• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Emotional influences on the scope of selective visual attention

Dissertation
Author: Ezra Saul Dusseau Wegbreit
Abstract:
Emotion interacts with numerous areas of cognition, including memory, decision-making, and attention (Dolan, 2002). In this dissertation, I conduct four experiments investigating the role of emotion on selective visual attention. In Chapter 1, I describe basic emotion research and the "scope" of selective attention. I outline previous research on the effects of happiness and anxiety on selective attention, some potential mechanisms of these effects, and their practical significance. In Chapter 2, I present an experiment building on prior findings that positive moods broaden the scope of selective attention (Rowe, Hirsh, & Anderson, 2007). I also present advantages and disadvantages of methods to induce and assess emotional states. In my experiment, participants showed greater flanker interference in a happy mood and less flanker interference in an anxious mood, indicating broadened and narrowed attentional scopes, respectively. In Chapter 3, I present two experiments exploring the role of mood in featural selection. I outline evidence for a featural scope of attention and changes to the flanker task to incorporate featural selection. When in an anxious mood, participants showed reduced flanker interference when a feature distinguished the target from the flankers, but more interference when the flankers shared the feature. In contrast, in a positive mood, these participants showed relatively non-selective featural attention. A follow-up experiment found that this effect was modified by task context. In Chapter 4, I examine mood influences in a visual search task similar to the flanker task. I outline the importance of visual search and reasons why mood states might influence it. In this search task, no reliable mood effects were found on visual search efficiency, except for a potential between-subjects effect. I interpret these findings in light of attentional theory and potential issues with the search task. In Chapter 5, I present overall conclusions and limitations, as well as how this research fits into the existing literature. Overall, participants. spatial and featural selection is broadened by happy moods and narrowed by anxiety, but this may not influence preattentive processing in visual search. I conclude with areas of improvement for future studies and research directions based on this work.

Table of contents Abstract ........................................................................................................................................... 3 Dedication ....................................................................................................................................... 5 Acknowledgments........................................................................................................................... 6 List of Tables ................................................................................................................................ 11 Chapter 1. Interactions between emotion and attention ............................................................... 12 1.1 Basic emotions and the dimensional structure of affect ...................................................... 16 1.2 The scope of visual attention. .............................................................................................. 21 1.3 Influence of happy and anxious mood states on the perceptual locus of selection ............. 26 1.4 Underlying mechanisms of the effects of mood states on selective visual attention. ......... 31 1.5 Applied relevance of research on the influences of mood on selective attention. .............. 32 Chapter 2: Mood influences on spatial attention scope ................................................................ 34 2.1 Experiment 1 Overview ...................................................................................................... 34 2.1.1 Differences in laboratory techniques to elicit emotion ................................................. 39 2.1.2 Methods of emotion assessment ................................................................................... 45 2.2 General experimental setup and equipment ........................................................................ 55 2.3 Experiment 1 Methods ........................................................................................................ 56 2.3 Experiment 1 Results .......................................................................................................... 59 2.4 Experiment 1 Discussion ..................................................................................................... 67

9

Chapter 3: Mood influences on featural attention scope .............................................................. 73 3.1.1 Experiment 2a Overview .............................................................................................. 73 3.1.2 Experiment 2a Methods ................................................................................................... 77 3.1.3 Experiment 2a Results ...................................................................................................... 81 3.1.4 Experiment 2a Discussion and Experiment 2b Overview ................................................ 88 3.2.1 Experiment 2b Methods ................................................................................................... 91 3.2.2 Experiment 2b Results ..................................................................................................... 92 3.2.3 Experiment 2b Discussion ............................................................................................ 99 3.3 Experiment 2 General Discussion ..................................................................................... 101 Chapter 4: Mood influences on the mechanisms of visual search .............................................. 105 4.1 Experiment 3 Overview .................................................................................................... 105 4.2. Experiment 3 Methods ..................................................................................................... 114 4.3. Experiment 3 Results ....................................................................................................... 117 4.4. Experiment 3 Discussion .................................................................................................. 128 Chapter 5. General Discussion .................................................................................................... 134 References ................................................................................................................................... 146 Appendix ..................................................................................................................................... 164 Curriculum Vita .......................................................................................................................... 167

