TOP-DOWN INFLUENCES ON SELECTIVE ATTENTION

ACROSS THE EXTENDED VISUAL FIELD

by

Jing Feng

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Graduate Department of Psychology

University of Toronto

© Copyright by Jing Feng 2011

ii

TOP-DOWN INFLUENCES ON SELECTIVE ATTENTION

ACROSS THE EXTENDED VISUAL FIELD

Jing Feng

Doctor of Philosophy

Graduate Department of Psychology

University of Toronto

2011

Abstract

The research focuses on the role of top-down influences on selective attention across the

attentional visual field. The attentional visual field is the subset of the visual field in which

attentional processes take place. The size of the attentional visual field is relatively large

compared to the areas considered by most empirical studies of visual attention to date. Three

possible forms of top-down influence are examined: 1) the expectation of the size of the area in

which the target is likely to occur; 2) the expectation of the direction in which the target is likely

to occur; and 3) existing unconscious bias in the spatial distribution of attention. Results from

Experiment 1 suggest that participants modify the size of the attended area according to their

expectation of the location of the target. Experiment 2 demonstrates that focus of attention can

be oriented toward the expected target direction. Experiment 3 reveals that, even when no

conscious control is involved, the distribution of attention is biased toward certain areas.

Theoretical considerations are discussed, including the introduction of a simple statistical model

to assist in conceptualizing the modifications of the distribution of attention over the attentional

visual field. Practical applications of the results are also discussed.

iii

Acknowledgements

I am very grateful to my supervisor, Dr. Ian Spence. With his enthusiasm, inspiration,

encouragement and a good sense of humour, Ian provided me great guidance and support in

research. Ian is truly a super supervisor. It is difficult to overstate my gratitude to him.

I would also like to sincerely thank the rest of my thesis committee members, Dr.

Jennifer Ryan and Dr. Matthias Niemeier, for their encouragement, insightful comments and

suggestions on my research.

I would further like to express my thanks to all faculty members who taught me and led

me to the fascinating world of psychological science during my undergraduate and graduate

studies. I would like to particularly thank Dr. Jay Pratt and Dr. Morris Moscovitch, whose

guidance shaped my interest in research.

I wish to thank my family: my husband, Liang Song, and my son, Alex. Alex is the most

precious gift. He has brought so much joy and meaning to my life.

Lastly, and most importantly, I want to thank my parents, my dad Baiwan Feng and my

mum Meifang Zhou. They gave birth to me, raised me, taught me, supported me, and always

loved me. They are my heroes. To them, I dedicate this thesis.

iv

Table of Contents

Title i

Abstract ii

Acknowledgements iii

Table of Contents iv

List of Figures vii

Chapter 1: Motivation and Overview 1

Motivation for the research 1

Overview of the thesis 1

Chapter 2: Introduction 3

The visual field 3

Visual selective attention 4

The attentional visual field 4

The selection function 6

Selecting spatial locations 7

Models of attention 8

Top-down influences 10

Chapter 3: The Distribution of Attention 14

Chapter 4: Experiments 16

Experiment 1 16

Experiment 1A 17

Method 17

Results 19

Discussion 20

v

Experiment 1B 21

Method 21

Results 22

Discussion 23

General discussion 23

Experiment 2 24

Experiment 2A 25

Method 26

Results 27

Discussion 28

Experiment 2B 29

Method 29

Results 30

Discussion 31

General discussion 32

Experiment 3 33

Method 34

Results 36

Discussion 39

Chapter 5: Discussion 41

Top-down influence in the attentional visual field 41

Expectation of the size of the area 41

Expectation of the target direction 43

Spatial bias in the distribution of attention 45

vi

New findings from investigating attention across the extended visual field 50

Neuropsychological basis of top-down influence 51

The distribution of attention 53

Practical applications of research on attention across the visual field 54

Chapter 6: Contributions 57

References 59

Figures 75

vii

List of Figures

Figure 1. Distribution of attention as a bivariate Gaussian

Figure 2. Illustration of how top-down influences affect the distribution

Figure 3. The attentional visual field (AVF) task (area size cue by trial) in Experiment 1A

Figure 4. Performance on the AVF task (area size cue by trial) in Experiment 1A

Figure 5. The AVF task (area size cue by block) in Experiment 1B

Figure 6. Performance on the AVF task (area size cue by block) in Experiment 1B

Figure 7. The AVF task (target direction cue) in Experiment 2A

Figure 8. Performance on the AVF task (target direction cue) in Experiment 2A

Figure 9. The AVF identification task (target direction cue) in Experiment 2B

Figure 10. Performance on the AVF identification task (target direction cue) in Experiment 2B

Figure 11. The AVF task (no cue, various exposures) in Experiment 3

Figure 12. Analysis of spatial bias in Experiment 3

Figure 13. Performance on the AVF task (no cue, various exposures) in Experiment 3 (in left vs.

right analysis)

Figure 14. Performance on the AVF task (no cue, various exposures) in Experiment 3 (in upper

vs. lower analysis)

1

Chapter 1: Motivation and Overview

Motivation for the Research

Most previous research in visual attention has been confined to a relatively small

central area of the visual field. For instance, among the 50 research reports randomly selected

from the reference list in a recent review on attention (Chun, Golomb & Turk-Browne, 2011),

only three reports claimed to have presented the stimulus at a visual angle greater than 15o

wide. As a result, our knowledge of how attention functions is limited to a small central area

of the visual field. Thus the operation of attentional processing in the much larger remaining

area of the visual field is relatively unexplored. Without consideration of the entire visual

field, our understanding of how attention is deployed is necessarily incomplete. This has

important consequences for the application of existing knowledge to many practical problems

(e.g. interface design for aircraft cockpit instrumentation where information is presented

across the extended visual field; see Boot, Kramer & Becic, 2007).

Selective attention over the visual field is affected by both bottom-up processes (inputs

from the visual environment) and top-down influences (effects that inhere in the participant).

Bottom-up processes have been examined from a computational perspective (Itti & Koch,

2001). In contrast, investigations on top-down influences across the visual field are scarce.

