2007),
producing emotion-induced blindness. In contrast, if the second of two targets is emotional,
attentional blink effects weaken because emotional stimuli are less susceptible to the effects of
suppression (Anderson & Phelps, 2001; Anderson, 2005; De Martino, Kalisch, Rees, & Dolan,
2009; Keil & Ihssen, 2004; Milders, Sahraie, Logan, & Donnellon, 2006; Stein, Peelen, Funk, &
Seidl, 2010). Thus across both space and time emotional stimuli dominate attention.
In summary, evidence from a variety of experimental procedures indicates that emotional
stimuli are selectively processed to a greater extent than neutral stimuli. However, a unique set of
observations have surfaced to reveal that emotion-related increases in attention, which are
usually observed for emotional stimuli, can carry over into subsequent visual processing,
resulting in increased attention to emotionally neutral stimuli.
1.3 Emotion and Subsequent Visual Processing
As mentioned in section 1.2, attention is not only limited across space but it is also
limited across time. Studies using RSVP have uncovered evidence revealing that emotional
stimuli are prioritized in the competition for attention. Emotional stimuli are less subject to
temporary suppression when they are the second of two subsequent targets (Anderson & Phelps,
2001; Anderson, 2005; De Martino, et al., 2009), and they are also more likely to interfere with
subsequent visual processing when presented right before a neutral target (Most, et al., 2005;
Most, et al., 2006; Most, et al., 2007). However, a wave of recent findings show the opposite
effect, namely that presenting an emotional stimulus before a neutral target enhances, rather than
suppresses, visual processing of the neutral target.
9
The first study to demonstrate such an effect used low-level visual gabor targets, and
faces expressing fear to elicit the emotional response. Gabor stimuli are lined gratings that vary
in frequency, contrast and spatial orientation and are commonly used to study early stages of
vision. When searching for an oddball gabor stimulus that differs in spatial orientation from
distracter gabor stimuli, presenting a fearful face beforehand enhances performance (Phelps, et
al., 2006), showing that exposure to an emotionally arousing stimulus enhances subsequent
attention at an early stage of visual processing. And similar effects have been observed with
aversively conditioned gabor stimuli. When aversively conditioned auditory or visual cues
precede the presentation of a gabor stimulus, the orientation of the gabor stimulus is more easily
detected (Padmala & Pessoa, 2008). And follow up studies show that emotional stimuli enhance
visual processing of subsequently presented gabor stimuli with low spatial frequencies (i.e.
thicker gratings), while impairing processing of those with high spatial frequencies (i.e. thin
gratings) (Bocanegra & Zeelenberg, 2009b; Lee, Baek, Lu, & Mather, in press), suggesting that
the coarse properties of a stimulus, like the general outline of an image, are the visual properties
that emotions enhance, while emotions may suppress visual processing of more detailed
properties.
Similarly visual search for oriented lines are enhanced by a preceding emotional stimulus,
but only when the difference in orientation is high rendering the oddballs perceptually salient
(Lee, Itti, & Mather, 2012). When the difference in orientation is low, emotional arousal
decreases attention to oriented lines, indicating that emotional arousal increases and decreases
subsequent attention to high-salient and low-salient stimuli, respectively. Similarly, exposure to
an emotionally arousing stimulus increases attention biases to perceptually salient targets and
10
decreases attention to less salient targets (Sutherland & Mather, 2012). Thus emotional arousal
enhances subsequent attention biases to perceptually salient stimuli that are emotionally neutral.
In addition, other studies reveal that emotional arousal enhances and impairs subsequent
visual processing, depending on the amount of time separating the emotional stimulus form the
neutral target. For instance, when an emotionally arousing word is briefly shown, the ability to
identify a neutral word that immediately follows is impaired (Bocanegra & Zeelenberg, 2009a),
for the emotional word dominates attention leading to emotion-induced blindness (see Most, et
al., 2005). Yet when enough time elapses (i.e. 1000 ms), the representation of the emotional
word subsides and the ability to identify the neutral word is enhanced, which is referred to as
emotion-induced hypervision (Bocanegra & Zeelenberg, 2009a).
And other studies have demonstrated emotion-induced blindness and hypervision within
the same experiment. Negative arousing photographs were used as arousing cues, and complex
visual scenes were used as targets (Ciesielski, Armstrong, Zald, & Olatunji, 2010). When the ISI
ranged from 200-600 ms, emotion-induced blindness occurred but when the ISI ranged from
800-1000 ms emotion-induced hypervision occurred, closely replicating the findings of
Bocanegra and Zeelenberg (2009a). And there is evidence that emotion-induced hypervision
immediately occurs when the emotional stimulus and the neutral target are presented through
different sensory modalities (Zeelenberg & Bocanegra, 2010), for there is less competition across
auditory and visual cortices.
Taken together these findings suggest that attention biases are enhanced when they are
preceded by an emotionally arousing stimulus. But it is unclear whether similar effects are
observed when positive arousing stimuli are used to elicit emotion, for all of the studies
discussed in this section used negative arousing stimuli. Moreover, none of these studies have
11
teased apart the effects of emotional arousal on bottom-up and top-down attention, with the
exception of Sutherland and Mather (2012), whom presented a group of targets that were equally
goal-relevant, but differed in perceptual salience. This allowed attention to both high-salient
stimuli and low-salient stimuli to be measured simultaneously. Yet no studies have presented
multiple targets of equal salience that differed in goal-relevance, or manipulated goal-relevance
and perceptual salience orthogonally to tease apart how emotional arousal influences both forms
of attention. Therefore, although there is convincing evidence that negative arousal amplifies
subsequent attention biases that are determined by differences in perceptual salience, it is unclear
whether similar effects will occur when attention biases are determined by goal-relevance, and
whether positive and negative arousal have the same effects on subsequent attention biases.
1.4 Arousal-Biased Competition (ABC) Theory
Section 1.2 outlined the different procedures that have been used to demonstrate attention
biases towards emotionally arousing stimuli. And section 1.3 outlined evidence for emotion-
induced biases in subsequent attention towards neutral stimuli. In particular, the evidence
suggested that emotional arousal enhanced subsequent attention biases when the target stimulus
was presented alone (Bocanegra & Zeelenberg, 2009a; Ciesielski, et al., 2010; Padmala &
Pessoa, 2008; Zeelenberg & Bocanegra, 2010), among a group of homogenous distracters that
were all similar (Lee, Itti, & Mather, 2012; Phelps, et al., 2006), or among other targets that had
relatively lower levels of perceptual salience (Sutherland & Mather, 2012). In each of these
studies the target stimulus was perceptually salient, either because the distracters were
homogenous leading to visual ‘pop out,’ the targets were presented alone, or because the targets
contrasted more strongly than the competing targets with the background on which they were
12
displayed. Thus each of these findings, along with attention biases to emotionally arousing
stimuli can be explained by ABC theory.
According to ABC theory (Mather & Sutherland, 2011), emotional arousal amplifies
biased competition processes in visual cortex. Visual perception is highly sensitive to
emotionally arousing stimuli, and like perceptually salient stimuli, attract attention reflexively
with no cognitive effort. But emotional stimuli also elicit a physiological arousal response in the
brain and body that works to keep the representation of the emotional stimulus strengthened
(Bradley, Codispoti, Cuthbert, & Lang, 2001; Bradley, 2009). This is why emotional stimuli not
only immediately dominate attention, but it is also why they continue to dominate over time.
Such effects are demonstrated in binocular rivalry and eye tracking procedures where not only is
the emotional stimulus either the first representation to enter awareness or the first stimulus to
attract a fixation, but they also continue to dominate awareness and continue to attract more
fixations over time (Alpers, et al., 2005; Alpers & Pauli, 2006; Bannerman, et al., 2008; Alpers
& Gerdes, 2007; Knight, et al., 2007; LaBar, et al., 2000; Sheth & Pham, 2008). However, ABC
theory hypothesizes that when an emotional stimulus is briefly presented, removed, and then
replaced by competing neutral stimuli, whichever individual stimulus dominates attention
because it is perceptually salient does so to an even greater extent.
But when attention is driven by differences in goal-relevance, it is unclear whether
exposure to an emotionally arousing stimulus will increase or decrease subsequent attention
biases. Unlike bottom-up attention, top-down attention depends more on cognitive control
processes because a target stimulus has to be held in working memory, and bottom-up attention
to distracters must be inhibited. There is some indication that exposure to negative arousing
stimuli increases subsequent attention biases that are driven by goal-relevance, but these effects
13
have only been observed with faces expressing fear and are only observed when emotional trials
and neutral trials are blocked rather than intermixed (Becker, 2009; Quinlan & Johnson, 2011).
Moreover, these studies failed to control for the effects of perceptual salience among the targets
and irrelevant distracters.
Alternatively, there is also reason to assume that emotional arousal may impair
subsequent attention biases that driven by goal-relevance, particularly when the preceding
emotional stimulus has high levels of arousal. For example, the same neural structures that are
involved in top-down selective attention are also activated by exposure to emotionally arousing
stimuli (Pessoa, 2009). Moreover, prior exposure to an emotionally arousing stimulus interferes
with subsequent inhibitory responses that rely on cognitive control (Verbruggen & De Houwer,
2007). Thus exposure to an emotionally arousing stimulus may interfere with subsequent top-
down attention because the processing of the emotionally arousing stimulus consumes cognitive
control resources, which are also needed for top-down attention.
ABC theory explains how emotional arousal increases subsequent attention biases that
are driven by perceptual salience, but regarding top-down attention, such biases are only
enhanced when competing neutral stimuli are encoded before, rather than after, emotional
arousal is elicited, resulting in greater memory for goal-relevant items presented beforehand (see
Sakaki, Fryer, & Mather, in press). When emotional arousal has already been elicited, and top-
down attention is used to bias visual processing to a subset of neutral competitors, attention
biases may be weakened, as processing an emotionally arousing stimulus consumes from the
same pool of resources that govern top-down attention biases (Pessoa, 2009). However, it is
unclear how emotional arousal influences subsequent attention biases when both bottom-up
attention and top-down attention are used to direct visual processing to a given neutral stimulus.
14
When a goal-relevant stimulus is also perceptually salient, then emotional arousal may increase
subsequent attention biases. On the other hand, when a goal-relevant stimulus has relatively low-
salience, then emotional arousal may interfere with subsequent attention biases. Moreover, a
third possibility is that emotional arousal may interfere with top-down attention biases regardless
of whether the goal-relevant stimulus is salient or not.
In addition, the effects of emotional arousal on subsequent attention biases in older adults
is a relatively unexplored topic, and thus it is unclear whether ABC theory can account for the
effects that arousal has on the subsequent selectivity of attention in older adults. There is
evidence that biases in attention to emotional stimuli remain intact in older age (Hahn, et al.,
2006; Mather & Knight, 2006; Ruffman, et al., 2009), however in some instances these
emotional biases differ from those observed in young adults (Bannerman, Regener, & Sahraie,
2011; Isaacowitz, Wadlinger, Goren, & Wilson, 2006a, 2006b; Knight, et al., 2007; Mather &
Carstensen, 2003) due to an added emphasis on emotional regulation (see Mather & Carstensen,
2005). Nevertheless, it remains to be seen if cognitive aging changes the influence that emotional
arousal has on subsequent attention biases. However, in the next section I discuss what is known
about the effects of emotion on attention in older adults, and develop several hypotheses about
the effects emotional arousal may have on subsequent visual processing in older adults.
1.5 Emotion and Visual Processing in Older Adults
One of the more interesting findings in the emotion and cognitive aging literature is that
emotional influences on attention and memory change as individuals approach older age. Most of
these changes are related to emotional regulation, which becomes more emphasized at the end of
the lifespan (Carstensen, Fung, & Charles, 2003; Reed & Carstensen, 2012). In young adults a
wide range of evidence suggests that both positive emotional stimuli and negative emotional
15
stimuli are more likely to attract attention, and are more likely to be remembered, compared to
neutral stimuli (Mather, 2007; Kensinger, 2009). But studies examining the effects of cognitive
aging on emotional attention and emotional memory indicate that negative stimuli less often
attract attention and less often dominate memory, while positive stimuli attract increased
attention and are remembered to a greater extent in older adults compared to young adults (see
Mather & Carstensen, 2005). This increased emphasis on positive stimuli and this decreased
emphasis on negative stimuli is known as the ‘positivity effect’ in aging.
Initial evidence for positivity effects in attention arose from studies using dot-probe
procedures where multiple images are briefly presented and then removed, followed by a probe
that replaces either image. How quickly one responds to the probe indicates whether attention
was directed to the image it replaced. It has been observed that when older adults are shown two
images, unlike younger adults, they are slower to respond to probes replacing a negative image
(Mather & Carstensen, 2003), suggesting that older adults inhibit attention away from negative
stimuli.
Studies using eye tracking to measure attention report similar effects. When several
stimuli are presented, older adults are less likely to fixate on an emotionally negative stimulus,
and are more likely to fixate on a positive stimulus, as measured by the initial fixation and by the
duration of each fixation (Isaacowitz, et al., 2006a, 2006b; Rösler, et al., 2005). Similarly,
studies using binocular rivalry show that unlike young adults, older adults suppress attention to
negative stimuli, but like young adults, show increases in attention to positive stimuli
(Bannerman, et al., 2011). Moreover, studies measuring event-related potentials (ERP) in
response to viewing negative and positive arousing stimuli indicate that across the lifespan
16
neural responses to negative stimuli decline in later age, but neural responses to positive stimuli
remain consistent (Kisley, Wood, & Burrows, 2007).
And evidence suggests that positivity effects arise from cognitive control processes,
despite the fact that cognitive control processes decline with age (Braver & Barch, 2002; Hedden
& Gabrieli, 2004). When attention is consumed by a secondary task, the tendency to fixate less
on a negative stimulus is no longer observed (Knight, et al., 2007; but see Allard & Isaacowitz,
2008), suggesting that cognitive control is used to direct attention away from negative stimuli.
Moreover, neural activity in regions associated with cognitive control, particularly the anterior
cingulate cortex (ACC), increase in activity when older adults show biases towards positive
stimuli (Brassen, Gamer, & Buchel, 2011), suggesting that cognitive control is also used by older
adults to direct attention to positive stimuli. In addition, studies examining visual search indicate
that attention is initially biased towards negative stimuli, but over time these biases become
suppressed (Ortega, 2011), suggesting that older adults show the same biases as younger adults,
but use cognitive control to inhibit attention away from negative stimuli. Moreover, these
behavioral findings are consistent with evidence that neural structures underlying emotional
regulation show little to no age-related decline, unlike other regions associated with other forms
of cognitive control (Fjell, et al., 2009). Taken together, these findings indicate that increases in
attention to positive stimuli, and decreases in attention to negative stimuli both arise from
cognitive control processes that enhance or suppress attention to emotional stimuli based to the
valence of the stimulus.
Nevertheless, the possibility that older adults initially show biases in attention to negative
stimuli before cognitive control is utilized to inhibit these biases (i.e. Ortega, 2011) has been
shown in other studies using visual search. When an oddball face is embedded within an array of
17
homogenous distracter faces, the presence or absence of the target is more quickly and accurately
reported if it is negative (Hahn, et al., 2006; Mather & Knight, 2006; Ruffman, et al., 2009). And
similar effects have been demonstrated with more complex images, such as pictures of snakes
and spiders (Leclerc & Kensinger, 2008). These measures of attention reflect immediate
attention biases that occur before cognitive control can be implemented, indicating that attention
biases to negative arousing stimuli are still present in older adults. In addition, it has been
demonstrated that identifying the font color of a word is more difficult for older adults if the
word is highly arousing, and this occurs regardless of the valence of the word (Wurm, Labouvie-
Vief, Aycock, Rebucal, & Koch, 2004). Similarly, in a meta-analysis performed on studies
examining emotional salience, or the degree to which emotional stimuli dominate attention and
memory among different age groups, it was revealed that older and young adults showed
consistent biases in attention and memory for both positive and negative stimuli (Murphy &
Isaacowitz, 2008). So although positivity effects show that attention biases to emotional stimuli
change in older age, this only occurs when attention is measured in such a way as to allow
cognitive control to influence attention. Otherwise, older and young adults show the same
emotional biases in attention, indicating that emotional attention is preserved in older age.