10

List of Figures Figure 1: Visual analog mood scale .............................................................................................. 58 Figure 2: Experiment 2 Flanker Congruency Effects (FCEs) by Mood and Distance ................. 60 Figure 3: Experiment 1 Participants‟ moods immediately after the film clips ............................. 64 Figure 4: Experiment 1 participants‟ residual moods immediately after the flanker blocks ........ 65 Figure 5: Flanker displays for Experiments 2a and 2b ................................................................. 80 Figure 6: Experiment 2a Flanker Congruency Effects (FCEs) by Mood and Color condition .... 83 Figure 7: Experiment 2a participants‟ moods immediately after the film clips .......................... 85 Figure 8: Experiment 2a participants‟ residual moods immediately after the flanker blocks ...... 86 Figure 9: Experiment 2b Flanker Congruency Effects (FCEs) by Mood and Color condition .... 94 Figure 10: Experiment 2b Participants‟ moods immediately after the film clips ........................ 96 Figure 11: Experiment 2b participants‟ residual moods immediately after the flanker blocks .... 98 Figure 13: Schematic search display for Experiment 3 .............................................................. 113 Figure 14: Experiment 3 mean search RT by trial type across mood blocks ............................. 119 Figure 15: Experiment 3 mean search RT slope (ms/item) by trial type across mood blocks ... 120 Figure 16: Experiment 3 mean accuracy by trial type across mood blocks ............................... 123 Figure 17: Experiment 3 mean search accuracy slope by trial type across mood blocks ........... 123 Figure 18: Experiment 3 participants‟ moods immediately after the film clips ........................ 125 Figure 19: Experiment 3 participants‟ residual moods immediately after the search blocks ..... 126 Figure 12: Comparison of Flanker Congruency Effects (FCEs) in Experiments 2a and 2b...... 165

11

List of Tables Table 1: Experiment 1 raw RTs (SEMs) in ms ........................................................................... 164 Table 2: RGB Color Codes used in Flanker Tasks in Experiment 2 .......................................... 164 Table 3: Experiment 2a raw RTs (SEMs) in ms ......................................................................... 164 Table 4: Experiment 2b raw RTs (SEMs) in ms ......................................................................... 164 Table 5: Quadrant and Angular Position of Search Stimuli in Experiment 3 ............................. 165 Table 6: Mean RTs (SEMs) in ms for the Search Task in Experiment 3 ................................... 166 Table 7: Mean Percentage Accuracy (SEM) for the Search Task in Experiment 3 ................... 166

12

Chapter 1. Interactions between emotion and attention For many years cognitive psychologists thought of emotion and cognition as relatively non-interacting systems instantiated in separate, modular areas of the brain. However, in recent years researchers have been discovering interactions between cognitive and emotional processing, both behaviorally and neurally (Pessoa, 2008, 2009). This dissertation focuses on the influence of emotional states on the cognitive process of attention specifically (i.e. rather than memory, decision making, etc.). Attention is an important determinant of cognition because in information processing many stimuli compete for limited processing resources, and attention determines which information is processed further and which information is discarded (Desimone & Duncan, 1995; Driver, 2001; Vecera, 2000; Wells & Matthews, 1994). Indeed, the proper functioning of attention is necessary for us to make sense of the world (Lavie, 2005). Otherwise, we would be lost in the “blooming, buzzing confusion” described over a century ago by William James (1890), as some people with impaired attention due to brain damage seem to be (Robertson, Treisman, Friedman-Hill, & Grabowecky, 1997). Thus, the study of the influence of emotions on cognition should include investigations of emotion-attention interactions (Vuilleumier, 2005; Vuilleumier & Huang, 2009). Evidence suggests that people‟s spatial attention has a measurable scope or breadth and that this scope is malleable depending on people‟s goals and on task demands (C.W. Eriksen & St. James, 1986; C. W. Eriksen & Yeh, 1985). The scope of people‟s selective attention can be influenced by their emotional status (Derryberry & Reed, 1998; Fredrickson & Branigan, 2005; Gasper, 2004a; Rowe, et al., 2007). These emotional influences on attention can come from the emotional content of environmental stimuli (Frischen, Eastwood, & Smilek, 2008; Vuilleumier & Huang, 2009; Weierich, Treat, & Hollingworth, 2008), from the current emotional state of the