This thesis examines the deployment of attention over the visual field with an emphasis on

top-down influences that modify the distribution of attention.

Overview of the Thesis

Various forms of top-down influence can alter the deployment of attention across the

visual field. These include: 1) the expectation of the size of the area in which the target is likely

to occur; 2) the expectation of the likely direction of the target; and 3) existing unconscious bias

in the spatial distribution of attention. The operation of each of these forms of top-down

2

influence over the visual field was examined by experiment:

1) Experiment 1 investigated how the expectation of the size of the attended area

modifies the distribution of attention;

2) Experiment 2 examined how the expectation of the direction of the target modifies the

distribution of attention;

3) Experiment 3 explored possible unconscious bias in the spatial distribution of selective

attention.

The results demonstrated that top-down influences modify the distribution of attention

over the visual field in various ways. A participant can: 1) consciously control the size of the

area in which attention is deployed; 2) consciously orient attention in a particular direction;

and 3) unconsciously favour certain regions in the visual field thus demonstrating existing

bias in the distribution of attention. The results confirm that top-down influences modify the

distribution of attention over the visual field and possible mechanisms are considered.

Potential applications and directions for future research are discussed.

3

Chapter 2: Introduction

The Visual Field

The visual system must deal with a vast amount of information. Each second, around 10

billion bits of information arrive at the retina (Raichle, 2010). Of this information, only 10,000

bits arrive for processing at the level of the visual cortex. Remarkably, most individuals have no

problem completing activities that require processing and distilling the large amount of

information that comes from all areas of the visual field. For example, when driving through an

intersection, a driver must identify traffic signs and lights and also be aware of vehicles and

pedestrians simultaneously moving in several directions. Superior driving performance relies on

the ability to select the important information (and ignore the unimportant) wherever it occurs in

the visual field. Failure to pick out important targets among distracting information can cause

problems. For example, older adults are generally less able to identify important events in a

cluttered visual environment and this decline in selective attention can lead to poorer driving

performance (Ball, Owsley, Sloane, Roenker & Eruni, 1993; Ball, Roenker & Bruni, 1990;

Bedard et al., 2006) and higher risks of falls (Broman et al., 2004; Lajoie, Teasdale, Bard &

Fleury, 1996; Owsley & McGwin, 2004).

Other tasks also rely on our ability to process information over the visual field to achieve

superior performance. For instance, during spatial navigation, blocking a participant’s peripheral

vision leads to severe impairment in wayfinding (Fortenbaugh, Hicks, Hao & Turano, 2006). In

yet another context, computer users reported greater comfort and satisfaction when using larger

displays (Robertson et al., 2005); they also worked faster and more accurately with larger

displays (Colvin, Tobler & Anderson, 2004; Czerwinski et al., 2003). Indeed, the ability to

select and process critical information over the visual field is necessary to complete many

important perceptual and cognitive tasks.

4

Visual Selective Attention

Visual selective attention is the ability to concentrate on task relevant information in the

visual field while ignoring other information that is irrelevant or disruptive (e.g. Deutsch &

Deutsch, 1963; Johnson & Proctor, 2004). Attention assigns a greater proportion of its limited

resource to selected sensory inputs. The selection function is mediated by both top-down

influences and bottom-up processes. Top-down influences refer to those effects that are inherent

in the participant. For example, when looking for a cup, a participant may deploy more attention

to the kitchen table, rather than the kitchen chairs, because prior knowledge suggests that the

table is a more likely location for a cup. In contrast, bottom-up processes are precipitated by

events in the visual environment. For example, when there is a salient change in the visual

periphery, the event may capture a participant’s attention (Lambert, Spencer & Mohindra, 1987).

Selective attention is a very important component of visual cognition. The selection

function operates from the early stages of information processing until the later stages; the

selection function alters perception, information integration and response choice (Johnson &

Proctor, 2004). Among the vast quantity of visual information received at any moment, attention

selects the subset that is most crucial to the current task and discards the rest, thus preventing

overload. The selection function also integrates visual features to form a representation of the

attended object (Treisman & Gelade, 1980). In addition, the selection function is involved in

choosing the most appropriate action (Gibson, 1941; Neumann, 1987).

The attentional visual field.

The visual field is the fraction of the world that is seen within one fixation. Similarly, the

attentional visual field is that fraction of the visual field within which attention may be deployed.

This area may be as large as the visual field but, under most conditions, it is smaller. Thus the

attentional visual field is the spatial coverage of visual selective attention within one fixation.

5

The attentional visual field may be measured using one of several possible attentional visual

field (AVF) tasks. In a typical AVF task, a participant’s ability to detect, localize and identify a

target among distractors at any arbitrary location within a wide visual field is assessed. In

general, performance declines as a function of eccentricity from fixation (Ball, Beard, Roenker,

Miller & Griggs, 1988; Ball et al., 1993). A plot of performance over a number of widely

distributed locations reveals a unique area, namely the attentional visual field. The boundary of

this area may be defined as the locus of points where performance has fallen to a chance level.

Visual information outside this area is not processed by visual selective attention. Thus an AVF

task provides a method for mapping the domain of visual selective attention.

The attentional visual field, as measured by an AVF task, is dynamic; it changes in size

and shape according to a number of factors, such as foveal processing demands (Ikeda and

Takeuchi, 1975; Wickens & Hollands, 1999), number of targets (Chan, Courtney & Ma, 2002),

similarity between the target and distractors (Chan & Tang, 2007), participant’s age (Ball et al.,

1988; Clay et al., 2005), gender (Feng, Spence & Pratt, 2007), anxiety level (Shapiro & Lim,

1989), and training (Ball et al., 1988; Feng et al., 2007; Green & Bavelier, 2003). Because the

size of the attentional visual field has significant impact on visual search performance (Chan &

So, 2007), the attentional visual field size has been found to be a reliable predictor of

performance in a variety of activities, including collisions with obstacles during walking

(Broman et al., 2004), self-reported falls (Owsley & McGwin, 2004), on-road driving (Clay et

al., 2005; Myers, Ball, Kalina, Roth & Goode, 2000; Wood & Troutbeck, 1995), simulated

driving and self-reported crashes (Clay et al., 2005), reading food labels and even using

screwdrivers (Owsley, McGwin, Sloane, Stalvey & Wells, 2001).