But very few studies have examined the effects of cognitive aging on emotion and
subsequent visual processing. Mickley-Steinmetz and colleagues (2010) recently reported
evidence demonstrating emotion-induced hypervision in older adults. In this study words were
presented in rapid succession (RSVP), and it was revealed that words of either positive or
negative valence led to increases in the ability to identify subsequently presented neutral words.
This finding reveals initial evidence that cognitive aging does not impact emotion-induced
increases in attention to subsequently presented neutral targets. However, this experiment
18
controlled for the effects of arousal and therefore only demonstrates valence-based effects on
subsequent attention. Nevertheless this finding gives some indication that older adults adhere to
the hypotheses of ABC theory, which suggests that emotional arousal, once elicited, enhances
subsequent attention to neutral stimuli when they appear salient due to there being no competing
stimuli in view.
In summary, a number of studies show that emotional biases in attention are preserved in
older adults (Hahn, et al., 2006; Leclerc & Kensinger, 2008; Mather & Knight, 2006; Murphy &
Isaacowitz, 2008; Wurm, et al., 2004). However, when cognitive control resources become
available, these biases are modulated to inhibit processing of negative stimuli and to enhance
processing of positive stimuli (Brassen, et al., 2011; Knight, et al., 2007; Ortega, 2011). And
there is evidence that exposure to stimuli of positive or negative valence leads to subsequent
increases in visual processing of neutral stimuli (Mickley-Steinmetz, Muscatell, & Kensinger,
2010), but it is still unclear whether prior exposure to emotionally arousing stimuli will have the
same effect, and whether such effects will be observed when the subsequent neutral stimulus
competes spatially for attention with other stimuli in view.
1.6 Overview
The three experiments reported had three primary goals. The first was to examine
whether older adults and young adults show similar increases in attention biases that are driven
by differences in perceptual salience following exposure to an emotionally arousing stimulus.
The second was to examine whether negative and positive arousal have similar effects on
emotion-induced increases in attention biases driven by perceptual salience. The third goal was
exploratory and involved examining whether attention biases driven by differences in goal-
relevance is influenced by emotional arousal, regardless of the perceptual salience of the goal-
19
relevant targets. Moreover, we examined whether subjective arousal responses and physiological
arousal responses could predict attention biases on a trial-by-trial basis, and whether emotion-
induced changes in attention are accompanied by changes in memory.
Previous studies examining emotional influences on subsequent attention biases have
used visual search, where a single target is presented among distracters. An alternative to visual
search is to present multiple targets, and to bias attention to a subset of the targets while still
obtaining a measure of attention for the perceptually salient or goal-relevant stimuli, as well as
the competing stimuli that are less salient and less relevant. Only one such study has used this
type of procedure to measure emotional influences on subsequent attention biases, and attention
was manipulated by presenting equally relevant targets with relatively high and low perceptual
salience (Sutherland & Mather, 2012). Moreover, so far only one study has examined age
differences regarding emotion-induced changes in subsequent attention, however, this study
controlled for differences in arousal, meaning the observed effects were valence-based (Mickley-
Steinmetz, et al., 2010). Thus it is still unclear whether positive and negative arousal leads to
different effects on subsequent attention biases in young and older adults, and whether the ABC
hypothesis can explain subsequent attention biases in older adults. Furthermore, no studies have
directly examined the effects of emotional arousal on top-down selective attention biases while
simultaneously manipulating the perceptual salience of the goal-relevant targets, and whether
cognitive aging impacts such biases.
Experiment 1 will examine whether increases in subsequent attention biases that are
driven by differences in perceptual salience can be replicated in an older adult sample. To
directly examine the question of age, data from Sutherland and Mather (2012) are pooled with a
new sample of older adults to compare the two age groups, allowing the use of age group as a
20
factor in the experimental analysis. Given that older adults and young adults have different
subjective and physiological responses to emotional stimuli (Grühn & Smith, 2008; Keil &
Freund, 2009; Levenson, Carstensen, Friesen, & Ekman, 1991), it was of interest to control for
these possible differences by using participant’s subjective ratings of the emotional stimuli as a
marker of emotional arousal, rather than the predefined norms that reflect subjective ratings of
young adults. Therefore, hierarchical linear models (HLM) are used to examine attention biases
on a trial-by-trial basis depending on each participant’s subjective arousal and valence rating of
the emotional stimulus presented on that particular trial.
The goal of Experiment 2 was to replicate the findings of Experiment 1, and to examine
whether positive arousal and negative arousal have similar effects on subsequent attention biases,
and whether age has an influence on the outcome. The procedure was identical to the procedure
used in Experiment 1, however the number of trials was doubled to accommodate the inclusion
of both positive and negative stimuli. Moreover, as in Experiment 1 we use individual valence
and arousal ratings to predict recall of high-salient targets and low-salient targets.
The goal of Experiment 3 was to examine how emotional arousal influences subsequent
attention biases that are determined by goal relevance, and whether this effect interacts with
differences in perceptual salience. With too much in view to process, attention is directed to the
most perceptually salient stimulus in view, and emotional arousal increases the selectivity of
bottom-up attention (see Sutherland & Mather, 2012). But unlike bottom-up attention, top-down
attention requires cognitive control. Thus it is unclear whether emotional arousal will have the
same effect on top-down attention, as there is evidence that the processing of emotional stimuli
consumes from the same limited pool of resources that underlie cognitive control processes
21
(Pessoa, 2009). Moreover, it is also of interest to determine whether the effects of emotional
arousal on top-down attention depend upon the perceptual salience of the goal-relevant stimulus.
To examine these alternatives, in Experiment 3 we exposed participants to positive and
negative arousing stimuli before displaying two items that competed for attention. One of the
items was always goal-relevant but we manipulated its perceptual salience to examine if top-
down and bottom-up signals interact when being influenced by emotional arousal. Moreover,
rather than having participants behaviorally respond to the goal-relevant stimulus, their eye
movements were instead recorded as they were directed to remember one of the two stimuli
displayed. In addition, rather than measuring emotional arousal through subjective report, it was
instead measured by indexing physiological arousal responses to emotional stimuli, which was
done by measuring changes in pupil size. And participants also completed a recognition memory
test for the distracter items and the goal-relevant items.
According to ABC theory, exposure to an emotionally arousing stimulus should increase
attention to high-salient targets in Experiments 1 and 2. On the other hand, when attention biases
are driven by differences in goal-relevance and perceptual salience, it is unclear what kind of
influence emotional arousal will have on subsequent attention biases. According to the dual
competition framework (Pessoa, 2009), prior exposure to an emotionally arousing stimulus
should interfere with top-down attention, as processing emotional stimuli draws from the same
limited resources that are used to implement cognitive control, which is needed for top-down
attention biases. On the other hand, according to ABC theory (Mather & Sutherland, 2011) prior
exposure to emotionally arousing stimuli should enhance bottom-up attention, as expected in
Experiments 1 and 2. Yet it is unclear how emotional arousal will influence subsequent biases
when both forms of attention are used to manipulate visual processing.
22
As for the effects of aging on subsequent attention, it is unclear whether older adults will
perform similarly as young adults in all three experiments. Evidence suggests that older adults
use cognitive control processes to increase attention to positive arousing stimuli (Brassen, et al.,
2011), and to decrease attention to negative arousing stimuli (Knight, et al., 2007), resulting from
an increased emphasis on regulating emotions in later life (Mather & Carstensen, 2005). From
this perspective one might expect exposure to emotionally arousing stimuli of either valence to
weaken subsequent top-down attention biases to a greater extent in older adults compared to
young adults. Although top-down attention biases are hypothesized to be weakened in young
adults due to prior exposure to an emotionally arousing stimulus (Pessoa, 2009), this effect
should be stronger in older adults, for older adults use cognitive control processes to up-regulate
or down-regulate attention biases to emotional stimuli, depending on valence. With respect to
bottom-up attention, because cognitive control resources are not needed for this type of attention,
older adults should replicate the findings previously observed in young adults (Sutherland &
Mather, 2012), as both age groups should show stronger biases in attention to perceptually
salient stimuli after being exposed to an emotionally arousing stimulus.
This leaves us with two hypotheses. The first is the ABC hypothesis, which predicts that
in both age groups exposure to an emotionally arousing stimulus, regardless of the emotional
valence, will increase subsequent attention biases to perceptually salient stimuli. Cognitive
control is not needed to facilitate bottom-up attention biases, thus even if older adults use
cognitive control to influence perceptual processing of the emotional stimulus, simply being
exposed to an arousing stimulus should carry over and amplify subsequent bottom-up attention,
as has been observed in young adults (Sutherland & Mather, 2012). Thus the ABC hypothesis
applies to both age groups in Experiments 1 and 2. As for Experiment 3, according to the dual-
23
competition framework (Pessoa, 2009), processing an emotional stimulus will consume from the
same pool of resources that are needed to control top-down attention, so biases in attention based
on differences in goal-relevance should be weakened by prior exposure to an emotionally
arousing stimulus. And these interference effects should be even greater in older adults, as there
is evidence that they use cognitive control to up-regulate and down-regulate emotional
processing to regulate their emotional experience (Brassen, et al., 2011; Knight, et al., 2007).
Thus in Experiment 3 we have the cognitive control hypothesis, which predicts that emotional
arousal will inhibit subsequent biases to goal-relevant stimuli. Moreover, since both top-down
and bottom-up attention biases will be simultaneously manipulated, the cognitive control
hypothesis and the ABC hypothesis are integrated to predict that emotional arousal will decrease
attention biases to goal-relevant stimuli, and this should interact with bottom-up attention such
that the decrease in attention is greater when the goal-relevant stimulus is not perceptually salient.
CHAPTER 2
2.1 Experiment 1
2.2 Methods
2.2.1 Stimuli
Twenty negative arousing and twenty neutral non-arousing audio clips were chosen from
the International Affective Digital Sound (IADS) system (Bradley & Lang, 2007) to manipulate
negative arousal (see Table 1). Twenty-five letters from the alphabet were used as visual targets.
The letter ‘I’ was omitted due to its close resemblance with lower case ‘L.’ All letters were
presented in uppercase Arial font, with high-salience letters appearing in a dark grey font (RGB:
102 102 102) and low-salience letters appearing in a light grey font (RGB: 204 204 204). On
each trial 8 letters were randomly selected to appear in a circular array that surrounded the
24
fixation cross (see Figure 1), subtending 11.08° x 14.58° of visual arc. Three high-salience and 5
low-salience letters were presented on each trial, and salience type was randomly assigned.
2.2.2 Participants
Fifty-five older adults (27 female) ranging from ages 61 - 80 (M = 70.7, SD = 5.1) years
in age were recruited from the University of Southern California Healthy Minds database and
were compensated monetarily for their participation. Years of education ranged from 12 – 28
years (M = 16.4, SD = 3.1), and their vocabulary aptitude (Brown, Fishco, & Hanna, 1993)
varied from 0.24 – 0.96 (M = 0.73, SD = 0.17). In addition, on a scale from 1 – 10 self-reported
stress ranged from 1 – 9 (M = 4.0, SD = 2.2), with 10 representing highest stress. Self-reported
mental health ranged from 4 – 9 (M = 7.2, SD = 1.3) with 10 representing the highest level of
mental health. As for the young adults 110 participants were pooled from Sutherland and Mather
(2012). They were recruited from the USC student volunteer research pool and the University of
Southern California Healthy Minds database, ranging from 18 - 29 (M = 20.3, SD = 2.3) in age
(80 female) and their subjective stress ranged from 1 – 9 (M = 4.2, SD = 2.0), with 10
representing the highest level of stress.
2.2.3 Procedure
Five practice trials preceded the 40 experimental trials. Each trial began with a fixation
cross (4 seconds), followed by a 6-second audio recording presented via headphones (see Figure
1). Following the audio clip an inter-stimulus-interval (ISI) ranging from 750 – 3000
milliseconds (ms) occurred, followed by the array of letters that displayed for 200 ms, and
included 3 high-salient letters and 5 low-salient letters, which were randomly selected on a trial-
by-trial basis. Self-paced recall then took place after a 200 ms delay, and targets were reported
25
via key press. The fixation cross was displayed for the duration of the trial until the recall phase.
Participants were directed to fixate on the cross until the recall phase.
After the 40 experimental trials were completed, valence and arousal ratings were
provided for the audio clips, which was self-paced. Both the valence and the arousal scales
ranged from 1 – 9. On the arousal scale 1 indicated no arousal (9 representing most arousing) and
on the valence scale a 1 indicated most negative and 9 represented most positive (5 representing
neutral valence). Following the rating participants were debriefed and then excused from the
study.
2.2.4 Data Analysis
A two-tier hierarchical linear model (HLM) (Singer, 1998) and a 2 × 2 × 2 repeated-
measures ANOVA were used to test our competing hypotheses. In the ANOVA the emotion type
of the trial and the salience type of the competing targets were within-subject factors and age
group was a between-subject factor. In the HLM arousal and valence ratings and the mean
decibel (dB) level of each sound, along with age group, were used to predict recall of high-
salience targets, low-salience targets and difference scores (high-salience minus low-salience).
Difference scores were modeled to capture simultaneous increases and decreases in recall of
either target type. The HLM allowed the use of subjective arousal and valence responses of
participants to predict changes in recall due to increases or decreases in arousal or valence on a
trial-by-trial basis. The dB predictor captured changes in recall due to differences in the physical
intensity of the negative arousing and neutral sounds (see Bradley & Lang, 2000). The ANOVA
was based on the predefined emotion categories that were normalized with young adults, but
allowed a direct comparison between recall across the negative arousing and neutral trial types.
26
2.3 Results and Discussion
The first analysis was a 2 × 2 × 2 repeated-measures ANOVA that compared age group,
emotion type and salience type, and their respective interactions. The main effect of salience type
was significant, F(1,163) = 183.27, p < 0.001, η
p
2
= 0.529, indicating that more high-salience
targets (M = 0.577, SD = 0.157) were recalled compared to low-salience targets (M = 0.344, SD
= 0.137). In addition, salience interacted with arousal type F(1,163) = 8.227, p < 0.01, η
p
2
=
0.048, but no other factors in the model reached significance. To explore the arousal-salience
interaction, older adults were analyzed separately in a 2-way ANOVA, as it was previously
observed that arousal interacted with salience separately in the young adult group (see
Sutherland & Mather, 2012). Thus it was possible that young adults were driving the observed
interaction, and the results did in fact reveal that this was the case. Arousal and salience failed to
interact when older adults were analyzed in a separate model, F(1,54) = 1.537, p = 0.22. Thus
according to these results ABC effects were not observed in older adults. However, this finding
may have resulted from older adults having different subjective emotional reactions to stimuli
categorized as neutral and negative arousing, as previous studies have shown that older adults
have different subjective responses and physiological responses to stimuli that are designed to be
emotional, compared to younger adults (Grühn & Smith, 2008; Keil & Freund, 2009; Levenson,
et al., 1991).
For this reason we followed up the ANOVA analysis with a HLM analysis that used
subjective arousal and valence ratings of the audio clips to predict recall of high and low-salience
targets, as well as difference scores (high-salience minus low-salience) on a trial-by-trial basis.