13

observer (Rowe, et al., 2007), and from the observer‟s chronic emotional traits (e.g. trait anxiety; Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, & van IJzendoorn, 2007). Experimental research on the interactions between attention and emotion has direct relevance for improving models of emotion and emotional disorder (Wells & Matthews, 1994). Understanding these interactions can help us understand the etiology and maintenance of many mood disorders, as well as better informing us about the way normal mood states impact our daily lives (Fredrickson, 2003; Fredrickson & Branigan, 2005). In Chapter 1, section 1, I briefly explain two major theoretical approaches to the study of emotion and make some distinctions within this literature which are important to describe why and how I induced and assessed specific affect states. In section 2, I briefly describe what is meant by the “scope” of selective attention, as attention scope can refer to selection of locations, objects, and features. A complete review of the nature of attentional selection is well beyond the scope of this dissertation. In section 3, I outline previous manipulations of mood on selective attention, focusing on the influence of two specific mood states: happiness and anxiety. I conclude Chapter 1 with some putative mechanisms of these effects and the practical significance of research on the influences of mood on attentional selection. Chapter 2 describes Experiment 1, which builds on previous findings in the literature of emotion-attention interactions. The previous research (Rowe, et al., 2007) induced happy, sad, and neutral mood states with music and examined their influence on flanker interference in various amounts of flanker distance on this task. I interpret their results and note some advantages and disadvantages of the research design. Next, I describe methods by which researchers have induced emotional states in participants, and the advantages and disadvantages

14

of each of these. Finally, I describe the methods by which researchers have measured mood and the effects of their mood inductions, and I describe the reasons why I chose to use films to induce moods and to use visual analog scales to measure anxiety, happiness, and arousal. Participants in Experiment 1 showed greater flanker interference while in a happy mood, indicative of a broadened attentional scope, and showed less flanker interference while in an anxious mood, indicative of a narrowed attentional scope. There were some differences between the results of Experiment 1 and the findings by Rowe et al. (2007), but the results of the two studies were not inconsistent, and I outline some potential reasons for the differences that were found. Chapter 3 describes two new experiments (Experiments 2a and 2b) that investigate the role of featural attention on the effect found in Experiment 1. First, I outline some relevant findings that suggest that participants can take a featural scope of attention, in addition to a spatial one. Next, I describe the changes I made to the flanker task in Experiment 2a to examine the role of featural selection in modifying the influence of anxious and positive moods on flanker interference. In Experiment 2a, I found that when participants are in an anxious mood and a helpful feature (color) distinguishes the target from the flankers then they show reduced response conflict on the flanker task, but when the same feature that was helpful previously is shared by the flankers, participants experiencing anxiety are more impacted. In contrast, when in a positive mood these same participants did not show a difference in flanker interference regardless of the colors of the letters, indicating relatively broad non-selective attention to features on the flanker task. Finally, Experiment 2b describes evidence that the benefit of an anxious mood found in Experiment 2b can be modified by task context in that it is only in the