6

The selection function.

Selection is an important function often used to measure attention (e.g. Posner, 1980;

Posner & Cohen, 1984; Johnson & Proctor, 2004). “Research on selection, perhaps more than

any other aspect of attentional processing, has dictated how attention is characterized” (Johnson

& Proctor, 2004, p. 93). Studies on the selection function have focused on three aspects: why

selection is needed, what is selected and how selection is achieved.

Why selection is necessary?

Although there are 10 billion bits of information registered at the retina each second,

only 10,000 bits of information may arrive at the visual cortex (Raichle, 2010). These 10,000

bits are processed further, while the remaining information is not. These processes involve

selection, which is a critical function that defines attention (e.g. Posner & Cohen, 1984; Johnson

& Proctor, 2004). Attentional selection is the process of identifying important information for

the current task and assigning processing priority to this information.

What is selected?

It is not generally agreed whether attention selects objects (e.g. Baylis & Driver, 1992;

Bundesen, 1990; Bundesen, Pedersen & Larsen, 1984) or spatial locations (e.g. Cave & Pashler,

1995; Eriksen & Yeh, 1985; LaBerge & Brown, 1989; Posner, 1980). In a target identification

task (Moore, Yantis & Vaughan, 1998), although only one location inside a large object was

cued, participants responded faster to targets presented at any location within the object

compared to locations outside the object. This was interpreted as evidence for deploying

attention onto the object containing the cued location, rather than the location itself. However,

findings from many other studies have supported the notion that attention is deployed to spatial

locations. For example, a spatial cue can lead to benefits in performance when it is valid, or

impairment when it is invalid, compared to baseline performance when the cue is uninformative

7

(Cave & Pashler, 1995; LaBerge & Brown, 1989; Posner, 1978). When a cue is presented,

attention selects an area around the cue (Eriksen & Yeh, 1985; LaBerge & Brown, 1999). The

size of the selected area may vary according to the task demand (Eriksen & Yeh, 1985). The

minimal size seems to be around one degree of visual angle, within which a distractor cannot be

rejected (Eriksen & Yeh, 1985).

How selection is achieved?

Selection involves both enhancement and inhibition mechanisms. Attentional selection

enhances processing of some information (Kimberg & Farah, 2000), while inhibiting processing

of other information (Vandenberghe et al., 1997, 2000). Whether a piece of information is

selected or not is determined by both the bottom-up saliency of the information and top-down

influences. In a conceptualization of the enhancement and inhibition mechanisms (Johnson &

Proctor, 2004, p.71), enhancement and inhibition are described as increasing attentional

“weights” for selected information and decreasing attentional “weights” for unselected

information.

Selecting spatial locations.

As “the primary visual property” (Pylyshyn, 1999, p. 360), spatial location is probably

more important than other features of an object for attentional selection (Johnson & Proctor,

2004; Van der Heijiden, 1993). Although there is evidence to show that features like color and

shape can provide the basis for effective selection, it is possible that selection based on these

features operates, at least partially, through selection of the spatial location(s). For example,

although knowing the color of the target in advance facilitates target identification, this

facilitation is due to the selection of the location of the target and its adjacent locations, rather

than the color (Cepeda, Cave, Bichot & Kim, 1998; Luck, Fan & Hillyard, 1993; Shih &

Sperling, 1996; Tsal & Lamy, 2000; Tsal & Lavie, 1993). Tsal and Lamy (2000) compared the

8

benefit when the target appeared close to the previously attended object or when the target

shared the same color with the previously attended object. Being close produced much greater

facilitation. Similarly, Luck and his colleagues (1993) observed better performance when the

target occurred at the same location as in the previous trial. Enhanced activity in an early visual

component was also found when event-related potentials (ERP) were measured, suggesting that

the location was selected (Luck et al., 1993).

Some studies even found selection based solely on spatial location but not other object

features (e.g. Posner, Snyder & Davidson, 1980; Theeuwes, 1989). In Posner et al. (1980),

participants were faster at identifying the target letter when its location was known, but not

when the letter itself was known. Theeuwes (1989) also found faster response when information

regarding the target location was provided, but not when shape information was provided.

However, a possible criticism is that object features like shape and identity were too similar to

be distinguished (e.g. Bundesen, 1990; Bundesen et al., 1984). On the other hand, one can still

argue that spatial location may be more important than other object features for attentional

selection, especially when other object features were not distinctive enough to serve as effective

cues.

In the visual periphery, visual acuity and color vision are greatly degraded (Anstis, 1998;

Hansen, Pracejus & Gegenfurtner, 2009). Therefore, the color and shape of an object become

much less useful cues at large eccentricities and selection based on spatial location becomes

even more important.

Models of attention.

One of the most popular descriptions of visual selective attention is the spotlight model

(Eriksen & Hoffman, 1973). It views attention as a spotlight that projects attention onto selected

locations in the visual field. The size of the spotlight varies according to the task demand

9

(Eriksen & St. James, 1986). Because the overall amount of the attentional resource is limited,

the attentional intensity at locations in the selected area also changes with the size of the area

(Eriksen & St. James, 1986). This limited resource constraint was incorporated into the model as

the zoom-lens supplementary (Eriksen & St. James, 1986), and is usually referred to as the

zoom-lens model.

The distribution of attention has also been described in terms of a gradient of the

attentional resource around the focus of attention (Downing & Pinker, 1985; Eriksen & Yeh,

1985; Palmer, 1990). This conceptualization dates back to Baldwin (1889) and James (1890),

who suggested that visual attention has a mountain-shaped distribution, with greatest

concentration at the center, gradually falling off toward the periphery. The concept of a gradient

has been supported by experimental evidence (LaBerge & Brown, 1989; Müller, Mollenhauer,

Rösler & Kleinschmidt, 2005).

Both the spotlight and gradient models were developed on the basis of experimental

findings from a relatively small area in the center of the visual field. However, visual processing

at large eccentricities is very different from that in the center (Palmer & Rosa, 2006; Previc,

Beer, Liotti, Blakemore & Fox, 2000; Wickens & Hollands, 1999) and, as a result, these models

may not generalize to the extended visual field.