In Table 2A we see that higher arousal ratings predicted greater difference scores, and we also
see that age predicted smaller difference scores, suggesting that arousal ratings led to greater
27
biases to recall high-salience targets compared to low-salience targets in both age groups.
Moreover, older age led to a relative decline in attention to high-salience targets. In addition,
higher arousal ratings predicted greater recall of high-salience targets (Table 2B), while arousal
had no influence on the recall of low-salience targets (Table 2C), suggesting that the increase in
difference scores was driven by participants recalling more high-salience targets on trials that
were more emotionally arousing, rather than a decline in the number of low-salience targets
detected on arousing trials. Moreover, older age predicted lower recall of high-salience targets
and lower recall of low-salience targets, suggesting that not only did older adults show weaker
biases in recalling high-salience targets compared to low-salience targets, but they also
demonstrated lower recall of both target types compared to young adults. In addition, average dB
levels of the sounds failed to predict recall in any of the three models, indicating that differences
in the physical intensity of the sounds had no influence in recalling high and low-salience targets.
Finally, in a separate HLM we modeled the number of incorrect letters reported on each
trial. This was done to ensure that participants that were reporting more letters correctly were not
also reporting more letters incorrectly, which could negate our interpretation that emotional
arousal increases biases to attend to salient stimuli. As expected, arousal ratings and mean dB
failed to predict incorrect responses, however the valence rating and age predictors did reach
significance (see Table 3). Thus when the effects of negative valence are separated from the
effects of arousal, negative valence can lead to more false reports of targets, suggesting that
targets are more easily confused with potential targets when the emotional stimulus is rated to be
more negative. In addition, the older age group was less likely to report incorrect letters,
suggesting that older adults were relatively less susceptible to false recall.
28
Taken together, the results of the HLM analysis support the ABC hypothesis, as higher
arousal ratings led to greater recall of high-salience targets in both age groups. However, unlike
the results reported in Sutherland and Mather (2012) arousal did not lead to decreases in recall of
low-salience targets. Furthermore, older adults gave fewer incorrect responses than young adults,
and lower valence ratings led to an increase in incorrect responses. Nevertheless, these results
indicate that negative arousal enhances subsequent attention biases in both older and young
adults when attention is driven by differences in perceptual salience.
CHAPTER 3
3.1 Experiment 2
According to ABC theory (Mather & Sutherland, 2011), arousing stimuli of either
valence should enhance subsequent attention biases. And given the results of Experiment 1,
namely that negative arousal increases subsequent attention biases to salient targets in both age
groups, similar results should be observed with positive arousing stimuli. Therefore, we
conducted this experiment using the exact procedure used in Experiment 1, but included positive
arousing sounds as well as negative arousing sounds as a manipulation of emotion.
3.2 Methods
3.2.1 Stimuli
A total of 80 IADS audio clips were used (Bradley & Lang, 2007), 40 of which were
emotionally neutral along with 20 positive arousing and 20 negative arousing clips (see Table 4).
Other than that, the materials used were identical to those reported in Experiment 1.
3.2.2 Participants
A total of 55 young adults (Female = 39) and 55 older adults (Female = 34) participanted
for a monetary compensation, or for course credit. Young adults ranged from 18 – 29 (M = 20.4,
29
SD =2.0) in age and older adults ranged from 60 – 86 (M = 72.2, SD = 7.4). On a scale from 1 –
10 measuring subjective stress young adults ranged from 1 – 9 (M = 4.53, SD = 2.00), and older
adults ranged from 1 – 9 (M = 3.71, SD = 2.36), however one older adult participant failed to
report their stress level. As for perceived health, young adults ranged from 3 – 10 (M = 7.97, SD
= 1.20) while older adults ranged from 4 – 10 (M = 8.47, SD = 1.23), with 10 representing the
highest level of health. Older adults also completed the attention network task (ANT) (Fan,
McCandiss, Fossela, Flombaum, & Posner, 2005), and they ranged from -163 to 71 on the
alerting effect (M = 1.29, SD = 42.43), from -89 to 116 on the orienting effect (M = 44.85, SD =
48.98) , and from -145 to 716 on the conflict effect (M = 167.55, SD = 122.15). However, for the
ANT 6 participants were excluded from this anaysis because of incomplete data.
3.2.3 Procedure
The procedure was also identical to the procedure reported in Experiment 1, however 80
experimental trials were completed instead of 40.
3.2.4 Data Analysis
The analyses performed are identical to those reported in Experiment 1, however the
emotion factor in the ANOVA model consisted of 3 levels (positive, negative, neutral) instead of
2.
3.3 Results and Discussion
The first analysis was a 2 × 3 × 2 repeated-measures ANOVA that compared salience
type and emotion type as within-subject factors and age group as a between-subject factor.
Salience type was significant, F(1,108) = 172.10, p < 0.001, η
p
2
= 0.614, indicating that fewer
low-salient letters (M = 0.292, SD = 0.143) were recalled compared high-salient letters (M =
0.573, SD = 0.164). However, no other effects reached significance. Next we collapsed positive
30
and negative trials and ran the same ANOVA, thus emotion type had only two levels in this
analysis. And again, no significant results were observed apart from salience type F(1,108) =
172.347, p < 0.001, η
p
2
= 0.615, showing that high-salience letters (M = 0.573, SD = 0.164) were
more often recalled compared to low-salience letters (M = 0.292, SD = 0.143).
Next we submitted high-salience recall and low-salience recall scores, as well as
difference scores (high-salience minus low-salience) to a HLM, using valence and arousal
ratings, the mean dB level of the sounds and age group as predictors. In Table 5A, Table 5B and
Table 5C we see that none of the predictors reached significance, apart from age predicting lower
recall of high-salience targets and low-salience targets. However, unlike Experiment 1 older age
did not predict smaller difference scores, suggesting that the relative attention bias produced by
differences in visual salience is similar in both age groups, despite older adults showing an
overall decrease in the number of high and low-salience targets reported.
We then analyzed the young adults and older adults separately in the same HLM but
omitted the age group predictor. In older adults no effects reached significance in predicting
difference scores, high-salience targets or low-salience targets. However, in young adults greater
arousal ratings predicted greater difference scores (Table 6A). In addition, greater arousal ratings
predicted greater recall of high-salience targets (Table 6B), but had no effect on recall of low-
salience targets (Table 6C). Moreover, greater mean dB levels predicted lower recall of low-
salience targets. No other predictors reached significance. Furthermore, physical intensity levels
of the sounds predicted lower recall of low-salience targets in young adults, suggesting that
louder audio clips inhibited attention to low-salience targets. In summary, the findings suggest
that in young adults greater arousal ratings on a particular trial led to greater recall of high-
31
salience targets on that trial, indicating that emotional arousal, regardless of valence, leads to
increases in attention to perceptually salient targets, thus supporting the ABC hypothesis.
However, no such effects were observed in older adults. When positive and negative
arousing trials were intermixed within the same experimental procedure, unlike the results
observed in Experiment 1, emotional arousal failed to increase subsequent attention biases in
older adults. Moreover, no significant effects in older adults were observed when negative and
neutral trials were analyzed separately, which given the findings of Experiment 1, should have
led to emotion-induced increases in the reporting of high-salience targets. Thus unlike the young
adult group that replicated the results of Experiment 1, which support the ABC hypothesis, older
adults failed to replicate the findings of Experiment 1.
We then determined whether arousal and valence ratings predicted the number of letters
falsely reported. In Table 7 we see that none of the predictors reached significance, however it
was of particular interest to examine whether arousal ratings or valence ratings predicted an
increase in the number of letters falsely reported when young adults were analyzed separately. In
Table 8 we see that none of the predictors reached significance, suggesting that higher recall of
high-salience targets on arousing trials was not driven by young adult participants simply
reporting more letters on those trials, regardless of their accuracy.
Taken together, we observed additional evidence for the ABC hypothesis, as positive and
negative arousal increased subsequent attention biases in young adults. Given the results of
Experiment 1, we expected older adults to show the same effects in Experiment 2. However, no
such effects were observed, suggesting that when positive and negative arousing trials are mixed
together in the same experiment, emotion-induced increases in attention biases do not occur in
older adults.
32
CHAPTER 4
4.1 Experiment 3
To examine the effects of emotional arousal on subsequent attention biases that are
determined by a combination of goal-relevance and perpetual salience, we used a procedure that
was similar to the procedure used in Experiment 1 and Experiment 2, however there were several
differences. Instead of using letters to measure attention, more complex visual images depicting
animals or inanimate objects were presented, and participants were directed to remember either
the animal or the object on each given trial. A free recall test of the to-be-remembered items was
given every ten trials as a manipulation check to ensure that participants were attending to the
goal-relevant items. The contrast level of the competing neutral items was also manipulated,
such that on half of the trials the goal-relevant item was also more perceptually salient than the
competing distracter item. And attention was measured via eye tracking, allowing a direct
measure of how attention was allocated to both competing items as they were displayed.
Moreover, at the end of the experiment participants were given a recognition memory test for all
of the animal and object images, as no studies have previously examined whether emotional
influences on subsequent attention biases to targets and distracters is associated with memory
biases for the targets and distracters. The more attention an item receives, the more likely it is to
be remembered (Logan, 2002), thus we expected emotion-induced changes in attention biases to
be reflected in stronger memory biases for goal-relevant items. Finally, rather than using
subjective ratings to measure different emotional responses, pupil sizes were recorded in
response to the emotionally arousing stimuli and neutral stimuli, which functioned as a
physiological measure of emotional arousal for each participant on a trial-by-trial basis.
33
4.2 Methods
4.2.1 Stimuli
A total of 120 IAPS images were used to manipulate emotional arousal. Forty of the
images were emotionally neutral, while the remaining eighty images were split between high-
arousal negative images and high-arousal positive images (Table 9). The neutral items used to
measure attention and memory consisted of three hundred and sixty individual objects presented
with no background. Half of the items consisted of animals, like a deer and a cat, while the other
half consisted of common objects, such as a hat and a spoon.
4.2.2 Participants
Thirty-six young adults (Females = 28) ranging from ages 19 to 33 (M = 23.6, SD = 4.0)
participated, as well as 36 older adults (Females = 26) ranging from age 65 to 90 (M = 74.1, SD
= 6.4). Participants were recruited from the University of Toronto and from the local community.
Young adults scored from 22 to 30 (M = 27.64, SD = 2.23) on the Montreal Cognitive Assesment
(MoCA), while older adults scored from 20 to 30 (M = 26.25, SD = 2.24). And young adults
ranged from 0.75 to 34.5 (M = 15.42, SD = 8.02) on the Extended Range Vocabulary Test
(ERVT) and older adults ranged from 1.75 to 48 (M = 28.60, SD = 10.81).
4.2.3 Procedure
Eye movements and pupil sizes were recorded throughout the experiment. An Eyelink 2
system (SR Research Ltd ) recorded eye movements at 500 hz with a spatial resolution of 0.5°. A
chin rest stabalized the head of each participant to ensure that eye movements were recorded
with minimal interference. Calibration was performed using a 9-point scale at the beginning of
both the encoding and recognition memory phases.
34
The encoding phase consisted of 120 experimental trials. Each trial began with the
presentation of a fixation cross for 4 seconds, followed by a 1-second presentation of an IAPS
image (Figure 2). After the IAPS presentation and a 1-second inter-stimulus-interval (ISI), an
object item and an animal item were simultaneously presented for 2 seconds. The ‘encoding’
phase of the experiment was divided into 2 separate blocks. In each block participants were
directed to remember either the animal or the object item (order counter-balanced across
participants). To ensure that participants prioritized remembering the goal-relevant target, a free
recall test was given every 10 trials for the target item. After the encoding phase, the recognition
memory phase of the experiment began. This part of the experiment was self-paced. The two-
hundred and forty previoulsy displayed object and animal items were presented individually,
along with 240 lure items (120 animals, 120 objects).
4.2.4 Data Analysis
Fixation count was used to measure attention biases towards the goal-relevant items.
Difference scores were calculated on a trial-by-trial basis to reflect a top-down bias. The total
number of fixations towards the competing item was subtracted from the total number of
fixations towards the goal-relevant item, providing a single score on each trial for fixation count,
which reflected the bias to attend to the goal-relevant item.
Similarly, recognition memory was calculated by creating difference scores for all 120
trials. On each trial the hit/miss score for the distracter item was subtracted from the hit/miss
score of the goal-relevant item. Thus if both items were missed or correctly recognized, a zero
bias score was assigned to that trial, while if the goal-relevant item was recognized and the
distracter item was missed, a bias score of 1 was assigned. Likewise, if the distracter item was
recognized and the goal-relevant item was missed, a score of -1 was assigned to that trial.
35
Arousal responses to the IAPS stimuli were measured by calculating the mean pupil size
during the ISI of each trial. Measuring pupil size during the ISI rather than during the
presentation of the IAPS images provides a better measure of pupil size changes that occur in
response to the emotional aspect of the image, as differences exist between the physical intensity
(i.e. luminance and contrast) of each image, which can influence pupil size. Nevertheless, to
address these differences the contrast level within each image as well as the overall brightness of
each image was modeled on a trial-by-trial basis in the HLM to hold constant the variance in
pupil size change that was associated with the physical properties of the IAPS stimuli.
4.3 Results and Discussion
First we examined attention biases to targets by comparing the number of times
participants fixated on the target compared to the distracter item. As expected, the number of
fixations towards the targets (M = 4.49, SE = 0.08) was greater compared to the number of
fixations made towards the distracters (M = 1.34, SE = 0.05), F(1,70) = 1376.03, p < 0.001, η
p
2
=
0.95, and age group did not interact with this main effect, F(1,70) = 0.752, p = 0.389, η
p
2
= 0.01.
This indicates that in both age groups attention biases towards the goal-relevant items compared
to the competing items were observed. Then we examined whether the salience of the items
influenced attention. As expected the number of fixations made towards full contrast items (M =
3.02, SD = 0.49) was greater than the number of fixations made towards low contrast items (M =
2.81, SD = 0.45), F(1,70) = 53.72, p < 0.001, η
p
2
= 0.434, and this effect did not interact with age
group, F(1,70) = 0.061, p = 0.805, η
p
2
= 0.001. Thus high-salience items attracted more attention
than low-salience items, and age group did not influence this effect. Therefore, both the bottom-
up and top-down attention manipulations worked, as more fixations were made towards target
items and items presented in full contrast.
36
The next step was to examine whether attention biases to the goal-relevant stimulus was
accompanied by greater recognition memory for goal-relevant stimulus. The fixation count on
each trial was submitted to a HLM to predict recognition memory biases for goal-relevant targets,
compared to competing targets, on a trial-by-trial basis. In Table 10 we see that fixation count
biases towards goal-relevant items (relative to distracters) predicted greater recognition memory
biases for those items, indicating that top-down attention led to better memory for goal-relevant
items, while controlling for differences in perceptual salience. In addition, age did not predict
memory biases, and age also did not interact with fixation count biases to predict memory biases
for goal-relevant items, suggesting that the effects of cognitive aging have no influence on
attention biases leading to memory biases for goal-relevant stimuli. Moreover, perceptual
salience had no influence on subsequent memory biases. Taken together, the results in Table 10
support the common notion that increases in attention during encoding leads to better recognition
memory (Logan, 2002), and that this is the case for both young and older adults.