15

presence of a helpful feature that anxious participants can adopt a task set and show a selection benefit. In Chapter 4, I examine the role that these emotions play in a visual search task that was putatively analogous to the flanker task. First, I briefly outline the importance of the visual search task in measuring attentional deployment. Next, I describe the reasons why mood states might influence visual search performance. Then, I present my version of a search task designed to parallel the flanker tasks in which existing results were found. In this search task, no reliable mood effects were found on the efficiency of visual search, except for a potential between- subjects effect. I conclude Chapter 4 with an interpretation of these findings in light of attentional theory, and some of the potential issues with the implementation of the search task. In Chapter 5, I present my overall conclusions about these findings, some of their limitations, and how they fit into the existing literature. In brief, when focused attention was required, participants showed relatively broadened spatial and featural attention in a happy mood, and showed an increased ability to narrow attention to spatial locations or features when in an anxious mood. In contrast, when required to search for an item across the display, anxious participants showed more accurate performance than did happy participants. Some key limitations of these studies include the lack of a neutral mood condition, the lack of truly neutral flankers, and differences in motivational focus between happy and anxious moods. However, these findings make a novel contribution to the literature by examining how happy and anxious mood states influence focusing attention to affectively-neutral stimulus features and influence visual searches for them. Chapter 5 concludes with some areas of improvement for future studies, and some exciting new research directions that are based on this work, including the role

16

of affective traits, such as trait anxiety and motivational focus, as well as investigations of the neural mechanisms of these effects.

1.1 Basic emotions and the dimensional structure of affect

Emotions are complex. On the surface most people have some idea of what an emotion is, but there is still substantial disagreement in scientific circles about what specifically constitutes an emotion (Izard, 2009; Larsen & Fredrickson, 2003; Watson & Clark, 1997) and exactly which brain structures are involved in emotion (Pessoa, 2008, 2009). However, researchers do agree that emotions involve a complex interplay between psychological and physiological factors, such as subjective experiences, central and peripheral nervous system activation, facial expressions, and changes in cognitive processing (Gable & Harmon-Jones, 2010b; Izard, 2009; Larsen & Fredrickson, 2003; Rottenberg, Ray, & Gross, 2007). Additionally, emotions such as anxiety, anger, and disgust have been defined as states that elicit specific action tendencies, such as fleeing, attacking, or expelling noxious substances from the body, respectively. In contrast, positive moods tend to elicit a more vague and diffuse set of action tendencies, such as exploring the environment and attending to something that one has started doing, but also aimlessness, inactivity, and feeling content or complacent (Fredrickson, 2001). These complexities of emotion classification mean that it is important for researchers to specify a working definition of what is meant by a discrete emotion (Izard, 2009) or ascribe to at least a basic model of emotion in order to lay the groundwork for their work (Larsen & Fredrickson, 2003; Watson & Clark, 1997). A prominent way to classify human emotional experience is to carve it up into its constituent basic emotions (Izard, 2009). Different

17

classification systems have been proposed, but it is commonly thought that “basic emotions” include fear, sadness, happiness, anger, surprise, and disgust (Ekman, 1992; Ekman, et al., 1987). Critics of these basic emotion theorists point out that there are numerous other emotion classification schemes and that other researchers postulate a plethora of additional basic emotions, such as pride, contempt, or shame (Ortony & Turner, 1990). Although, there is also disagreement as to whether these so-called basic emotions exist (Ekman, 1992; Ortony & Turner, 1990), evidence for the existence of the six basic emotions listed above comes from numerous studies showing convergent and discriminant validity of ratings of these six basic emotions (Watson & Clark, 1997). In addition, the evidence from a large number of studies of emotional expression recognition suggests that, although there do exist culturally specific patterns of expression within different cultures (Elfenbein & Ambady, 2002), the basic six emotional expressions are recognized across a broad array of Eastern and Western cultures (Ekman, et al., 1987; Elfenbein & Ambady, 2002). However despite evidence of the apparent independence of the six basic emotions, actual measurements of self-reported affect tend to a large extent correlate systematically with each other. People who tend to exhibit high levels of anxiety also tend to report higher levels of other negative emotions, such as sadness and anger, and people who report higher levels of happiness also tend to report higher levels of other positive emotions (Watson & Clark, 1997). Thus, theories of discrete emotions may not capture some of the underlying structure or dimensions of affect. One influential model postulates that there are two major dimensions of affect that lie orthogonal to each other, with various basic emotions arranged around them in a circle (Russell,