One model, however, may fare better in the extended visual field. The saliency model

(Itti & Koch, 2001; Koch & Ullman, 1985) is a computational model of selective attention, built

primarily to mimic human attentional process when viewing natural scenes. According to this

model, a visual scene is analyzed using feature detection, integration and competition for

saliency. A “saliency map” is formed and it serves as a guide for attentional deployment.

However, the saliency model describes attentional selection based only on bottom-up processes

(Itti & Koch, 2001), but does not consider top-down influences.

10

Top-down influences.

Top-down influences play a significant role in attentional selection. These influences are

powerful and operate at very early stages of processing, often even before the stimulus is

present (Mangun, Buonocore, Girelli & Jha, 1998). Top-down influences affect attentional

selection by enhancing the neural response at some locations, thus biasing competition

favourably toward stimuli at these locations (Kastner & Ungerleider, 2001). This enhancement,

as measured by electroencephalogram (EEG) and single cell response, has been seen at both

lower and higher levels of the visual cortex including V1 (Motter, 1993), V2 (Luck, Chelazzi,

Hillyard & Desimone, 1997; Motter, 1993) and V4 (Connor, Gallant, Preddie & Van Essen,

1996; Connor, Preddie, Gallant & Van Essen, 1997; Motter, 1993).

Although top-down influences have been described as goal-directed control of attention,

the definition is broader than that. Top-down influences include any effect on attentional

selection that is initiated by the participant, as opposed to effects initiated by events in the visual

environment. Goal-directed control of attention is merely one form of top-down influence which

is consciously managed by the participant and can be modified constantly according to the goal.

For example, when a participant knows that the target is highly likely to occur at a particular

location in the visual field, the distribution of attention is then biased toward the location (e.g.

Posner, 1978, 1980; Posner et al., 1980). Another potential source of top-down influence, that is

not specifically goal-directed, is spatial bias (e.g. Dickinson & Intraub, 2009; Jewell & McCourt,

2000). Such bias is not likely under conscious control; it is not modified without extensive

training; and it is generally the consequence of relatively stable neurophysiological

characteristics of the participant. These are the product of genetics or learning, or a combination

of both.

11

Goal-directed control modulates attentional selection (e.g. Heinze, Luck, Mangun &

Hillyard, 1990; Posner et al., 1980). By guiding attention to sensory inputs that are relevant to

the goal, we are able to stay focused on a task until completion. For instance, when we are

talking to a friend in a club, we are able to focus on the conversation, rather than being diverted

by the music or the flickering lights. Without goal-directed control, we would be distracted by

salient events in the environment. The ability to control selection is weak in young children and

they are more vulnerable to distractors (Wendelken, Baym, Gazzaley & Bunge, 2011; Ruff &

Lawson, 1990; Ruff & Capozzoli, 2003). This ability to control selection develops as an

individual grows up (Ruff & Capozzoli, 2003), thus adults are better at staying focused on

important information in complex task situations. This ability to control selection also declines

with aging, as older adults become more vulnerable to distractors (Gazzaley, Clapp, McEvoy,

Knight & D’Esposito, 2008).

Goal-directed control often occurs when a participant holds expectations regarding

certain aspects of the target. There are at least two possible forms of such goal-directed control

that depend on either an expectation regarding: 1) the size of the area in which the target is

likely to occur, or 2) the likely direction of the target. Thus a participant may modify the

attended area to conform to these expectations.

Expectation of the size of the attended area.

The idea of variability in the attentional area was described by Titchener (1908) in his

Law of Two Levels. He proposed that the size of the attended area can be modified by the

participant depending on the task condition. A participant can choose to concentrate attention

with a narrow focus, or distribute it across a relatively large visual area. A similar idea was

incorporated in the zoom-lens model of selective attention (Eriksen & St. James, 1986).

According to the zoom-lens model, the size of the attended area varies with the demands of the

12

task and the attentional intensity within the focus area decreases with increasing size of the

focus.

Expectation of the likely direction.

Expectation of the target direction is another source of goal-directed control. When an

expectancy of direction is formed, the participant orients attention in the expected direction.

This is described as endogenous orienting of attention, since it is triggered by the participant’s

intention. When the attentional focus is oriented away from the fixation without an eye

movement, the orienting is referred to as covert orienting (Wright & Ward, 2008). At small

eccentricities, endogenous covert orienting of attention is known to enhance attentional

performance (e.g. Posner, 1978, 1980); and the enhancement increases with a longer interval

between the onset of a valid direction cue and the appearance of the stimulus (Posner, 1980). It

is possible that at larger eccentricities endogenous covert orienting of attention has an even

more significant effect, given the increased importance of spatial attributes (relative to other

attributes such as color) for information processing in the periphery. In fact, Shepherd & Müller

(1989) showed that endogenous covert orienting of attention enhanced target detection at

eccentricities of 10o and 20

o. There was a trend of greater enhancement of target detection at the

larger eccentricity (Figure 1 and Table 1, Shepherd & Müller, 1989, pp.148–149, no statistical

analysis was provided).

Spatial bias.

Unconscious bias in the spatial distribution of attention may be another source of top-

down influence. Spatial bias in the performance of a variety of perceptual and cognitive tasks

has been shown in a number of studies. For example, Dickinson & Intraub (2009) found that

participants were more likely to make their first eye movement toward the left when viewing a

picture freely; also, their memory was better for the left-side objects. The researchers attributed

13

these findings to a possible leftward bias in the spatial distribution of attention. There is also a

general leftward bias when bisecting a line (Bowers & Heilman, 1980; Bradshaw, Nettleton,

Nathan & Wilson, 1983, 1985; Bradshaw, Nettleton, Wilson & Bradshaw, 1987). In a

qualitative review of 73 studies, Jewell & McCourt (2000) suggested that bias in the spatial

distribution of attention may be one of the underlying causes for the bias in line bisection

performance. Although a leftward bias was dominant in most studies (as reviewed in Jewell &

McCourt, 2000), significant individual differences in the direction and magnitude of bias were

found in some line bisection studies (Manning, Halligan & Marshall, 1990; McCourt & Olafson,

1997). For example, some individuals consistently demonstrated a leftward bias, while others

showed a rightward bias (Manning, Halligan & Marshall, 1990). Men in general demonstrated a

stronger bias than women in the line bisection task (Jewell & McCourt, 2000; McCourt &

Olafson, 1997). These individual differences suggest that there may be great individual

variability in the bias of the distribution of attention over the attentional visual field.