Next we examined whether physiological arousal influenced attention towards the goal-
relevant targets. Mean pupil sizes were submitted to a HLM to predict fixation count biases
towards the goal-relevant items, relative to the competing items (i.e. difference scores). In Table
11 we see that greater mean pupil sizes predicted weaker attention biases towards goal-relevant
items, and that mean pupil size failed to interact with age group, suggesting that physiological
arousal interfered with top-down attention similarly in both age groups. Interestingly, the
emotion type of the trial failed to predict attention biases. Moreover, given that pupil size
predicted fixation count biases while holding the variance in contrast and brightness of images
constant, the observed pupil size changes that predicted attention biases can be attributed to
differences in emotional arousal, rather than the physical properties of the images. Furthermore,
37
there was no influence of age on attention biases, nor did age interact with pupil size to predict
top-down selective attention. Yet the salience predictor was significant indicating that fixation
biases towards goal-relevant items were greater when they had relatively high perceptual
salience compared to the distracter image. But in a separate model it was observed that salience
type did not interact with mean pupil sizes to predict fixation count biases, thus decreases in
attention to goal-relevant targets was not modulated by differences in perceptual salience. Taken
together, the results in Table 11 indicate that the more arousal that was elicited in response to the
IAPS images, the fewer fixations that were made to the goal-relevant stimulus, indicating that
emotional arousal interferes with top-down selective attention, regardless of the perceptual
salience of the goal-relevant stimulus and regardless of age. Therefore our findings suggest that
emotional arousal interferes with subsequent top-down attention biases, regardless of the
perceptual salience of the goal-relevant stimulus.
We then examined whether mean pupil sizes predicted biases in recognition memory for
goal-relevant items, compared to distracter items. Given that greater attention biases led to
greater recognition memory biases in general (Table 10), and given that greater arousal responses
led to weaker attention biases to goal-relevant items (Table 11), we expected that greater
emotional arousal would also lead to weaker memory biases for goal-relevant items, which is
what was observed. The results in Table 12 show that greater mean pupil sizes predicted weaker
recognition memory biases for goal-relevant items. And similar to the results observed in Table
11, the emotion type of the trial did not predict memory biases for goal-relevant items, and
likewise, age group had no influence on memory biases and age group did not interact with pupil
size. Thus physiological arousal responses predicted weaker top-down recognition memory
biases, in addition to weaker attention biases for goal-relevant targets, and this was the case for
38
both young adults and older adults. As for the salience type of the goal-relevant item, although
greater fixations were made to the goal-relevant items when they were salient, memory biases
were not greater for goal-relevant items when they were more salient, compared to when they
were less salient. Thus although top-down attention biases were greater towards goal-relevant
items when they were salient compared to when they were not (Table 11), the salience type of
the goal-relevant item had no influence on memory biases for the goal-relevant items (Table 12).
What these findings indicate is that physiological arousal suppresses subsequent top-down
attention biases as measured by the number of fixations made to each image, which carries over
into memory, for goal-relevant targets were less often remembered on trials where more
physiological arousal was elicited (Table 12).
Finally, we compared mean pupil sizes across the emotion type of the trials (Figure 3) to
verify that negative arousing and positive arousing stimuli led to greater physiological arousal
responses. For young adults pupil sizes were greatest on positive trials compared to neutral and
negative trials. There was no difference in the amount of physiological arousal elicited from
negative and neutral images. As for older adults, mean pupil sizes were also greatest on positive
trials, however, mean pupil sizes were also lowest on negative trials. Thus in older adults,
positive images produced more arousal and negative images produced less arousal, compared to
non-arousing neutral control images, which is indicative of an increased emphasis on emotion
regulation goals that is associated with older age (Mather & Carstensen, 2005). However, the
result that for young adults exposure to negative arousing images did not lead to greater
physiological responses than neutral images was unexpected.
In summary, the results of Experiment 3 indicate that there are no age differences in how
emotional arousal influences subsequent attention and memory biases. Physiological arousal
39
responses to IAPS images led to weaker subsequent attention biases towards emotionally neutral
items that were goal-relevant, as well as weaker memory biases for goal-relevant items.
Moreover, both of these effects were observed regardless of the perceptual salience of the goal-
relevant target. Thus, instead of increasing subsequent attention biases that are driven by
differences in perceptual salience, emotional arousal has the opposite effect on subsequent
attention biases that are driven by differences in goal-relevance, regardless of the perceptual
salience of the goal-relevant target. This indicates that when both bottom-up attention and top-
down attention are orthogonally manipulated, prior exposure to an emotionally arousing stimulus
weakens top-down attention biases but has no influence on bottom-up attention.
4.4 Results Summary
Across three experiments we addressed several related questions concerning how
attention biases are modulated as a result of being exposed to an emotionally arousing stimulus,
and whether cognitive aging impacts this influence. In Experiment 1 and Experiment 2 the main
goal was to test hypotheses derived from ABC theory, and to determine if the ABC hypothesis
explains emotional influences on attention biases that are driven by differences in perceptual
salience in both young and older adults. The goal of Experiment 3 was to examine the effects of
emotional arousal on subsequent attention biases that were driven by a combination of goal-
relevance and perceptual salience, and to examine whether these effects differed by age group.
In Experiment 1 it was shown that when goal-relevance was held constant among targets
competing for attention, and attention biases were driven by differences in perceptual salience,
exposure to stimuli that were rated as more arousing increased subsequent attention biases to
high-salient targets. The novel finding in this study was that the effect of arousal was not
40
modulated by the age of participants, indicating that emotion-induced increases in bottom-up
attention biases do not decline in older age.
In Experiment 2 we sought to extend the findings of Experiment 1 by including
emotionally arousing stimuli that were positive in emotional valence. According to ABC theory,
the valence of the emotionally arousing stimulus should not modulate the effects of arousal on
subsequent attention biases. Therefore it was predicted that the results of Experiment 2 would
replicate the results of Experiment 1. The findings revealed that the results did replicate for the
young adult age group, however no effects were observed for the older adult age group. Thus the
findings of Experiment 2 indicated that both positive and negative arousing stimuli enhance
subsequent attention biases in young adults, but not in older adults. It is unclear why the older
adult group is Experiment 2 failed to replicate the results observed with older adult group in
Experiment 1.
The purpose of Experiment 3 was to manipulate attention via goal-relevance and
perceptual salience. The findings revealed that exposure to emotionally arousing stimuli led to
decreases in attention to subsequently presented goal-relevant targets, regardless of the
perceptual salience of the goal-relevant target. Moreover, these emotion-induced decreases in
attention carried over into recognition memory, as greater physiological arousal responses to the
emotional stimuli also predicted weaker memory biases for goal-relevant targets, compared to
distracters.
In summary, the findings of all three experiments reveal that negative arousal increases
subsequent attention biases that are driven by perceptual salience, and that this effect is not
influenced by cognitive aging. However, when negative and positive stimuli are used together to
elicit emotional arousal, increases in attention to salient stimuli are increased in young adults, but
41
cognitive aging modulates this effect, as no effects were observed with older adults. Thus
positive arousal and negative arousal enhance subsequent attention biases when the relative
dominance among competing stimuli is high, but for older adults this effect was only observed
with negative arousing stimuli. Furthermore, when attention is driven by goal-relevance and
perceptual salience, emotional arousal inhibits subsequent attention biases, and this effect is
preserved in older adults. Moreover, this effect carries over into memory, as the bias to recognize
goal-relevant items compared to distracters was smaller on trials where participants produced
greater physiological arousal responses before the competing stimuli were encoded.
CHAPTER 5
5.1 General Discussion
Attention is a key component of human cognition, and its purpose is to filter out
irrelevant distractions so that perceptual processing is directed to what is most relevant. Two
primary forms of perceptual biases underlie selective attention—bottom-up processing and top-
down processing (Corbetta & Shulman, 2002; Desimone & Duncan, 1995; Fecteau & Munoz,
2006; Kastner & Ungerleider, 2001). But another important influence is emotional processing, as
perception is highly sensitive to emotionally arousing stimuli that are motivationally relevant and
related to threat and reward (Vuilleumier, 2005; Pessoa, 2009). A wide pattern of findings
related to cognitive aging indicates that emotional influences on attention remain intact in
healthy older adults (Hahn, et al., 2006; Mather & Knight, 2006; Ruffman, et al., 2009).
However, there is evidence that in young adults exposure to negative arousing stimuli can
enhance subsequent attention biases (Bocanegra & Zeelenberg, 2009a; Ciesielski, et al., 2010;
Padmala & Pessoa, 2008; Phelps, et al., 2006; Zeelenberg & Bocanegra, 2010), particularly
when the bias arises from differences in perceptual salience (Lee, et al., 2012; Sutherland &
42
Mather, 2012). But it is unclear how bottom-up and top-down influences interact when being
modulated by emotion. Moreover, no studies have examined whether emotional influences on
subsequent attention biases are the same in young and older adults.
Bottom-up attention refers to attention biases that arise from differences in perceptual
salience (Itti & Koch, 2000; Nothdurft, 2000; Proulx & Egeth, 2008), as when certain objects
have perceptual features that highly contrast with surrounding features. This form of attention
requires little effort on the part of the viewer, as the source of the bias arises externally. On the
other hand, top-down attention arises from internal cognitive processes, thus more cognitive
effort is required to maintain a target stimulus in working memory during visual search (Corbetta
& Shulman, 2002; Kastner & Ungerleider, 2001; Pessoa, Kastner, & Ungerleider, 2003). These
two influences on attention are quite different, and there is evidence that they rely on different
neural mechanisms, with bottom-up attention arising from activity in visual cortex while top-
down biases arise from frontal and parietal regions known as the fronto-parietal attention
network (Buschman & Miller, 2007).
Perceptual processing is also tuned to visual objects or features that signify threat or
reward, thus in addition to bottom-up and top-down processing there is emotional processing,
which is referred to as emotional attention (Pourtois, Schettino, & Vuilleumier, 2013;
Vuilleumier, 2005). Emotional stimuli are relevant to motivational drives that ensure survival,
and attention is reflexively biased towards environmental stimuli that may in some way lead to
reward, harm or death (see Bradley, 2009). Thus even in the presence of visually salient stimulus
(Niu, et al., 2012), or when searching for a goal-relevant stimulus (Hodsoll, et al., 2011; Muller,
et al., 2008), attention can be immediately directed to an emotionally arousing stimulus once it is
perceived, which then becomes the focus of attention to ensure an appropriate behavioral
43
response. On the other hand, there is evidence that emotional attention influences visual
processing after the emotional stimulus is removed and no longer competing for attention.
Exposure to an emotional stimulus produces an arousal response in the brain and body
that influences cognition (Bradley, et al., 2001), and this influence lasts for a short while after the
emotion-eliciting stimulus has been removed. The three experiments reported here examined
how selective attention is influenced by prior exposure to an emotionally arousing stimulus, and
whether cognitive aging impacts this influence. A number of studies were described showing that
when an emotional stimulus is briefly presented and followed by neutral targets, with a large
enough interval in time, the subsequent neutral targets are more quickly and accurately identified
than if a neutral stimulus was presented beforehand (Becker, 2009; Bocanegra & Zeelenberg,
2009a; Ciesielski, et al., 2010; Lee, et al., 2012; Olatunji, Ciesielski, Armstrong, & Zald, 2011;
Padmala & Pessoa, 2008; Phelps, et al., 2006; Quinlan & Johnson, 2011; Zeelenberg &
Bocanegra, 2010). These findings suggest that attention biases that are usually observed for
emotionally arousing stimuli carry over and enhance subsequent visual processing, particularly
for targets that are perceptually salient (Lee, et al., 2012; Sutherland & Mather, 2012). However,
no previous studies have directly examined the influence of emotional arousal on bottom-up
attention processing in older adults.
According to ABC theory (Mather & Sutherland, 2011), when emotional arousal is
elicited and the emotional stimulus is no longer competing for attention, for a short while
subsequent attention biases increase, particularly when they are driven by differences in
perceptual salience. Several experiments provide direct evidence that bottom-up attention biases
become amplified for a short duration after one is exposed to an emotional stimulus (Lee, et al.,
2012; Sutherland & Mather, 2012). But it is unclear whether top-down attention biases are
44
influenced by emotional arousal in the same way, and whether the influence of top-down
attention depends on the perceptual salience of the goal-relevant target. Moreover, no studies
have examined whether cognitive aging influences the way emotional arousal modulates
subsequent attention biases when top-down and bottom-up influences work together to bias
attention.
In the studies that have examined emotional influences on subsequent visual processing
in young adults, the target stimulus could be described as being both goal-relevant and
perceptually salient. When a subsequent neutral stimulus is presented alone (Bocanegra &
Zeelenberg, 2009a; Padmala & Pessoa, 2008; Zeelenberg & Bocanegra, 2010), it is salient
because it is not competing spatially with other targets in view as it occupies empty space.
Moreover, when an oddball target is presented among a group of homogenous distracters (Lee, et
al., 2012; Phelps, et al., 2006), the oddball contrasts perceptually with the other stimuli in view,
which makes it appear salient. Thus in the previous studies demonstrating emotion-induced
increases in subsequent visual processing, the target not only differed from the distracters in
terms of goal-relevance, but they were also more perceptually salient. Therefore it is unclear
whether emotional arousal enhances subsequent attention biases that are driven by goal-
relevance. However, there is some indication that emotional arousal may enhance the effects of
subsequent goal-relevant attention biases, such as when a target stimulus is presented among a
heterogeneous group of distracters, where visual ‘pop out’ effects are less likely to occur.
Presenting a fearful facial expression has been show to enhance subsequent visual search for
images presented among individually unique distracters (Becker, 2009; Quinlan & Johnson,
2011). However, this effect was only observed when emotional and neutral trials were blocked,
meaning a longer lasting mood was most likely induced throughout the emotional blocks, which
45
led to the increases in visual search (Quinlan & Johnson, 2011). Moreover, the findings were
only demonstrated with faces expressing fear—other emotional expressions like happiness,
disgust and anger did not influence subsequent attention.
On the other hand, there is also evidence suggesting that rather than increasing
subsequent attention biases that are driven by goal-relevance, emotional arousal may actually
interfere with subsequent top-down attention. It has been shown that the same neural structures
that are involved in top-down selective attention are also used to process emotionally arousing
stimuli (Pessoa, 2009). Moreover, prior exposure to an emotionally arousing stimulus interferes
with subsequent inhibitory responses that rely on cognitive control (Verbruggen & De Houwer,
2007). The ‘dual competition’ framework predicts that because highly arousing emotional
stimuli consume from the same pool of resources as cognitive control (Pessoa, 2009), prior
exposure to an emotionally arousing stimulus may interfere with subsequent attention biases that
are top-down, as top-down attention relies on cognitive control processes whereas bottom-up
attention does not.
Thus we had two separate hypotheses regarding the effects of emotional arousal on
subsequent top-down and bottom-up attention. The ABC hypothesis predicts that emotional
arousal of either valence amplifies subsequent attention when attention is bottom-up (Mather &
Sutherland, 2011). On the other hand the dual competition hypothesis suggests that exposure to
emotionally arousing stimuli of either valence will interfere with subsequent attention when
attention is top-down (Pessoa, 2009). However, in most situations attention is driven by a
combination of goal-relevance and perceptual salience, thus it was also of interest to examine
how emotional arousal influenced subsequent attention when both bottom-up and top-down
influences were simultaneously manipulated.