18

1980). The first factor of emotion is the valence dimension, which is whether one is feeling pleasant or unpleasant. The second factor is arousal, which is the extent to which the person is feeling engaged, activated, or energized (Larsen & Fredrickson, 2003; Russell, 1980; Watson & Clark, 1997). Weak evidence for another dimension, that of dominance or aggression, is sometimes found (Russell, 1980), but this dimension does not always replicate across cultures as the other two do, and thus is not often measured (Watson & Tellegen, 1985). Evidence for the veracity of this model comes from factor analyses of self-reports of affect, overt facial and vocal expression, and analyses of affective words existing in languages (Russell, 1980; Watson & Clark, 1997). Another conceptualization of the circumplex model of affect rotates these pleasantness and arousal axes by 45 degrees and postulates that the principal components of affect are two orthogonal dimensions of positive affect and negative affect (Watson & Clark, 1997; Watson & Tellegen, 1985). This second circumplex model was created to update the first model because of a problem with the first circumplex model of affect – namely, that positive mood and negative mood ratings often fail to correlate negatively with one another (Watson & Tellegen, 1985), even when methodological problems with affect scales are controlled (Watson, Clark, & Tellegen, 1988). “Positive Affect” represents how much joy, enthusiasm, and other positive emotions one is feeling, whereas “Negative Affect” represents the extent to which one is feeling sad, angry, scared, and other negative emotions. The orthogonality of these two dimensions allows for the possibility of positive and negative affect to be uncorrelated, such that one could be feeling a mixture of positive and negative affect states (e.g. happiness and anxiety). The Positive Affect- Negative Affect dimensional model shows high reliability and discriminant validity across a

19

variety of response formats and time frames (Watson, 1988), good convergent validity between self- and peer ratings of trait affect (Watson & Clark, 1991), and reliability in ratings between- and within-participants and across cultures (Watson & Clark, 1997; Watson & Tellegen, 1985). As a result, the circumplex model has become the most influential model of the underlying “dimensional” structure of subjective affective experience (Gray & Watson, 2007; Watson & Clark, 1997; Watson & Tellegen, 1985). Nevertheless, basic emotions theorists point out that dimensional models may miss some distinctions between emotional states. For example, fear and anger are both highly arousing, negatively valenced states (see also Gable & Harmon-Jones, 2010b). However, the dimensional theorists point out that discrete emotions tend to be correlated when measured together, suggesting an underlying structure of affect (Watson & Clark, 1997). Despite this seeming conflict between theories of “basic” emotions and the circumplex model, these two principal approaches to affective experience may not be mutually exclusive. Each of the basic emotions cited by theorists may have a valence component and an arousal component while also possessing attributes unique to that emotion (Nielsen & Kaszniak, 2007). In this dissertation, the influence of two discrete emotions, anxiety and happiness, on the scope of visual attention will be assessed. In addition, the role of arousal, a potentially confounding affective dimension, will be measured and discussed. Some studies have claimed that differences in attentional deployment to emotional and affectively-neutral stimuli are in fact caused by the differences in arousal elicited by these types of stimuli (e.g. Anderson, 2005). In addition, other researchers have pointed out that some affective influences on attention depend on specific interactions between emotional valence and arousal (e.g. Jefferies, Smilek, Eich, &