14

Chapter 3: The Distribution of Attention

A simple statistical model describing the spatial distribution of attention across the

visual field is proposed in this chapter. The model views the distribution of attention as a

bivariate probability distribution over the visual field (Figure 1). The probability density at any

particular location in the visual field represents the attentional intensity corresponding to that

location. In general, the attentional intensity decreases with an increase in the distance from the

fixation. The volume under the distribution represents the total attentional resource and remains

constant except for long-term change factors (e.g. aging, training). Note that the distribution

may not be symmetrical. Its size and shape may vary among individuals (Feng et al., 2007), and

may also change within an individual by focusing attention (e.g. focused vs. diffused mode,

Eriksen & St. James, 1986; Titchener, 1908), shifting attention (Wright & Ward, 2008), or by

aging (Ball et al., 1988; Clay et al., 2005) and training (Ball et al., 1988; Feng et al., 2007;

Green & Bavelier, 2003).

The model assumes that the distribution of attention is continuous in space and time. The

continuity in space implies that the attentional intensity at any given location within the attended

area is greater than zero. In addition, the focus of attention (often the fixation point, but can be

other locations when there are multiple foci) is associated with higher attentional intensity than

surrounding locations. Continuity in time suggests that the distribution of attention may change

to reflect not only the current attentional process, but also the history of previous processes

(LaBerge & Brown, 1989; LaBerge, Carlson, Williams & Bunny, 1997). The distribution of

attention is modified constantly. And this modification is time consuming (Eriksen & St. James,

1986; Shepherd & Müller, 1989).

Top-down influences can modify the distribution of attention, and consequently affect

the attentional intensity at any given location across the visual field. These modifications

15

include varying the spatial extent of the distribution, increasing the attentional intensity at

particular locations or along particular directions by changing the shape of the distribution. Also

existing spatial bias in the observer will help determine the shape of the distribution of attention.

A preliminary description of some of the top-down influences that can affect the distribution is

as follows:

1) The expectation of the size of the area in which the target is likely to occur

The size of the area of the distribution of attention is modified according to expectation.

When a participant narrows the attended area, the spread of the distribution becomes

smaller (Figure 2a). Because the attentional resource is assumed to be fixed, the

attentional intensity consequently increases close to fixation and decreases at points

distant from fixation. Thus, attentional performance is enhanced around fixation and

impaired in the periphery. In contrast, when a participant enlarges the attended area, the

spread of the distribution becomes greater (Figure 2b) and the attentional intensity

decreases at locations close to fixation and intensifies in the periphery.

2) The expectation of the likely direction of the target

An expectation of the direction of the target induces covert orienting of attention (Ponser,

1980). When attention is covertly oriented, the attentional intensity increases along that

direction (Figure 2c). Meanwhile, the attentional intensity decreases in other directions,

particularly along the opposite direction. As a result, performance along the expected

direction is enhanced, while performance along other directions is impaired.

3) Spatial bias in the distribution of attention

The attentional resource may not be symmetrically distributed across the visual field

even in the absence of cueing. For example, if there is a rightward bias, the distribution

of attention elongates toward the right (Figure 2d).

16

Chapter 4: Experiments

Three experiments examined top-down influences on selective attention across the visual

field. Experiment 1 investigated how the participant’s expectation of the size of the area in

which the target is likely to occur modifies the distribution of attention. A cue indicating the size

of the area where the target would likely appear was presented before the AVF task. Experiment

2 examined how the participant’s expectation of the likely direction of the target modifies the

distribution of attention. A cue indicating the likely direction of the target was presented before

the AVF task. Experiment 3 explored possible existing bias in the distribution of attention

across the visual field.

Experiment 1

This experiment investigated the influence of the participant’s expectation of the size of

the area in which the target was likely to occur. The AVF task was used to measure the

distribution of attention within a relatively large visual area (63.1o in diameter). Information

regarding the size of the area in which the target was likely to occur was given in two ways. In

Experiment 1A, on each trial, a cue indicating the likely size was presented at the center of the

display before the AVF stimulus. Trials with various expected sizes were randomized. In

Experiment 1B, the likely size was cued before each block of trials. Each block contained trials

with the same expected size. Thus these two experiments differed in the time available to the

participant to make use of the cue which indicated the size of the area in which the target would

likely appear. In Experiment 1A, this time was much shorter as the expected area size varied

from trial to trial. Since changing the size of the attended area takes time (Eriksen & St. James,

1986), the magnitude of influence from the two types of instructions on the distribution of

attention may differ.

17

Experiment 1A.

On each trial, a cue indicating the size of the area in which the target was likely to occur

was presented at the center of the display before the stimulus appeared. Participants were

instructed to make use of the cue (80% valid) and perform as well as possible on the AVF task.

The percentage of valid cues was set high to encourage participants to use the cue (Jonides,

1981; Kröse & Julesz, 1989; Wright & Ward, 2008).

Method.

Participants. 15 undergraduates at the University of Toronto (five males, ten females;

age range: 17-22 years) participated for course credit.

Task. The AVF task was used to examine the distribution of attention. In addition, a cue

was presented after the fixation and before the stimulus on each trial to induce an expectation of

the size of the area in which the target was likely to occur.

Design. The experimental design was a within-subject 3×3 repeated-measures. The

factors were expected area size (small/medium/large) and target eccentricity (10o/20

o/30

o).