46
As for the effects of cognitive aging, very few studies have examined emotional
influences on subsequent attention biases in older adults. Despite declines in selective attention
and the speed at which sensory information is processed (Li, Lindenberger, & Sikström, 2001;
Madden, 2007), emotional influences on attention are in many instances spared (Hahn, et al.,
2006; Mather & Knight, 2006; Ruffman, et al., 2009), yet nevertheless modulated by an
increased emphasis on emotion regulation (Isaacowitz, et al., 2006a; Knight, et al., 2007; Mather
& Carstensen, 2003). Research on cognitive aging and emotional attention has revealed that
negative stimuli are actively ignored and that positive stimuli attract increased attention in older
adults compared to young adults—the so-called positivity effect in aging (Mather & Carstensen,
2005). But other studies have revealed that under certain conditions older adults and young
adults show the same increases in attention to negative stimuli (Hahn, et al., 2006; Mather &
Knight, 2006; Ruffman, et al., 2009), indicating that positivity effects are not always observed in
older adults, particularly when the procedures measure immediate attention biases, as in visual
search. Whether or not positivity effects are observed in a selective attention task appears to
depend upon whether cognitive control processes can be used on the task to control attention to
emotional stimuli (Brassen, et al., 2011; Knight, et al., 2007). Therefore, cognitive aging has an
effect on emotional attention that leads to increases in attention to positive stimuli and decreases
in attention to negative stimuli, but only when cognitive control processes are available.
Otherwise, older adults and young adults appear to show the same effects on measures of
emotional attention (Hahn, et al., 2006; Mather & Knight, 2006; Ruffman, et al., 2009).
However, cognitive aging and emotion-induced changes in subsequent attention is a
relatively unexplored topic, for the research aimed at examining emotional attention and
cognitive aging focuses on examining how emotional stimuli are processed differently from
47
neutral stimuli, and how this changes with age and is related to emotion regulation goals. No
research in the domain of cognitive aging has examined how exposure to an emotionally
arousing stimulus influences attention processes in general, or in other words how attention is
influenced when an emotional response has been elicited but the participant is no longer being
exposed to the emotionally arousing stimulus. Only one study has examined whether exposure to
emotional stimuli enhances or impairs subsequent visual processing in older adults (Mickley-
Steinmetz, et al., 2010), however in this study the emotional stimuli and neutral targets were
matched on levels of arousal, making it difficult to interpret whether emotional arousal has the
same effect. But nevertheless this study indicates that emotional arousal may enhance subsequent
visual processing, as has been observed in young adults (Bocanegra & Zeelenberg, 2009a;
Ciesielski, et al., 2010; Padmala & Pessoa, 2008; Zeelenberg & Bocanegra, 2010). Moreover,
given that emotional attention biases are preserved in older age (Hahn, et al., 2006; Mather &
Knight, 2006; Ruffman, et al., 2009), there is reason to suspect that emotional influences will
have the same effect on subsequent attention biases in both age groups. However, this may only
be observed when the attention bias is determined by differences in perceptual salience, for there
is evidence that older adults use cognitive control processes to up-regulate or down-regulate
attention biases to emotional stimuli (Brassen, et al., 2011; Knight, et al., 2007), which may
interfere with attention biases in older adults to a greater extent than in young adults when
attention is top-down. Therefore, the effects of emotionally arousing stimuli on subsequent visual
processing were examined across all three experiments, and both older and young adults
participated to examine the effects of cognitive aging.
In Experiment 1 the effects of negative arousal on subsequent bottom-up attention biases
was examined in older adults. In young adults exposure to a negative arousing stimulus led to
48
increases in biases to high-salient targets. Given that older adults show initial biases in attention
towards negative stimuli, it was expected that they would also demonstrate enhanced attention
biases for salient stimuli due to prior exposure to a negative arousing stimulus. And although
older adults are known to use cognitive control to inhibit attention away negative stimuli, we
suspected that this would not influence how the emotional response to the stimulus influenced
subsequent attention, for bottom-up attention biases do not rely on cognitive control. And as
expected, older adults performed better in identifying high-salient targets on trials where they
were exposed to sounds that were rated to be more emotionally arousing.
However, in Experiment 1 these effects were only observed when subjective ratings of
the sounds were used to predict performance, as direct comparisons in performance that were
based on the predefined emotional categories of the IADS sounds revealed no evidence that
negative arousal amplified bottom-up attention biases in older adults. This is likely due to the
fact that these predefined emotional categories are derived from normalized emotional ratings
obtained from young adults. Moreover, it has been shown that young adults and older adults
respond differently to emotional stimuli, both subjectively and physiologically (Grühn & Smith,
2008; Keil & Freund, 2009; Levenson, et al., 1991). Therefore, older adult participants provided
subjective arousal ratings to predict their performance on a trial-by-trial basis, as these results
best reflect how negative arousal influences subsequent attention biases in older adults. So to
summarize, the results of Experiment 1 indicate that prior exposure to an emotionally arousing
stimulus enhances subsequent attention biases that are bottom-up, and this effect occurs in both
older and young adults. Thus the predictions of ABC theory appear to explain subsequent
attention biases in older adults as well as biases in young adults when subsequent attention biases
are driven by perceptual salience.
49
The goal of Experiment 2 was to replicate and extend the findings of Experiment 1 by
exposing participants to both positive arousing and negative arousing sounds to elicit arousal.
Based on the predictions of ABC theory we expected positive arousal to have the same effect as
negative arousal on subsequent attention biases. And as for the effects of cognitive aging, we
expected to observe the same effects as those observed in Experiment 1. Although older adults
have been shown to use cognitive control to increase attention to positive stimuli, similar to the
way cognitive control is used to inhibit attention to negative stimuli, because bottom-up attention
does not rely of cognitive control, we expected the findings of Experiment 2 to replicate the
findings of Experiment 1.
However the results of Experiment 2 only replicated for the young adults, as greater
arousal ratings in this age group predicted increases in attention to high-salient targets. To further
explore the null effects in older adults, negative arousing and neutral trials based on the
predefined negative and neutral categories were separated from the positive trials and submitted
to the same analysis. Still, no effects were observed, suggesting that when negative and positive
arousing stimuli were intermixed to elicit arousal, no influence on subsequent attention biases are
observed in older adults. In addition, median splits were conducted on subjective stress and
affect levels reported by older adults prior to completing the main attention task. Those
measuring on the high end and low end of the PANAS and CESD measures were submitted to
separate analyses, however, these separate analyses all revealed null effects. Moreover, the same
procedure was performed with the ANT task, and likewise, no significant results were observed
in those scoring on the high and low ends of the ANT. Thus it is unclear why the older adults in
Experiment 2 did not replicate the findings of older adults in Experiment 1. Nevertheless, the
results of Experiment 2 indicate that positive arousal has the same effect as negative arousal on
50
subsequent attention biases in young adults, and corroborates the predictions derived from ABC
theory.
The goal of Experiment 3 was to examine whether top-down and bottom-up attention
interact in being influenced by prior exposure to positive or negative arousing stimuli, and
whether these effects are similar in older and young adults. A number of studies show that
exposure to negative arousing stimuli increases attention to subsequent neutral targets when
individually presented (Becker, 2009; Bocanegra & Zeelenberg, 2009a; Ciesielski, et al., 2010;
Lee, et al., 2012; Olatunji, et al., 2011; Padmala & Pessoa, 2008; Phelps, et al., 2006; Quinlan &
Johnson, 2011; Zeelenberg & Bocanegra, 2010), and there is also some evidence that exposure to
negative arousing stimuli may increase subsequent visual search (Becker, 2009; Quinlan &
Johnson, 2011), which reflects a top-down attention bias. But there is also evidence that
emotional arousal may interfere with subsequent attention biases that are driven by goal-
relevance, for exposure to highly arousing emotional stimuli activates the same neural structures
that underlie cognitive control. Thus simply being exposed to an arousing stimulus may consume
enough resources related to cognitive control to interfere with subsequent top-down attention
biases. Moreover, it has been shown that attending to multiple goal-relevant items that are
equally salient is more difficult when a stressful state has been induced (Morelli & Burton, 2009).
Thus the goal of Experiment 3 was to examine whether attention biases to goal-relevant items is
enhanced or weakened due to prior exposure to an emotionally arousing stimulus. However,
since there is strong evidence that emotional arousal enhances the effects of subsequent attention
biases that bottom-up (Lee, et al., 2012; Sutherland & Mather, 2012), it was also of interest to
examine whether bottom-up and top-down attention biases interacted in being influenced by
prior exposure to an emotionally arousing stimulus. Moreover, given that there is little to no
51
evidence regarding the effects of cognitive aging on emotion-induced changes in subsequent
attention, as in Experiments 1 and 2, both young adults and older adults participated.
What was observed in Experiment 3 was that emotional arousal suppresses subsequent
attention biases that are driven by goal-relevance, and this occurs regardless of whether the
emotionally arousing stimulus is perceptually salient or has relatively low-salience. Moreover,
these effects were not influenced by the age group of the participants, suggesting that weaker
top-down attention biases due to prior exposure to an emotionally arousing stimulus is an
emotional influence on attention that is preserved in older age. And recognition memory for
goal-relevant items was similarly weakened by emotional arousal, suggesting that emotional
influences on attention are reflected in recognition memory biases. These findings are novel for
they demonstrate that top-down attention is weakened by emotional arousal, regardless of the
salience of the goal-relevant target. In addition, no other studies have measured physiological
arousal responses and used the responses to predict attention biases on a trial-by-trial basis.
The findings of Experiment 3 conflict with other findings showing that in young adults
exposure to fearful faces increases subsequent visual search for a single goal-relevant item
presented among distracters (Becker, 2009; Quinlan & Johnson, 2011). However, these findings
most likely reflect the influence of negative mood rather than the influence of negative arousal
per se, as emotion only influenced subsequent attention when emotional trials and neutral trials
were blocked. Thus it is possible that goal-relevant attention is enhanced by negative mood, but
that immediate negative arousal responses have the opposite effect and interfere with subsequent
attention biases that are driven by goal-relevance. These conflicting findings indicate that the
effects of mood and brief arousal responses have the opposite effect on top-down attention.
Moreover, the results of Experiment 2 also indicate that mood and immediate emotional arousal
52
responses have different effects on attention, for previous studies show that positive mood
broadens attention (Fredrickson & Branigan, 2005), resulting in greater distraction to irrelevant
information (Rowe, Hirsh, & Anderson, 2007). However, the results of Experiment 2 indicate
that in young adults positive and negative arousing stimuli have the same effect on selective
attention. Given these alternative findings between the effects of mood on attention and the
effects of immediate emotional responses on attention, direct comparisons between the effects of
positive and negative moods and positive and negative arousal responses on attention would be a
very intriguing question to address in future studies examining emotional influences on attention.
Another interesting finding in Experiment 3 is that emotion-induced decreases in
subsequent attention biases that are top-down are not modulated by the perceptual salience of the
goal-relevant target. This was contrary to expectations, as emotional arousal was expected to
weaken top-down attention biases, but it was expected that this effect might interact with the
salience of the goal-relevant stimulus. However, our results suggest that greater arousal led to
weaker top-down attention biases, and this effect was not influenced by perceptual salience.
Yet there were no age-differences observed in Experiment 3. Given that older adults
show age-related declines in cognitive control (Braver & Barch, 2002; Hedden & Gabrieli, 2004),
and because prior exposure to emotionally arousing stimuli were expected to consume additional
resources needed for subsequent top-down attention (Pessoa, 2009), emotion-induced decreases
in subsequent attention were expected in both age groups, but interference was expected to be
greater in older adults given the decline in cognitive control resources that are associated with
cognitive aging. However, this was not observed, as both age groups showed weaker top-down
attention biases on trials where physiological arousal responses were greater to the emotional
stimuli, which was not influenced by age group. This finding along with the results of
53
Experiment 1 indicate that emotional influences on subsequent attention biases is preserved in
older age, as are immediate biases in attention to emotion-eliciting stimuli (Hahn, et al., 2006;
Mather & Knight, 2006; Ruffman, et al., 2009). These results corroborate the hypothesis derived
from dual competition theory, which predicts that exposure to emotional arousing stimuli will
decrease attention biases when the bias is not directed to the emotion-eliciting stimulus (Pessoa,
2009).
In addition to measuring emotional influences on top-down and bottom-up attention
biases, in Experiment 3 we also examined the effects of memory biases towards stimuli that were
goal-relevant. No studies in the past have examined if emotional influences on subsequent
attention biases result in memory biases that reflect the initial bias in attention. It was observed
that weaker attention biases towards goal-relevant stimuli were reflected in the way that
participants remembered the items competing for attention, as weaker memory biases for goal-
relevant targets were predicted by arousal responses to the preceding emotional stimulus. Other
studies have shown that emotionally arousing stimuli impair memory for neutral stimuli that are
presented after an emotionally arousing stimulus, but only when the stimulus was irrelevant to
the task (Sakaki, et al., in press). When the subsequent neutral stimulus was goal-relevant,
memory for that stimulus was neither enhanced nor impaired. However, the findings of
Experiment 3 indicate that memory for goal-relevant items decrease due to prior exposure to an
emotionally arousing stimulus. But the procedure used by Sakaki and colleagues (in press) did
not directly measure attention to the subsequently presented neutral stimuli, nor did the stimuli
compete spatially with other stimuli during encoding. Thus there are a number of methodological
differences, however both studies indicate that memory for goal-relevant stimuli that follow an
emotionally arousing stimulus are not enhanced.
54
Taken together, two hypotheses derived from the emotion and cognition literature were
used to examine how exposure to an emotionally arousing stimulus influences the way
subsequently presented neutral targets are visually processed. ABC theory predicts that bottom-
up attention biases among competing neutral stimuli will be enhanced following exposure to
positive and negative arousing stimuli, and the results of Experiments 1 and 2 supported this
prediction, but only in young adults. As for older adults, the results were mixed and unclear. In
the first experiment older adults showed the same effects as young adults, as negative arousal
increased subsequent attention biases to perceptually salient targets. However, when positive
arousing trials were mixed with negative trials, no effects were observed in older adults, while
the performance of young adults replicated the results of Experiment 1. Thus the results of
Experiment 1 in the older adult age group should be interpreted with caution, as these results did
not replicate in Experiment 2. Moreover, there was no evidence that individual differences in
selective attention, stress, or positive and negative affect could explain the incongruent findings
in older adults in Experiment 2.
As for Experiment 3, older and young adults were similarly influenced by emotional
arousal, as trials where more physiological arousal was elicited led to weaker attention and
memory biases for goal-relevant targets, and age did not modulate this effect. Therefore the
findings support the dual competition hypothesis (Pessoa, 2009), which predicts that emotional
arousal will weaken subsequent attention biases due to fewer cognitive control resources being
available for top-down attention, as exposure to emotionally arousing stimuli are hypothesized to
consume more cognitive control resources compared to neutral stimuli, thus interfering with
subsequent top-down attention biases that rely on cognitive control.
55
5.2 Concluding Remarks
Here we report three studies that contribute novel findings to the emotion, attention and
cognitive aging literature. The first novel finding is that when multiple targets compete for
limited attention, previous exposure to a negative arousing stimulus increased the tendency to
focus on high-salient targets, which was observed in both older adults and young adults. No
studies in the cognitive aging literature have examined emotion-induced changes on subsequent
attention when attention is manipulated by differences in perceptual salience, or when
subsequent neutral stimuli were simultaneously presented (i.e. spatial competition). Therefore
the findings of Experiment 1, which reveal that negative arousal amplifies subsequent attention
biases in older adults that result from differences in perceptual salience add novel and
informative insight into the effects of cognitive aging on emotional attention.
However, the results of Experiment 2 in some ways challenge and in some ways bolster
and extend the findings of Experiment 1. The results of young adults replicate the findings of
Experiment 1, showing that higher arousal ratings of an IADS sound played on a particular trial
led to greater biases to recall perceptually salient targets on that trial. This suggests that positive
and negative arousal have the same effects on subsequent attention processes in young adults. On
the other hand, with older adults the results of Experiment 2 failed to replicate the results of
Experiment 1, indicating that when positive and negative stimuli are used to elicit arousal, no
influences on subsequent attention is observed. Moreover, no increases in subsequent attention
biases were observed when the positive trials were removed from the analysis, further suggesting
that for older adults the results of Experiment 2 did not replicate the results of Experiment 1.