20

Enns, 2008). Thus, including an assessment of arousal is warranted as a way of testing these hypotheses in the present set of experiments. Another dimension of affective experience is the role of motivation. Motivations are what make an organism work to obtain a reward or avoid punishment (Pessoa, 2009). They vary in intensity (low to high) and direction (approaching a stimulus or withdrawing from it), and behavioral and neural evidence suggests that they may constitute a third dimension of affect distinct from valence and arousal (Gable & Harmon-Jones, 2008, 2010a, 2010b; Harmon-Jones & Gable, 2009) Although not specifically assessed empirically in this dissertation, the role of motivational factors in setting the scope of attention will be discussed in Chapter 5 as a possible additional variable to be included in future research. There are other distinctions that researchers make between emotion terms. While many researchers have treated mood and emotion as synonymous (e.g. Westermann, Spies, Stahl, & Hesse, 1996) or nearly so (Martin, 1990), other emotion researchers caution against equating these terms (Gray & Watson, 2007; Rottenberg, et al., 2007). These researchers accept that both “emotion” and “mood” are subsumed under the umbrella term of “affect”, but they argue that moods and emotions are best defined as occurring on different time scales and with different patterns of activation. The term “emotion” best corresponds to a briefer, typically more intense state (e.g. seconds, minutes), whereas the term “mood” best corresponds to the less intense way a person is feeling for a significant amount of time (e.g. hours, days). Emotions are felt less frequently than moods because it would be evolutionarily maladaptive to experience such intense affective states all the time (Fredrickson & Branigan, 2005). Moods, on the other hand, are probably ubiquitous during our everyday lives. Furthermore, emotions occur in response to a

21

particular stimulus (e.g. a snake, a funny film clip) whereas moods are not engendered by or connected to a specific thought or external stimulus, but rather are a summary of one‟s affective state (Gray & Watson, 2007). However, given the plethora of researchers using the terms “mood” or “mood induction” in their research (e.g. Martin, 1990; Mayer, Allen, & Beauregard, 1995; Westermann, et al., 1996), when they probably should be employing the terms “emotion” and “emotion induction”, these terms may be used interchangeably in this work while keeping in mind this distinction, with the terms “mood” and “mood induction” treated as synonyms for “emotion” and “emotion induction”. Finally, in addition to the distinction between the terms “mood” and “emotion”, there is also a distinction between state affect and trait affect. The term “state affect” corresponds to moods and emotions that are transiently felt by the individual and the term “trait affect” corresponds to long-term and relatively stable tendencies of the individual to feel various affective states (Gray & Watson, 2007). Thus, people‟s affective experience encompasses short- term emotional experiences, more stable mood states, and long-term trait dispositions.

1.2 The scope of visual attention.

Before one can understand how affect influences attention, it is important to have at least a working definition of visual selective attention. A more comprehensive treatment of selective attention is well beyond the scope of this work (for further review, see Cave, 1999; Driver, 2001; Lavie, 2005; Pessoa, Kastner, & Ungerleider, 2003; Wells & Matthews, 1994), but some specificity will aid in explaining mood and attention interactions. A fundamental aspect of attention is that it is selective, meaning that some information is selected for further processing, while other information is attenuated or discarded (Cave & Bichot, 1999; Weierich, et al., 2008;

22

Wells & Matthews, 1994). At the neural level, attention can be described the combination of a collection of bottom-up data-driven mechanisms that, in part, bind objects together (Treisman & Gelade, 1980) and top-down biasing signals that boost the neural signal for competing inputs (Desimone & Duncan, 1995; Vecera, 2000). However, attention is not a unitary concept and can be divided into several parts (Weierich, et al., 2008; Wells & Matthews, 1994). Three important attentional functions are alerting, orienting, and executive control (Posner & Petersen, 1990). They rely on networks that are separable behaviorally (Fan, McCandliss, Sommer, Raz, & Posner, 2002) and neurally (Fan, McCandliss, Fossella, Flombaum, & Posner, 2005), but that also interact with each other (Callejas, Lupiáñez, & Tudela, 2004). The alerting function allows people to achieve a state of readiness to process signals, also known as vigilance. The alerting function depends critically on the functioning of the right frontal lobe (Fan, et al., 2005), and particularly an optimal level of norepinephrine (Posner & Petersen, 1990) which is likely to be increased by anxiety and highly- arousing positive moods (Ashby, Isen, & Turken, 1999; Ashby, Valentin, & Turken, 2002). The orienting function allows us to direct our attention to a particular spatial location, and a network of brain areas in the dorsal “where” processing stream (e.g. right parietal lobe) and thalamus are involved in producing orienting responses (Fan, et al., 2005; Posner & Petersen, 1990). The orienting function involves disengaging from the current focus, shifting attention, and reengaging attention on the new object (Posner & Petersen, 1990). A problem could occur at any of these stages; for example, trait-anxious people have difficulty disengaging from threatening stimuli once they are detected (Fox, Russo, & Dutton, 2002).