Stimuli. In the AVF task (Figure 3), the stimuli were presented in a circular area (63.1o

diameter) centered on a uniform light-gray screen. Each trial began with a centered, unfilled

fixation square with a dark-gray border (3o×3

o). The fixation square was presented for 800 ms

and then replaced by the area size cue. The small size cue was a dark-gray unfilled circle

(2.2o×2.2

o); the medium size cue was two dark-gray concentric unfilled circles (3.6

o×3.6

o); and

the large size cue was three dark-gray concentric unfilled circles (4.5o×4.5

o). On each trial, a

size cue (small or medium or large, randomized) was displayed for 500 ms followed by a short

interval of 300 ms blank display. Then the stimulus display appeared. It consisted of 23 identical

distractors and one target, each uniquely localized at an eccentricity of 10o, 20

o, or 30

o in one of

eight equally spaced directions. The location of the target was randomly selected on each trial,

18

subject to the restriction that the target appeared an equal number of times in each possible

location. The target was a dark-gray filled square (1.5o×1.5

o) surrounded by an unfilled circle

with a dark-gray circumference (3o×3

o). The distractors squares were unfilled squares with dark-

gray borders (3o×3

o), identical to the fixation square. The stimulus display was presented for 30

ms, followed by a mask of randomly oriented lines for 200 ms. Participants indicated the

direction of the target after the mask disappeared. The next trial started 1000 ms after a response

was made.

Procedure. Participants were first asked to report their gender and age. Then they

completed the AVF task; the size of the area in which the target was likely to occur was cued.

The meanings of the size cues were explained: a small size cue indicating that the target was

likely to occur only at an eccentricity of 10o; a medium size cue indicating that the target was

likely to occur at an eccentricity of either 10o or 20

o; and a large size cue indicating that the

target was likely to occur at any eccentricity: 10o, 20

o or 30

o. Participants were required to

position their head on a chin rest at a distance of 35cm from the display. A practice session,

consisting 36 trials of randomized conditions, was required before the experimental session in

order to ensure that participants understood the task. In the experimental session, 720 trials were

presented in random sequence with the target appearing 30 times in each of the 24 locations.

80% of the trials were valid, in which the cued size was equal to or larger than the eccentricity

of presented target. Participants were given a two-minute rest after each block of 120 trials.

Responses from participants indicating the directions of the target on each trial and the reaction

times (RT) were recorded.

19

Results.

To examine the effect of the size cue, a 3×3 (cued size: small/medium/large, target

eccentricity: 10o/20

o/30

o) repeated-measures ANOVA was used to analyze the accuracy and RT

data.

The difficulty of increasing, decreasing, or keeping the to-be-attended area constant from

one trial to the next was examined using a single-factor (area change: increased, no-change,

decreased) repeated-measures ANOVA. To control for other factors (e.g. eccentricity, cued size)

that may have an influence on accuracy and RT, only valid trials with medium size cues were

included in this analysis. A trial with a medium sized cue that was preceded by a trial with a

large cue was an instance of the decreased condition. In the decreased condition, participants

had to reduce the size of the to-be-attended area from the large size on the previous trial to the

medium size for the current trial. Similarly, if the current trial with a medium size cue was

preceded by another trial with a medium cue, this was an instance of the no-change condition. A

trial preceded by a trial with a small cue was an instance of the increased condition.

Accuracy. Target detection was significantly poorer with an increase in target

eccentricity (10o: 82%, 20

o: 70%, 30

o: 49%) (Figure 4, left panel), F(2,28) = 50.34, p < .01.

Varying the size of the cued area did not change overall performance (small cue: 66%, medium

cue: 68%, large cue: 68%), F(2,28) = .40, p = .67; however, the interaction between cue and

eccentricity was significant (Figure 4, left panel), indicating that the distribution of attention was

modified according to the expected area size , F(4,56) = 5.34, p < .01.

In the inter-trial comparison, no differences in accuracy were found among the three

conditions (decreased: 74%, no-change: 74%, increased: 76%), F(2,28) = .97, p = .39.

RT. Responding was slower with an increase in target eccentricity (10o: 429 ms, 20

o: 484

ms, 30o: 580 ms) (Figure 4, right panel), F(2,28) = 8.68, p < .01. The interaction between cue

20

and eccentricity was also significant (Figure 4, right panel), F(4,56) = 4.68, p < .01. Slower

responding was associated with lower accuracy (Figure 4), suggesting that there was no speed-

accuracy trade-off.

In the inter-trial comparison, the differences in speed was almost significant (decreased:

551 ms, no-change: 504 ms, increased: 524 ms), F(2,28) = 3.08, p = .06. In particular, compared

to the no-change condition, responding was much slower when participants had to decrease the

to-be-attended area, F(1,14) = 7.49, p < .05, but comparable when they had to increase the to-

be-attended area, F(1,14) = .73, p = .41. The data suggest that responding may be slightly slower

in the decreased condition compared to the increased condition, F(1,14) = 3.05, p = .10.

Discussion.

The ability to identify a target decreases with an increase in target eccentricity; this is

typically observed in an AVF task (Ball et al., 1988; Green & Bavelier, 2003; Feng et al., 2007).

The size of the attended area can be modified under voluntary control. When the expected area

for the occurrence of the target was small, attention was more concentrated at smaller

eccentricities; when the expected area was large, attention was spread across a larger area.

These findings confirmed the opinion of Titchener (1908) and Eriksen & St. James (1986) that

the distribution of attention can be modified, according to the participant’s expectation.

In this experiment, the cue that indicated the size of the area within which the target

would likely appear was presented at the beginning of each trial and the cue was randomly

selected on every trial. When the cue changed from the preceding trial, participants had to

modify the size of the to-be-attended area. The inter-trial change made the task more difficult,

particularly when the to-be-attended area had to be reduced. This implies that when the cued

area size changed from trial to trial, the participant may not have had sufficient time to form a

21

clear sense of the boundaries of the to-be-attended area. This uncertainty was also reported by

several participants when asked to comment during the debriefing.

Experiment 1B.

Experiment 1B differed from Experiment 1A by using the same size cue for each block

of trials. The cue was presented at the beginning of each block. The cued size was maintained

through the entire block of trials (80 or 72 trials, depending on the cue condition). Because the

expectation of the size of the area is formed at the beginning of each block, there is a longer

time available to participants to make use of the cue. Otherwise, the task was identical to that in

Experiment 1A.