Future studies examining emotion-induced changes on subsequent attention may confirm or
disconfirm the conflicting findings of Experiment 1 and Experiment 2, but these studies
56
nevertheless serve as a reference to future studies examining cognitive aging and the effects of
emotional arousal on subsequent attention biases. In addition, the findings of Experiment 1 and
Experiment 2 indicate that for young adults both positive and negative arousal lead to increases
in subsequent attention biases that are driven by differences in perceptual salience.
As for Experiment 3, several novel findings were also observed. The first was that
positive and negative arousal led to decreases in attention to goal-relevant targets, and this
occurred regardless of the target’s relative perceptual salience. And Experiment 3 also measured
recognition memory for subsequent neutral targets that competed for attention. No previous
studies have examined the effects of emotion-related changes in attention biases to subsequently
presented neutral stimuli, while also measuring recognition memory for those stimuli. It was
observed that the emotion-induced decreases in attention to goal-relevant stimuli were reflected
in memory, as physiological arousal responses predicted lower recognition memory biases for
goal-relevant items. Thus the effects of emotional arousal on subsequent attention biases have
enduring consequences by influencing memory for the subsequently presented items. Moreover,
Experiment 3 was the first study to use a physiological marker of emotional arousal, which was
used to predict performance, rather than assuming that normalized stimuli classified as being
emotional or neutral has the same impression on the individuals participating in the experiment.
In conclusion, emotion and attention researchers are beginning to ask larger questions
beyond simply when and how attention is biased to emotionally arousing stimuli, as it has been
revealed that emotions influence attention after the emotional stimulus is removed, suggesting
that the emotional response elicited in the brain and body has a more general effect on attention
and the encoding of new information. Future studies examining emotion, attention and cognitive
aging will hopefully work to elucidate these effects more specifically and continue to push the
57
boundaries of experimental procedures to reveal the specific effects of emotional attention, and
how it changes in older age.
58
References
Allard,
E.
S.,
&
Isaacowitz,
D.
M.
(2008).
Are
preferences
in
emotional
processing
affected
by
distraction?Examining
the
age-‐related
positivity
effect
in
visual
fixation
within
a
dual-‐task
paradigm.
Aging,
Neuropsychology,
and
Cognition,
15(6),
725-‐743.
doi:
10.1080/13825580802348562
Alpers,
G.
W.,
&
Gerdes,
A.
B.
M.
(2007).
Here
is
looking
at
you:
Emotional
faces
predominate
in
binocular
rivalry.
Emotion,
7(3),
495-‐506.
doi:
10.1037/1528-‐
3542.7.3.495
Alpers,
G.
W.,
&
Pauli,
P.
(2006).
Emotional
pictures
predominate
in
binocular
rivalry.
Cognition
&
Emotion,
20(5),
596-‐607.
doi:
10.1080/02699930500282249
Alpers,
G.
W.,
Ruhleder,
M.,
Walz,
N.,
Muhlberger,
A.,
&
Pauli,
P.
(2005).
Binocular
rivalry
between
emotional
and
neutral
stimuli:
A
validation
using
fear
conditioning
and
EEG.
International
Journal
of
Psychophysiology,
57(1),
25-‐32.
doi:
10.1016/j.ijpsycho.2005.01.008
Anderson,
A.
K.
(2005).
Affective
influences
on
the
attentional
dynamics
supporting
awareness.
Journal
of
Experimental
Psychology-‐General,
134(2),
258-‐281.
doi:
10.1037/0096-‐3445.134.2.258
Anderson,
A.
K.,
&
Phelps,
E.
A.
(2001).
Lesions
of
the
human
amygdala
impair
enhanced
perception
of
emotionally
salient
events.
Nature,
411(6835),
305-‐309.
doi:
10.1038/35077083
Armony,
J.
L.,
&
Dolan,
R.
J.
(2002).
Modulation
of
spatial
attention
by
fear-‐conditioned
stimuli:
an
event-‐related
fMRI
study.
Neuropsychologia,
40(7),
817-‐826.
doi:
10.1016S0028-‐3932(01)00178-‐6
Bannerman,
R.
L.,
Milders,
M.,
De
Gelder,
B.,
&
Sahraie,
A.
(2008).
Influence
of
emotional
facial
expressions
on
binocular
rivalry.
Ophthalmic
and
Physiological
Optics,
28(4),
317-‐326.
doi:
10.1111/j.1475-‐1313.2008.00568.x
Bannerman,
R.
L.,
Regener,
P.,
&
Sahraie,
A.
(2011).
Binocular
rivalry:
a
window
into
emotional
processing
in
aging.
Psychology
and
Aging,
26(2),
372-‐380.
doi:
10.1037/a0022029
Beck,
D.
M.,
&
Kastner,
S.
(2005).
Stimulus
context
modulates
competition
in
human
extrastriate
cortex.
Nature
Neuroscience,
8(8),
1110-‐1116.
doi:
10.1038/nn1501
Beck,
D.
M.,
&
Kastner,
S.
(2009).
Top-‐down
and
bottom-‐up
mechanisms
in
biasing
competition
in
the
human
brain.
Vision
Research,
49(10),
1154-‐1165.
doi:
10.1016/j.visres.2008.07.012
59
Becker,
M.
W.
(2009).
Panic
search:
Fear
produces
efficient
visual
search
for
nonthreatening
objects.
Psychological
Science,
20(4),
435-‐437.
doi:
10.1111/j.1467-‐
9280.2009.02303.x
Blake,
R.,
&
Logothetis,
N.
K.
(2002).
Visual
competition.
Nature
Reviews
Neuroscience,
3(1),
13-‐23.
doi:
10.1038/nrn701
Bocanegra,
B.
R.,
&
Zeelenberg,
R.
(2009a).
Dissociating
emotion-‐induced
blindness
and
hypervision.
Emotion,
9(6),
865-‐873.
doi:
10.1037/a0017749
Bocanegra,
B.
R.,
&
Zeelenberg,
R.
(2009b).
Emotion
Improves
and
Impairs
Early
Vision.
Psychological
Science,
20(6),
707-‐713.
doi:
10.1111/j.1467-‐9280.2009.02354.x
Bradley,
M.
M.
(2009).
Natural
selective
attention:
Orienting
and
emotion.
Psychophysiology,
46(1),
1-‐11.
doi:
10.1111/j.1469-‐8986.2008.00702.x
Bradley,
M.
M.,
Codispoti,
M.,
Cuthbert,
B.
N.,
&
Lang,
P.
J.
(2001).
Emotion
and
motivation
I:
Defensive
and
appetitive
reactions
in
picture
processing.
Emotion,
1(3),
276-‐298.
doi:
10.1037/1528-‐3542.1.3.276
Bradley,
M.
M.,
&
Lang,
P.
J.
(2000).
Affective
reactions
to
acoustic
stimuli.
Psychophysiology,
37(2),
204-‐215.
doi:
10.1111/1469-‐8986.3720204
Bradley,
M.
M.,
&
Lang,
P.
J.
(2007).
The
international
affective
digitized
sounds
(2nd
Edition;
IADS-‐2):
Affective
ratings
of
sounds
and
instruction
manual.
Gainesville,
FL:
University
of
Florida.
Brassen,
S.,
Gamer,
M.,
&
Buchel,
C.
(2011).
Anterior
cingulate
activation
Is
related
to
a
positivity
bias
and
emotional
stability
in
successful
aging.
Biological
Psychiatry,
70(2),
131-‐137.
doi:
10.1016/j.biopsych.2010.10.013
Braver,
T.
S.,
&
Barch,
D.
M.
(2002).
A
theory
of
cognitive
control,
aging
cognition,
and
neuromodulation.
Neuroscience
and
Biobehavioral
Reviews,
26(7),
809-‐817.
doi:
10.1016/S0149-‐7634(02)00067-‐2
Broadbent,
D.
E.
(1958).
The
selective
nature
of
learning
Perception
and
communication.
(pp.
244-‐267):
Elmsford,
NY,
US:
Pergamon
Press.
Brosch,
T.,
Sander,
D.,
Pourtois,
G.,
&
Scherer,
K.
R.
(2008).
Beyond
fear:
Rapid
spatial
orienting
toward
positive
emotional
stimuli.
Psychological
Science,
19(4),
362-‐370.
doi:
10.1111/j.1467-‐9280.2008.02094.x
Brown,
J.
A.,
Fishco,
V.
V.,
&
Hanna,
G.
(1993).
Nelson-‐Denny
Reading
Test:
Manual
for
Scoring
and
Interpretation
Forms
G
&
H.
Rolling
Meadows,
IL:
Riverside
Publishing.
60
Bundesen,
C.
(1990).
A
theory
of
visual
attention.
Psychological
Review,
97(4),
523-‐547.
doi:
10.1037/0033-‐295X.97.4.523
Buschman,
T.
J.,
&
Miller,
E.
K.
(2007).
Top-‐down
versus
bottom-‐up
control
of
attention
in
the
prefrontal
and
posterior
parietal
cortices.
Science,
315,
1860-‐1862.
doi:
10.1126/science.1138071
Calvo,
M.
G.,
&
Lang,
P.
J.
(2005).
Parafoveal
Semantic
Processing
of
Emotional
Visual
Scenes.
Journal
of
Experimental
Psychology:
Human
Perception
and
Performance,
31(3),
502-‐
519.
doi:
10.1037/0096-‐1523.31.3.502
Carlson,
J.
M.,
Reinke,
K.
S.,
&
Habib,
R.
(2009).
A
left
amygdala
mediated
network
for
rapid
orienting
to
masked
fearful
faces.
Neuropsychologia,
47(5),
1386-‐1389.
doi:
10.1016/j.neuropsychologia.2009.01.026
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
Chelazzi,
L.,
Duncan,
J.,
Miller,
E.
K.,
&
Desimone,
R.
(1998).
Responses
of
neurons
in
inferior
temporal
cortex
during
memory-‐guided
visual
search.
Journal
of
Neurophysiology,
80(6),
2918-‐2940.
Chelazzi,
L.,
Miller,
E.
K.,
Duncan,
J.,
&
Desimone,
R.
(2001).
Responses
of
neurons
in
macaque
area
V4
during
memory-‐guided
visual
search.
Cerebral
Cortex,
11(8),
761-‐
772.
doi:
10.1093/cercor/11.8.761
Chun,
M.
M.,
&
Potter,
M.
C.
(1995).
A
2-‐stage
model
for
multiple-‐target
detection
in
rapid
serial
visual
presentation.
Journal
of
Experimental
Psychology:
Human
Perception
and
Performance,
21(1),
109-‐127.
doi:
10.1037/0096-‐1523.21.1.109
Ciesielski,
B.
G.,
Armstrong,
T.,
Zald,
D.
H.,
&
Olatunji,
B.
O.
(2010).
Emotion
modulation
of
visual
attention:
Categorical
and
temporal
characteristics.
PLoS
ONE,
11(5),
e13860.
doi:
10.1371/journal.pone.0013860
Corbetta,
M.,
&
Shulman,
G.
L.
(2002).
Control
of
goal-‐directed
and
stimulus-‐driven
attention
in
the
brain.
Nature
Reviews
Neuroscience,
3(3),
201-‐215.
doi:
10.1038/nrn755
De
Martino,
B.,
Kalisch,
R.,
Rees,
G.,
&
Dolan,
R.
J.
(2009).
Enhanced
processing
of
threat
stimuli
under
limited
attentional
resources.
Cerebral
Cortex,
19(1),
127-‐133.
doi:
10.1093/cercor/bhn062
Desimone,
R.,
&
Duncan,
J.
(1995).
Neural
mechanisms
of
selective
visual-‐attention.
Annual
Review
of
Neuroscience,
18,
193-‐222.
61
Deutsch,
J.
A.,
&
Deutsch,
D.
(1963).
Attention:
Some
theoretical
considerations.
[doi:10.1037/h0039515].
Psychological
Review,
70(1),
80-‐90.
doi:
10.1037/h0039515
Dolan,
R.
J.
(2002).
Emotion,
cognition,
and
behavior.
Science,
298(5596),
1191-‐1194.
doi:
10.1126/science.1076358
Duchowski,
A.
(2007).
Eye
tracking
methodology:
Theory
and
practice
(Vol.
373):
Springer.
Fan,
J.,
McCandiss,
B.
D.,
Fossela,
J.,
Flombaum,
J.
I.,
&
Posner,
M.
I.
(2005).
The
activation
of
attentional
networks.
NeuroImage,
26(2),
471-‐479.
doi:
10.1016/j.neuroimage.2005.02.004
Fecteau,
J.
H.,
&
Munoz,
D.
P.
(2006).
Salience,
relevance,
and
firing:
A
priority
map
for
target
selection.
Trends
in
Cognitive
Sciences,
10(8),
382-‐390.
doi:
10.1016/j.tics.2006.06.011
Fjell,
A.
M.,
Westlye,
L.
T.,
Amlien,
I.,
Espeseth,
T.,
Reinvang,
I.,
Raz,
N.,
et
al.
(2009).
High
consistency
of
regional
cortical
thinning
in
aging
across
multiple
samples.
Cerebral
Cortex,
19(9),
2001-‐2012.
doi:
10.1093/cercor/bhn232
Fredrickson,
B.
L.,
&
Branigan,
C.
(2005).
Positive
emotions
broaden
the
scope
of
attention
and
thought-‐action
repertoires.
Cognition
&
Emotion,
19(3),
313-‐332.
doi:
10.1080/02699930441000238
Grühn,
D.,
&
Smith,
J.
(2008).
Characteristics
for
200
words
rated
by
young
and
older
adults:
Age-‐dependent
evaluations
of
German
adjectives
(AGE)
Behavior
Research
Methods,
40(4),
1088-‐1097.
doi:
10.3758/BRM.40.4.1088
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(3),
312-‐336.
doi:
10.1016/j.actpsy.2006.01.008
Hedden,
T.,
&
Gabrieli,
J.
D.
E.
(2004).
Insights
into
the
ageing
mind:
A
view
from
cognitive
neuroscience.
Nature
Reviews
Neuroscience,
5,
87-‐96.
doi:
10.1038/nrn1323
Hodsoll,
S.,
Viding,
E.,
&
Lavie,
N.
(2011).
Attentional
capture
by
irrelevant
emotional
distractor
faces.
Emotion,
11(2),
346-‐353.
Isaacowitz,
D.
M.,
Wadlinger,
H.
A.,
Goren,
D.,
&
Wilson,
H.
R.
(2006a).
Is
there
an
age-‐
related
positivity
effect
in
visual
attention?
A
comparison
of
two
methodologies.
Emotion,
6(3),
511-‐516.
doi:
10.1037/1528-‐3542.6.3.511
Isaacowitz,
D.
M.,
Wadlinger,
H.
A.,
Goren,
D.,
&
Wilson,
H.
R.
(2006b).
Selective
preference
in
visual
fixation
away
from
negative
images
in
old
age?
An
eye-‐tracking
study.
Psychology
and
Aging,
21(1),
40-‐48.
doi:
10.1037/0882-‐7974.21.1.40
62
Itti,
L.,
&
Koch,
C.
(2000).
A
saliency-‐based
search
mechanism
for
overt
and
covert
shifts
of
visual
attention.
Vision
Research,
40(10-‐12),
1489-‐1506.
doi:
10.1016/S0042-‐
6989(99)00163-‐7
Itti,
L.,
&
Koch,
C.
(2001).
Computational
modelling
of
visual
attention.