23

Finally, the executive control function allows us to determine which objects are selected for further processing (e.g. for categorization or for a response). In executive control tasks, participants must generate and maintain a task set in order to resolve conflict between competing stimuli or response tendencies (Egner, 2008). The executive control function relies critically on the anterior cingulate cortex (ACC) in the brain (Botvinick, Braver, Barch, Carter, & Cohen, 2001; Botvinick, Cohen, & Carter, 2004; Bush, Luu, & Posner, 2000; Egner, 2008; Fan, et al., 2005; Weissman, Gopalakrishnan, Hazlett, & Woldorff, 2005; Woldorff, et al., 2004). Although the functions of alerting and orienting will be discussed to some extent in this dissertation, the principal focus will be on the topic of executive control and visual selection. The attentional “spotlight” can provide a useful analogy for the scope of visual selection (Posner, Snyder, & Davidson, 1980), though there is evidence that this analogy may obscure some important aspects of attention (Driver & Baylis, 1989). A strong version of this analogy implies that all of the information within the spotlight beam is selected and all of the information outside of the beam is discarded (Cave & Bichot, 1999). Weaker versions would allow for some information from outside the “beam” to be processed as well, i.e. a “leaky” attentional filter (Miller, 1991; Schmidt & Dark, 1998), or allow for a selection gradient moving outward from the center of the beam (e.g. Andersen & Kramer, 1993; LaBerge & Brown, 1989). Additionally, the spotlight might show an imperfect amount of focus and sometimes Although there are attentional resources that operate in parallel across the display for features such as motion and color (Saenz, Buracas, & Boynton, 2003), the spotlight represents a prioritization of limited processing capacity to a particular location (C.W. Eriksen & St. James, 1986; C. W. Eriksen & Yeh, 1985). At the neural level, selection in the spotlight involves both excitation of the

Full document contains 172 pages
Abstract: Emotion interacts with numerous areas of cognition, including memory, decision-making, and attention (Dolan, 2002). In this dissertation, I conduct four experiments investigating the role of emotion on selective visual attention. In Chapter 1, I describe basic emotion research and the "scope" of selective attention. I outline previous research on the effects of happiness and anxiety on selective attention, some potential mechanisms of these effects, and their practical significance. In Chapter 2, I present an experiment building on prior findings that positive moods broaden the scope of selective attention (Rowe, Hirsh, & Anderson, 2007). I also present advantages and disadvantages of methods to induce and assess emotional states. In my experiment, participants showed greater flanker interference in a happy mood and less flanker interference in an anxious mood, indicating broadened and narrowed attentional scopes, respectively. In Chapter 3, I present two experiments exploring the role of mood in featural selection. I outline evidence for a featural scope of attention and changes to the flanker task to incorporate featural selection. When in an anxious mood, participants showed reduced flanker interference when a feature distinguished the target from the flankers, but more interference when the flankers shared the feature. In contrast, in a positive mood, these participants showed relatively non-selective featural attention. A follow-up experiment found that this effect was modified by task context. In Chapter 4, I examine mood influences in a visual search task similar to the flanker task. I outline the importance of visual search and reasons why mood states might influence it. In this search task, no reliable mood effects were found on visual search efficiency, except for a potential between-subjects effect. I interpret these findings in light of attentional theory and potential issues with the search task. In Chapter 5, I present overall conclusions and limitations, as well as how this research fits into the existing literature. Overall, participants. spatial and featural selection is broadened by happy moods and narrowed by anxiety, but this may not influence preattentive processing in visual search. I conclude with areas of improvement for future studies and research directions based on this work.