Method.

Participants. 15 undergraduates at the University of Toronto (six males, nine females;

age range: 18-22 years), participated for course credit.

Stimuli. All settings were the same as in Experiment 1A, except the stimulus display

appeared immediately after the fixation (Figure 5b).The cue indicating the size of the area in

which the target was likely to occur was presented only at the beginning of each block before

the trials started (Figure 5a).

Design. The experimental design was a within-subject 3×3 repeated-measures design

with blocked trials. There was a between-block factor, cued area size (small/medium/large),

and a within-block factor, target eccentricity (10o/20

o/30

o). A block of trials of each cued area

size was repeated three times, resulting in nine blocks in total. The order of the nine blocks of

trials was counterbalanced using a Latin Square design, independently randomized for each

participant. Therefore, one round of blocks included three blocks, with one block of each of the

three cued area sizes, and there were three rounds of blocks in total.

22

Procedure. Only the order of conditions and presentation of the cue was different from

the procedure in Experiment 1A. 36 practice trials were grouped into three blocks, 12 trials for

each type of area size cues. Similarly, in the experiment session, trials with the same size cue

were blocked. There were 72 trials in each block with a large size cue. Because a large cue

would be valid for a target appearing at any of the three eccentricities, every trial in a block with

a large size cue was valid. The number of trials in the block, 72, was a multiple of the 24

possible locations of the target. In contrast, there were 80 trials in each block with a medium or

small cue. Since a small or medium size cue would be invalid for a target appearing at an

eccentricity outside the to-be-attended area (an eccentricity of 20o or 30

o with a small size cue or

an eccentricity of 30o with a medium size cue), blocks with medium or small size cues had both

valid and invalid trials. In each block with a small or medium size cue, 80% of the trials were

valid. Participants were given the size cue at the beginning of each block and were asked to

maintain the same expectation induced by this cue throughout the block. Accuracies and

reaction times during the AVF task were recorded.

Results.

Accuracy. Overall accuracy was significantly poorer with an increase in target

eccentricity (10o: 81%, 20

o: 68%, 30

o: 48%) (Figure 6, left panel), F(2,28) = 60.19, p < .01.

Overall accuracy varied with cue size (small cue: 62%, medium cue: 67%, large cue: 68%),

F(2,28) = 7.05, p < .01. The distribution of attention was modified by the size of the cued area.

This is indicated by the significant interaction between expected size and target eccentricity

(Figure 6, left panel), F(4,56) = 10.22, p < .01. In particular, accuracy at an eccentricity of 30o

increased significantly with a larger cued size (small cue: 38%, medium cue: 50%, large cue:

55%), F(2,28) = 14.79, p < . 01. Overall accuracy improved over rounds of blocks (first: 62%,

second: 67%, third: 68%), F(2,28) = 4.92, p < .05. This practise effect was significant between

23

the first and second rounds of blocks, F(1,14) = 6.65, p < .05. There was no change in the ability

of the size of the cued area to modify the distribution of attention, as indicated by an

insignificant three way interaction among round of blocks, expected size and target eccentricity,

F(8,112) = 1.01, p = .41.

RT. Responding was slower with an increase in the eccentricity of the target (10o: 499

ms, 20o: 538 ms, 30

o: 622 ms) (Figure 6, right panel), F(2,28) = 6.27, p < .01. The interaction

between cued size and eccentricity was significant (Figure 6, right panel), F(4,56) = 2.68, p

< .05. Responding became faster over rounds of blocks (first: 638 ms, second: 518 ms, third:

504 ms), F(2,28) = 6.64, p < .01, particularly between the first and second rounds of blocks,

F(1,14) = 9.94, p < .01. Slower responding was associated with lower accuracy (Figure 6),

suggesting that there was no speed-accuracy trade-off.

Discussion.

When the size cue was presented at the beginning of each block, the expectation of the

size of the area significantly altered the distribution of attention across the visual field. When

participants were expecting the target to appear within a larger area, accuracy increased slightly

at an eccentricity of 20o and quite significantly at an eccentricity of 30

o. Change at an

eccentricity of 10o was not obvious, probably because detecting the target was too easy at this

small eccentricity with an exposure of 30 ms. Indeed, accuracies at an eccentricity of 10o were

very high (> 80%).

General discussion.

The expectation of the size of the area in which the target is likely to appear affects the

spatial distribution of attention. Attention was more concentrated at small eccentricities when

participants expected a small area; while the distribution of attention was more spread out when

participants expected a large area. This modification of the spatial distribution of attention was

24

observed in Experiment 1A, where participants continuously changed their expectation of the

size of the to-be-attended area, in response to a changing cue, from trial to trial. In Experiment

1B, where participants did not have to change the expectation within each block of trials,

changes in the spatial distribution of attention according to the expectation were greater. The

expectation of the size of the area modified the spatial extent of the distribution of attention and

hence the attentional intensities at eccentricities of 20o and 30

o. Compared to complete

randomization (Experiment 1A), the blocking design (Experiment 1B) yielded a greater

modification of the distribution of attention. This is at variance with the findings of some

researchers (Downing & Pinker, 1985; Posner, 1978; Remington & Pierce, 1984). In

experiments with blocked trials, no differences were observed as the result of manipulations of

the spatial cues (Posner, 1978; Remington & Pierce, 1984); while similar experiments with

randomized trials yielded significant differences (Downing & Pinker, 1985). Thus Sperling &

Dosher (1986) suggested that blocking designs in spatial cueing paradigms may produce less

reliable data. Since the present experiment demonstrated a contrary result, it may be that the

conclusion of Sperling & Dosher (1986) does not apply in all cases.

Experiment 2

The influence of a cue indicating the direction of the target was examined. A cue

presented at the center of the display indicated the likely direction of the target. This directional

cue was highly valid (67% in Experiment 2A and 80% in Experiment 2B), leading the

participant to expect a particular target direction. In central areas, direction cues induce covert

orienting of attention toward the cued direction (e.g. Posner, 1978, 1980), benefiting

performance when the cue is valid, and impairing performance when the cue is invalid (e.g.