Nature
Review
Neuroscience,
2,
194-‐203.
doi:
10.1038/35058500
Kastner,
S.,
De
Weerd,
P.,
Desimone,
R.,
&
Ungerleider,
L.
C.
(1998).
Mechanisms
of
directed
attention
in
the
human
extrastriate
cortex
as
revealed
by
functional
MRI.
Science,
282(5386),
108-‐111.
doi:
10.1126/science.282.5386.108
Kastner,
S.,
&
Ungerleider,
L.
G.
(2001).
The
neural
basis
of
biased
competition
in
human
visual
cortex.
Neuropsychologia,
39(12),
1263-‐1276.
doi:
10.1016/S0028-‐
3932(01)00116-‐6
Keil,
A.,
&
Freund,
A.
M.
(2009).
Changes
in
the
sensitivity
to
appetitive
and
aversive
arousal
across
adulthood.
Psychology
and
Aging,
24(3),
668-‐680.
doi:
10.1037/a0016969
Keil,
A.,
&
Ihssen,
N.
(2004).
Identification
facilitation
for
emotionally
arousing
verbs
during
the
attentional
blink.
Emotion,
4(1),
23-‐35.
doi:
10.1037/1528-‐3542.4.1.23
Kensinger,
E.
A.
(2009).
Remembering
the
details:
Effects
of
emotion.
Emotion
Review,
1,
99-‐113.
doi:
10.1177/1754073908100432.
Kisley,
M.
A.,
Wood,
S.,
&
Burrows,
C.
L.
(2007).
Looking
at
the
sunny
side
of
life:
Age-‐related
change
in
an
event-‐related
potential
measure
of
the
negativity
bias.
Psychological
Science,
18(9),
838-‐843.
doi:
10.1111/j.1467-‐9280.2007.01988.x
Knight,
M.,
Seymour,
T.
L.,
Gaunt,
J.
T.,
Baker,
C.,
Nesmith,
K.,
&
Mather,
M.
(2007).
Aging
and
goal-‐directed
emotional
attention:
Distraction
reverses
emotional
biases.
Emotion,
7(4),
705-‐714.
doi:
10.1037/1528-‐3542.7.4.705
LaBar,
K.
S.,
Mesulam,
M.
M.,
Gitelman,
D.
R.,
&
Weintraub,
S.
(2000).
Emotional
curiosity:
modulation
of
visuospatial
attention
by
arousal
is
preserved
in
aging
and
early-‐
stage
Alzheimer's
disease.
Neuropsychologia,
38(13),
1734-‐1740.
doi:
10.1016/S0028-‐3932(00)00077-‐4
Leclerc,
C.
M.,
&
Kensinger,
E.
A.
(2008).
Effects
of
age
on
detection
of
emotional
information.
Psychology
and
Aging,
23(1),
209-‐215.
doi:
10.1037/0882-‐
7974.23.1.209
Lee,
T.-‐H.,
Baek,
J.,
Lu,
Z.-‐L.,
&
Mather,
M.
(in
press).
How
arousal
modulates
the
visual
contrast
sensitivity
function.
Emotion.
63
Lee,
T.-‐H.,
Itti,
L.,
&
Mather,
M.
(2012).
Evidence
for
arousal-‐biased
competition
in
perceptual
learning.
Frontiers
in
Emotion
Science,
3(241),
1-‐9.
doi:
10.3389/fpsyg.2012.00241
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
Li,
S.-‐C.,
Lindenberger,
U.,
&
Sikström,
S.
(2001).
Aging
cognition:
from
neuromodulation
to
representation.
Trends
in
Cognitive
Sciences,
5(11),
479-‐486.
doi:
10.1016/S1364-‐
6613(00)01769-‐1
Logan,
G.
D.
(2002).
An
instance
theory
of
attention
and
memory.
Psychological
Review,
109(2),
376-‐400.
doi:
10.1037//0033-‐295X.109.2.376
Madden,
D.
J.
(2007).
Aging
and
visual
attention.
Current
Directions
in
Psychological
Science,
16(2),
70-‐74.
doi:
10.1111/j.1467-‐8721.2007.00478.x
Mather,
M.
(2007).
Emotional
arousal
and
memory
binding:
An
object-‐based
framework.
Perspectives
on
Psychological
Science,
2(1),
33-‐52.
doi:
10.1111/j.1745-‐
6916.2007.00028.x
Mather,
M.,
&
Carstensen,
L.
L.
(2003).
Aging
and
attentional
biases
for
emotional
faces.
Psychological
Science,
15(5),
409-‐415.
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
Mather,
M.,
&
Knight,
M.
(2006).
Angry
faces
get
noticed
quickly:
threat
detection
is
not
impaired
among
older
adults.
Journal
of
Gerontology:
Psychological
Sciences,
61(1),
54-‐57.
Mather,
M.,
&
Sutherland,
M.
R.
(2011).
Arousal-‐biased
competition
in
perception
and
memory.
Perspectives
on
Psychological
Science,
6(2),
114-‐133.
doi:
10.1177/1745691611400234
Mickley-‐Steinmetz,
K.
R.,
Muscatell,
K.
A.,
&
Kensinger,
E.
A.
(2010).
The
effect
of
valence
on
young
and
older
adults’
attention
in
a
rapid
serial
visual
presentation
task.
Psychology
and
Aging,
25(1),
239-‐245.
doi:
10.1037/a0018297
Milders,
M.,
Sahraie,
A.,
Logan,
S.,
&
Donnellon,
N.
(2006).
Awareness
of
Faces
Is
Modulated
by
Their
Emotional
Meaning.
Emotion,
6(1),
10-‐17.
doi:
10.1037/1528-‐3542.6.1.10
64
Morelli,
F.,
&
Burton,
P.
A.
(2009).
The
impact
of
induced
stress
upon
selective
attention
in
multiple
object
tracking.
Military
Psychology,
21(1),
81-‐97.
doi:
10.1080/08995600802565769
Most,
S.
B.,
Chun,
M.
M.,
Johnson,
M.
R.,
&
Kiehl,
K.
A.
(2006).
Attentional
modulation
of
the
amygdala
varies
with
personality.
Neuroimage,
31(2),
934-‐944.
doi:
10.1016/j.neuroimage.2005.12.031
Most,
S.
B.,
Chun,
M.
M.,
Widders,
D.
M.,
&
Zald,
D.
H.
(2005).
Attentional
rubbernecking:
Cognitive
control
and
personality
in
emotion-‐induced
blindness.
Psychonomic
Bulletin
&
Review,
12(4),
654-‐661.
doi:
10.3758/BF03196754
Most,
S.
B.,
Smith,
S.
D.,
Cooter,
A.
B.,
Levy,
B.
N.,
&
Zald,
D.
H.
(2007).
The
naked
truth:
Positive,
arousing
distractors
impair
rapid
target
perception.
21(5),
964-‐981.
doi:
10.1080/02699930600959340
Muller,
M.
M.,
Andersen,
S.
K.,
&
Keil,
A.
(2008).
Time
course
of
competition
for
visual
processing
resources
between
emotional
pictures
and
foreground
task.
Cerebral
Cortex,
18(8),
1892-‐1899.
doi:
10.1093/cercor/bhm215
Murphy,
N.
A.,
&
Isaacowitz,
D.
M.
(2008).
Preferences
for
emotional
information
in
older
and
younger
adults:
A
meta-‐analysis
of
memory
and
attention
tasks.
Psychology
and
Aging,
23(2),
263-‐286.
doi:
10.1037/0882-‐7974.23.2.263
Niu,
Y.,
Todd,
R.
M.,
&
Anderson,
A.
K.
(2012).
Affective
salience
can
reverse
the
effects
of
stimulus-‐driven
salience
on
eye
movements
in
complex
scenes.
Frontiers
in
Psychology,
3(336),
1-‐11.
doi:
10.3389/fpsyg.2012.00336
Nothdurft,
H.
C.
(2000).
Salience
from
feature
contrast:
Additivity
across
dimensions.
Vision
Research,
40(10-‐12),
1183-‐1201.
doi:
10.1016/S0042-‐6989(00)00031-‐6
Nummenmaa,
L.,
Hyona,
J.,
&
Calvo,
M.
G.
(2006).
Eye
movement
assessment
of
selective
attentional
capture
by
emotional
pictures.
Emotion,
6(2),
257-‐268.
doi:
10.1037/1528-‐3542.6.2.257
Ohman,
A.,
Flykt,
A.,
&
Esteves,
F.
(2001).
Emotion
drives
attention:
detecting
the
snake
in
the
grass.
Journal
of
Experimental
Psychology:
General,
130(3),
466-‐478.
doi:
10.1037/0096-‐3445.130.3.466
Olatunji,
B.
O.,
Ciesielski,
B.
G.,
Armstrong,
T.,
&
Zald,
D.
H.
(2011).
Emotional
expressions
and
visual
search
efficiency:
Specificity
and
effects
of
anxiety
symptoms.
Emotion,
11(5),
1073-‐1079.
doi:
10.1037/a0021785
Ortega,
V.
(2011).
Avoiding
threat
in
late
adulthood:
Testing
two
lifespan
theories
of
emotion.
Experimental
Aging
Research,
37(4),
449-‐472.
doi:
10.1080/0361073X.2011.590759
65
Padmala,
S.,
&
Pessoa,
L.
(2008).
Affective
learning
enhances
visual
detection
and
responses
in
primary
visual
cortex.
The
Journal
of
Neuroscience,
28(24),
6202-‐6210.
doi:
10.1523/jneurosci.1233-‐08.2008
Pessoa,
L.
(2009).
How
do
emotion
and
motivation
direct
executive
control?
Trends
in
Cognitive
Sciences,
13(4),
160-‐166.
doi:
doi:10.1016/j.tics.2009.01.006
Pessoa,
L.,
Kastner,
S.,
&
Ungerleider,
L.
C.
(2003).
Neuroimaging
studies
of
attention:
From
modulation
of
sensory
processing
to
top-‐down
control.
The
Journal
of
Neuroscience,
23(10),
3990-‐3998.
Phelps,
E.
A.,
Ling,
S.,
&
Carrasco,
M.
(2006).
Emotion
facilitates
perception
and
potentiates
the
perceptual
benefits
of
attention.
Psychological
Science,
17(4),
292-‐299.
doi:
10.1111/j.1467-‐9280.2006.01701.x
Pourtois,
G.,
Grandjean,
D.,
Sander,
D.,
&
Vuilleumier,
P.
(2004).
Electrophysiological
correlates
of
rapid
spatial
orienting
towards
fearful
faces.
Cerebral
Cortex,
14(6),
619-‐633.
doi:
10.1093/cercor/bhh023
Pourtois,
G.,
Schettino,
A.,
&
Vuilleumier,
P.
(2013).
Brain
mechanisms
for
emotional
influences
on
perception
and
attention:
What
is
magic
and
what
is
not.
Biological
Psychology,
92(3),
492-‐512.
doi:
10.1016/j.biopsycho.2012.02.007
Proulx,
M.
J.,
&
Egeth,
H.
E.
(2008).
Biased
competition
and
visual
search:
the
role
of
luminance
and
size
contrast.
Psychological
Research-‐Psychologische
Forschung,
72(1),
106-‐113.
doi:
10.1007/s00426-‐006-‐0077-‐z
Quinlan,
P.
T.,
&
Johnson,
D.
P.
(2011).
The
effect
of
inducing
panic
search
on
the
detection
of
fear-‐relevant
and
neutral
images.
Visual
Cognition,
19(6),
762-‐784.
doi:
10.1080/13506285.2011.587846
Raymond,
J.
E.,
Shapiro,
K.
L.,
&
Arnell,
K.
M.
(1992).
Temporary
suppression
of
visual
processing
in
an
RSVP
task:
an
attentional
blink?
Journal
of
Experimental
Psychology:
Human
Perception
and
Performance,
18(3),
849-‐860.
doi:
10.1037/0096-‐
1523.18.3.849
Reed,
A.
E.,
&
Carstensen,
L.
L.
(2012).
The
theory
behind
the
age-‐related
positivity
effect.
Frontiers
in
Psychology,
3(339),
1-‐9.
doi:
10.3389/fpsyg.2012.00339
Rösler,
A.,
Ulrich,
C.,
Billino,
J.,
Sterzer,
P.,
Weidauer,
S.,
Bernhardt,
T.,
et
al.
(2005).
Effects
of
arousing
emotional
scenes
on
the
distribution
of
visuospatial
attention:
changes
with
aging
and
early
subcortical
vascular
dementia.
Journal
of
the
Neurological
Sciences,
229,
109-‐116.
66
Rowe,
G.,
Hirsh,
J.
B.,
&
Anderson,
A.
K.
(2007).
Positive
affect
increases
the
breadth
of
attentional
selection.
Proceedings
of
the
National
Academy
of
Sciences
of
the
United
States
of
America,
104(1),
383-‐388.
doi:
10.1073/pnas.0605198104
Ruffman,
T.,
Ng,
M.,
&
Jenkin,
T.
(2009).
Older
adults
respond
quickly
to
angry
faces
despite
labeling
difficulty.
Journal
of
Gerontology:
Psychological
Sciences,
64B(2),
171-‐179.
doi:
10.1093/geronb/gbn035
Sakaki,
M.,
Fryer,
K.,
&
Mather,
M.
(in
press).
Emotion
strengthens
high
priority
memory
traces
but
weakens
low
priority
memory
traces.
Psychological
Science.
Sheth,
B.
R.,
&
Pham,
T.
(2008).
How
emotional
arousal
and
valence
influence
access
to
awareness.
Vision
Research,
48(23-‐24),
2415-‐2424.
doi:
10.1016/j.visres.2008.07.013
Singer,
J.
D.
(1998).
Using
SAS
PROC
MIXED
to
fit
multilevel
models,
hierarchical
models,
and
individual
growth
models.
Journal
of
Educational
and
Behavioral
Statistics,
23(4),
323-‐355.
doi:
10.3102/10769986023004323
Stein,
T.,
Peelen,
M.
V.,
Funk,
J.,
&
Seidl,
K.
N.
(2010).
The
fearful-‐face
advantage
Is
modulated
by
task
demands:
evidence
from
the
attentional
blink.
Emotion,
10(1),
136-‐140.
doi:
10.1037/a0017814
Sutherland,
M.
R.,
&
Mather,
M.
(2012).
Negative
arousal
amplifies
the
effects
of
saliency
in
short-‐term
memory.
Emotion,
12(6),
1367-‐1372.
doi:
10.1037/a0027860
Todd,
R.
M.,
Cunninghan,
W.
A.,
Anderson,
A.
K.,
&
Thompson,
E.
(2012).
Affect-‐biased
attention
as
emotion
regulation.
Trends
in
Cognitive
Sciences,
16(7),
365-‐372.
doi:
10.1016/j.tics.2012.06.003
Tsuchiya,
N.,
&
Koch,
C.
(2005).
Continuous
flash
suppression
reduces
negative
afterimages.
Nature
Neuroscience,
8(8),
1096-‐1101.
doi:
10.1038/nn1500
Verbruggen,
F.,
&
De
Houwer,
J.
(2007).
Do
emotional
stimuli
interfere
with
response
inhibition?
Evidence
from
the
stop
signal
paradigm.
Cognition
and
Emotion,
21(2),
391-‐403.
doi:
10.1080/02699930600625081
Vuilleumier,
P.
(2005).
How
brains
beware:
Neural
mechanisms
of
emotional
attention.
Trends
in
Cognitive
Sciences,
9(12),
585-‐594.
doi:
10.1016/j.tics.2005.10.011
Wolfe,
J.
M.
(1994).
Guided
search
2.0
A
revised
model
of
visual
search
Psychonomic
Bulletin
&
Review,
1(2),
202-‐238.
doi:
10.3758/BF03200774
Wurm,
L.
H.,
Labouvie-‐Vief,
G.,
Aycock,
J.,
Rebucal,
K.
A.,
&
Koch,
H.
E.
(2004).
Performance
in
auditory
and
visual
emotional
stroop
tasks:
A
comparison
of
older
and
younger
adults.
Psychology
and
Aging,
19(3),
523-‐535.
doi:
10.1037/0882-‐7974.19.3.523
67
Yang,
E.,
Zald,
D.
H.,
&
Blake,
R.
(2007).
Fearful
expressions
gain
preferential
access
to
awareness
during
continuous
flash
suppression.
Emotion,
7(4),
882-‐886.
doi:
10.1037/1528-‐3542.7.4.882
Zeelenberg,
R.,
&
Bocanegra,
B.
R.
(2010).
Auditory
emotional
cues
enhance
visual
perception.
Cognition,
115(1),
202-‐206.
doi:
10.1016/j.cognition.2009.12.004
68
Appendix A: Tables
Table 1: Library numbers for IADS stimuli used in Experiment 1
Stimulus Type IADS Library Numbers
Negative Arousing 106 115 134 244 255 260 276 279 282 283
289 292 420 501 600 624 626 711 712 730
Neutral 102 113 130 132 170 225 246 250 252 322
358 373 375 377 382 701 708 720 723 728
69
Table 2: Hierarchical linear model (HLM) analysis of older adults (n = 55) and younger
adults (n = 110) from Experiment 1 using the arousal ratings (1 = least arousing, 9 = most
arousing) of the sounds, the valence ratings of the sounds (1 = most negative, 5 = neutral, 9 =
most positive), and age group as well as the mean dB (physical intensity) of the sounds to
predict (A) difference scores (high-salience minus low-salience), (B) the proportion of high-
salience letters recalled and (C) the proportion of low-salience letters recalled.
A. Difference Scores % Recalled (High-salience minus Low-salience)
Effect b SE t df p
Intercept 0.240894 0.017790 13.541 164 <0.001
Arousal Rating 0.005938 0.002362 2.514 6432 0.01195
Valence Rating - 0.004353 0.002492 - 1.746 6432 0.0808
Mean dB 0.000224 0.000302 0.742 6432 0.4582
Age 0.026409 0.017790 1.484 163 0.1396
B. High-salience % Recalled
Effect b SE t df p
Intercept 0.5566667 0.0121296 45.89 164 <0.001
Arousal Rating 0.0046819 0.0015780 2.97 6432 0.003018
Valence Rating - 0.0028761 0.0016652 - 1.73 6432 0.08419
Mean dB 0.0001448 0.0002018 0.72 6432 0.4731
Age - 0.0590909 0.0121296 -4.87 163 <0.001
C. Low-salience % Recalled
Effect b SE t df p
Intercept 0.3158 0.009202 34.31 164 <0.001
Arousal Rating -0.001256 0.001084 -1.16 6432 0.2464
Valence Rating 0.001476 0.001144 1.29 6432 0.1967
Mean dB -0.00007926 0.0001386 -0.57 6432 0.5674
70
Age -0.08550 0.009202 -9.29 163 <0.001
71
Table 3: Hierarchical linear model (HLM) analysis of Experiment 1 using the arousal and
valence ratings of the sounds, the mean dB of the sounds and age group as predictors of the
number of incorrectly reported letters.
Older and Younger Adults Incorrect Responses
Effect b SE t df p
Intercept 0.5529545 0.0457304 12.09 164 <0.001
Arousal Rating -0.0055707 0.0041893 -1.33 6432 0.184
Valence Rating -0.0092037 0.0044209 -2.08 6432 0.037
Mean dB 0.0004734 0.0005357 0.88 6432 0.377
Age -0.1511364 0.0457304 -3.31 163 0.001
72
Table 4: IADS library numbers for all 80 audio clips.
Stimulus Type IAPS Library Numbers
Low-Arousal Neutral 104 107 113 120 132 170 171 2002 2004 2010 2015 2019 2039 206 225
245 250 252 262 322 358 361 373 374 375 376 377 382 425 627 700 701
705 708 720 722 723 724 728 729
High-Arousal Negative 106 115 134 255 260 261 276 277 279 285 289 290 292 293 312 420 422
626 711 712
High-Arousal Positive 110 2101 2102 2103 2105 2106 2107 2108 2110 2111 2112 2114 2115
2117 2119 220 226 311 351 365
73
Table 5: Hierarchical linear model (HLM) analysis of older adults (n = 55) and younger
adults (n = 110) from Experiment 2 using the arousal ratings (1 = least arousing, 9 = most
arousing) of the sounds, the valence ratings of the sounds (1 = most negative, 5 = neutral, 9 =
most positive), and age group as well as the mean dB (physical intensity) of the sounds to
predict (A) difference scores (high-salience minus low-salience), (B) the number of high-
salience letters recalled and (C) the number of low-salience letters recalled.
A. Difference Scores (High-salience minus Low-salience)
Effect b SE t df p
Intercept 0.2810000 0.0214263 13.115 109 < 0.001
Arousal Rating 0.0029724 0.0020147 1.475 8687 0.1402
Valence Rating -0.0017453 0.0017792 -0.981 8687 0.3267
Mean dB 0.0003074 0.0002927 1.050 8687 0.3267
Age -0.0070758 0.0214263 -0.330 108 0.7419
B. High-salience Recall
Effect b SE t df p
Intercept 0.5730 0.01449 39.53 109 < 0.001
Arousal Rating 0.002349 0.001372 1.71 8687 0.08702
Valence Rating -0.001103 0.001212 -0.91 8687 0.363
Mean dB 0.00004734 0.0001994 0.24 8687 0.8123
Age -0.06167 0.01449 -4.25 108 <0.001
C. Low-salience Recall
Effect b SE t df p
Intercept 0.2919545 0.0126946 22.998 109 <0.002
Arousal Rating -0.0006237 0.0009229 -0.676 8687 0.4992
Valence Rating 0.0006428 0.0008150 0.789 8687 0.4303
Mean dB -0.0002601 0.0001341 -1.940 8687 0.05244
74
Age -0.0545909 0.0126946 -4.300 108 <0.001
75
Table 6: Hierarchical linear model (HLM) analysis of young adults (n = 55) from Experiment
2 using the arousal ratings (1 = least arousing, 9 = most arousing) of the sounds, the valence
ratings of the sounds (1 = most negative, 5 = neutral, 9 = most positive), and the mean dB
(physical intensity) of the sounds to predict (A) difference scores (high-salience minus low-
salience), (B) the proportion of high-salience letters recalled and (C) the proportion of low-
salience letters recalled.
A. Difference Scores % Recalled (High-salience minus Low-salience)
Effect b SE t df p
Intercept 0.2880758 0.0313715 9.183 54 <0.001
Arousal Rating 0.0060002 0.0028453 2.109 4342 0.03502
Valence Rating 0.0018774 0.0026790 0.701 4342 0.4835
Mean dB 0.0007580 0.0004316 1.756 4342 0.07912
B. High-salience % Recalled
Effect b SE t df p
Intercept 0.6346212 0.0188505 33.67 54 <0.001
Arousal Rating 0.0044626 0.0018999 2.35 4342 0.01887
Valence Rating 0.0015012 0.0017888 0.84 4342 0.4014
Mean dB 0.0002599 0.0002882 0.90 4342 0.3672
C. Low-salience % Recalled
Effect b SE t df p
Intercept 0.3465455 0.0188223 18.411 54 <0.001
Arousal Rating -0.0015376 0.0013330 -1.154 4342 0.2488
Valence Rating -0.0003762 0.0012551 -0.300 4342 0.7644
Mean dB -0.0004981 0.0002022 -2.464 4342 0.01379
76
Table 7: Hierarchical linear model (HLM) analysis of Experiment 2 using the arousal and
valence ratings of the sounds, the mean dB of the sounds and age group as predictors of the
number of incorrectly reported letters.
Older and Young Adults Incorrect Responses
Effect b SE t df p
Intercept 0.06364 0.005262 12.093 109 <0.001
Arousal Rating 0.00001969 0.0004351 0.045 8687 0.9639
Valence Rating 0.0006166 0.0003843 1.605 8687 0.1086
Mean dB 0.00006550 0.00006322 1.036 8687 0.3002
Age -0.0008523 0.005262 -0.162 108 0.8716
77
Table 8: Hierarchical linear model (HLM) analysis of Experiment 2 using the arousal and
valence ratings of the sounds, the mean dB of the sounds and age group as predictors of the
number of incorrectly reported letters in young adults.
Young Adults Incorrect Responses
Effect b SE t df p
Intercept 0.06449 0.007015 9.193 54 <0.001
Arousal Rating -0.0004611 0.0006071 - 0.760 4342 0.4476
Valence Rating 0.0006154 0.0005716 1.077 4342 0.2817
Mean dB 0.0001336 0.00009208 1.451 4342 0.1469
78
Table 9: IAPS library numbers for all 120 images used in Experiment 3.
Stimulus Type IAPS Library Numbers
Low-Arousal Neutral 2383 2396 2441 2480 2500 2575 2590 2593 2594 2595 2620 2840 2880
5130 5390 5510 5535 5661 5720 5731 5740 5750 5900 7002 7009 7025
7031 7036 7037 7038 7041 7052 7060 7130 7140 7150 7161 7175 7179
7705
High-Arousal Negative 1019 1026 1050 1052 1114 1200 1220 1300 1525 1931 2053 2352.2 2730
2981 3010 3030 3051 3060 3071 3100 3101 3120 3130 3150 3168 3170
3180 3266 3350 3400 6230 6570.1 8230 9040 9253 9300 9405 9410 9600
9635.1
High-Arousal Positive 1463 2050 4607 4608 4611 4650 4651 4652 4653 4656 4658 4659 4666
4670 4676 4680 4800 5260 5450 5470 5621 5623 5626 8030 8031 8040
8178 8179 8180 8185 8186 8193 8200 8210 8300 8341 8370 8400 8490
8496
79
Table 10: Recognition memory biases for the goal-relevant item predicted by salience type,
location type, emotion type, age group and fixation count biases towards the goal-relevant item,
as well as the interaction between age group and fixation count biases.
Predictor b SE t df p
Intercept 0.6049355 0.0401486 15.067 71 < 0.001
Salience -0.0027215 0.0053239 -0.511 8563 0.6092
Location -0.0008928 0.0053252 -0.168 8563 0.8669
Emotion -0.0070675 0.0065215 -1.084 8563 0.2785
Age Group -0.0198330 0.0401486 -0.494 70 0.6229
Fix Count 0.0335252 0.0023535 14.245 8563 < 0.001
Age Group × Fix
Count
0.0036925 0.0023531 1.569 8563 0.1166
80
Table 11: Fixation count bias scores predicted by the brightness and contrast levels of the
IAPS images, the salience type and the location of the goal-relevant item, the emotion type of
the IAPS image, the age group of the participants, the mean pupil sizes observed during the
ISI period, as well as the interaction between age group and mean pupil size.
Predictor b SE t df p
Intercept 3.1528935 0.0849774 37.10 71 <0.001
Brightness 0.0005338 0.0007797 0.68 8561 0.4936
Contrast 0.0002530 0.0016506 0.15 8561 0.8782
Salience 0.2092151 0.0244858 8.54 8561 <0.001
Location - 0.0476239 0.0244874 - 1.94 8561 0.05183
Emotion - 0.0190890 0.0303708 - 0.63 8561 0.5297
Age Group - 0.0737269 0.0849774 - 0.87 70 0.3886
Mean Pupil ISI - 0.0003756 0.0001334 - 2.82 8561 0.004881
Age Group ×
Mean Pupil ISI
- 0.0001289 0.0001329 - 0.97 8561 0.3321
81
Table 12: Recognition memory bias scores predicted by the brightness and contrast levels of
the IAPS images, the salience type and the location of the goal-relevant item, the emotion type
of the IAPS image, the age group of the participants, the mean pupil sizes observed during the
ISI period, as well as the interaction between age group and mean pupil size.
Predictor b SE t df p
Intercept 0.6049000 0.04015 15.068 71 < 0.001
Brightness 0.0000973 0.0001714 0.568 8561 0.5701
Contrast 0.0000920 0.0003628 0.253 8561 0.7999
Salience -0.0025240 0.005389 -0.468 8561 0.6396
Location -0.0023830 0.005389 -0.443 8561 0.6576
Emotion -0.0076400 0.006686 -1.143 8561 0.2532
Age Group 0.0198300 0.040150 -0.494 70 0.6229
Mean Pupil ISI -0.0000720 0.00002975 -2.418 8561 0.01561
Mean Pupil ×
Age
-0.0000478 0.00002964 -1.612 8561 0.1069
82
Appendix B: Figures
Figure 1. Schematic depiction of the procedure used in Experiment 1 and Experiment 2.
83
Figure 2: Schematic depiction of the procedure used in Experiment 3.
1
84
Figure 3: Mean pupil sizes for older adults and young adults across emotion type.
0
200
400
600
800
1000
1200
1400
1600
1800
Older
Adults
Mean
Pupil
Size
Young
Adults
Mean
Pupil
Size
Neutral
Negative
Positive
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Emotional arousal amplifies the selectivity of visual selective attention
PDF
The emotional arousal effects on visual processing
PDF
Influence of age and anxiety on recognition of facial expressions of emotion: exploring the role of attentional processes
PDF
Aging and emotion regulation in the judgment of facial emotion
PDF
Integrating top-down and bottom-up visual attention
PDF
Noradrenergic mechanisms of arousal-enhanced memory selectivity
PDF
Age differences in the effects of anxiety on selective attention: a test of the vigilance-avoidance hypothesis
PDF
Mechanisms of stress effects on learning and decision making in younger and older adults
PDF
The impact of emotion, valence, and arousal on differential memory processes in younger and older people
PDF
Dedifferentiation of emotion regulation strategies in the aging brain: an MVPA investigation
PDF
Emotional reactivity and memory biases in older and younger adults
PDF
Temporal dynamics of attention: attention gating in rapid serial visual presentation
PDF
How emotional arousal influences memory and learning in younger and older adults
PDF
How age-related changes in the locus coeruleus affect selective attention and memory
PDF
Influence of sad mood and old age schema on older adults' physical symptoms processing
PDF
Optimism when imagining the future in old age influences the positivity effect in memory
PDF
Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults
PDF
Structure and function of the locus coeruleus across the lifespan
PDF
Psychophysiological assessment of cognitive and affective responses for prediction of performance in arousal inducing virtual environments
PDF
The role of age, mood, and time- perspective on the interpretation of emotionally ambiguous information
Asset Metadata
Creator
Sutherland, Matthew Ryan
(author)
Core Title
Emotion, attention and cognitive aging: the effects of emotional arousal on subsequent visual processing
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Psychology
Publication Date
06/17/2014
Defense Date
04/09/2014
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
attention,cognitive aging,emotional arousal,memory,OAI-PMH Harvest
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Mather, Mara (
committee chair
), Dawson, Michael Edward (
committee member
), Itti, Laurent (
committee member
), Tjan, Bosco S. (
committee member
), Wilcox, Rand R. (
committee member
)
Creator Email
mattryansutherland@gmail.com,msutherland@research.baycrest.org
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-420227
Unique identifier
UC11285934
Identifier
etd-Sutherland-2553.pdf (filename),usctheses-c3-420227 (legacy record id)
Legacy Identifier
etd-Sutherland-2553.pdf
Dmrecord
420227
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Sutherland, Matthew Ryan
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
Tags
attention
cognitive aging
emotional arousal
memory