Posner, 1978, 1980). The effect of the cue is greater with a longer interval between the cue and

the stimulus (Müller & Findlay, 1988; Heinze et al., 1990). Similar effects from direction cues

25

on target detection have been demonstrated in the near periphery (Shepherd & Müller, 1989).

Shepherd & Müller (1989) examined the effect of direction cues at relatively large eccentricities

(10o and 20

o). In their study, participants detected a target in one of the four boxes horizontally

aligned with the center box. Each trial, a direction cue pointing to either the left or right was

given in the center box, indicating the likely direction of the target. Participants pressed a key as

soon as they saw a target appearing in one of the four boxes. In Shepherd & Müller (1989), no

distractor was presented along with the target, thus only detection but no discrimination or

identification of the target was involved. This experiment differs in that the distribution of

attention across the attentional visual field is defined by performance on a task which assesses a

participant’s ability to discriminate, localize and identify a target among distractors. In

Experiment 2, 11 distractors were presented simultaneously with the target, to test the effect of

endogenous direction cues on the ability to discriminate, localize and identify a target among

distractors.

Because properties like shape and color are less reliably discriminated at larger

eccentricities, expectation of the direction of the target may play a relatively more important

role. During covert orienting of attention, only the attentional focus but not the fixation is

shifted. To help ensure that the participant maintained fixation during each trial, the duration

from onset of the cue to offset of the stimulus was controlled within 200 ms. Thus the likelihood

of a saccade occurring during processing of the stimulus was low (Sparks, Rohrer & Zhang,

2000).

Experiment 2A.

The effect of the expectation of the direction of the target on the distribution of attention

was assessed. A symbolic cue (67% valid) indicating the likely direction of the target was

presented before the AVF stimulus appeared.

26

Method.

Participants. 14 undergraduates at the University of Toronto (five males, nine females;

age range: 18-21 years), participated for course credit.

Task. This experiment used the AVF task with an endogenous cue indicating the likely

direction of the target between the fixation and the stimulus display (Figure 7).

Design. The experimental design was a 3×3×2 within-subject repeated-measures, with

factors including target eccentricity (10o/20

o/30

o), cue-target interval (0/50/100 ms) and target

validity (valid/invalid).

Stimuli. In the AVF task, the stimuli were presented in a circular area (63.1o in diameter)

centered on a uniform light-gray screen. Each trial began with a centered, unfilled fixation

square with a dark-gray border (3o×3

o). The fixation square was presented for 800 ms and then

augmented by the directional cue, which was a dark-gray arrow (3o×3

o) centered on the screen.

The cue pointed to one of four directions (up, down, left, right) depending on the condition of

the particular trial. During each trial, the cue remained on screen for 80 ms and was followed by

a 0/50/100 ms blank display. Then the stimulus appeared, consisting of 11 identical distractors

and one target, each uniquely localized at an eccentricity of 10o, 20

o, or 30

o in one of four

equally spaced directions (up, down, left, right). The location of the target was randomly

selected on each trial, subject to the restriction that the target appeared an equal number of times

in each possible location. The target was a dark-gray filled square (1.5o×1.5

o) surrounded by an

unfilled circle with a dark-gray circumference (3o×3

o). The distractors were unfilled squares

with dark-gray borders (3o×3

o), identical to the fixation square. The stimulus display was

presented for 20 ms, followed by a mask of randomly oriented lines for 200 ms. Participants

indicated the direction of the target after the mask disappeared. The next trial started 800 ms

after a response was made.

27

Procedure. Participants were first asked to report their gender and age. Then they

completed the AVF task with directional cues. During the task, participants were required to

position their head on a chin rest at a test distance of 35cm from the screen. A practice session,

consisting 24 trials, was mandatory before the experimental session in order to ensure that

participants understood the task. In the experimental session, 648 trials were presented in a

random sequence with the target appearing 54 times in each of the 12 locations, for each cue-

target interval condition. 67% of the trials were valid, in which the cue arrow correctly pointed

to the target. Participants were instructed about this high validity of directional cues, to

encourage them to use the cue. In the invalid trials, the cue arrow pointed in one of the other

three directions. The arrow pointed (at random) in each incorrect direction with equal frequency.

Participants were given a two-minute rest after each set of 108 trials. Responses from

participants indicating the directions of the target and the reaction times on each trial were

recorded.

Results.

Accuracy. Target detection was significantly poorer with an increase in target

eccentricity (10o: 87%, 20

o: 78%, 30

o: 65%) (Figure 8a, three right panels), F(2,26) = 55.45, p

< .01. The mean accuracy of the valid trials was higher than that of the invalid trials (valid: 87%,

invalid: 66%) (Figure 8a), F(1,13) = 70.60, p< .01. The duration of the cue-target interval also

had a significant impact on the performance. With longer intervals, overall accuracy (including

both valid and invalid trials) became slightly worse at the longest interval (0 ms: 77%, 50 ms:

77%, 100 ms: 75%), F(2,26) = 3.48, p < .05. Although target detection in valid trials was not

enhanced at longer intervals (0 ms: 87%, 50 ms: 88%, 100 ms: 87%) (Figure 8a, left panel),

F(2,26) = .14, p = .87, accuracy was impaired in invalid trials at longer intervals (0 ms: 68%, 50

ms: 66%, 100 ms: 63%) (Figure 8a, left panel), F(2,26) = 6.59, p < .01. The disparity in

28

accuracy between valid and invalid trials increased with longer cue-target intervals (0 ms: 19%,

50 ms: 22%, 100 ms: 24%) (Figure 8a, left panel), F(2,26) = 3.96, p < .05, and also at larger

eccentricities (10o: 13%, 20

o: 22%, 30

o: 30%) (Figure 8a, three right panels), F(2,26) = 8.74, p

< .01.

RT. Participants responded fastest at the smallest eccentricity (10o: 297 ms, 20

o: 358 ms,

30o: 358 ms) (Figure 8b, three right panels), F(2,26) = 6.71, p < .01. They also responded faster

in valid trials than in invalid trials (valid: 297 ms, invalid: 379 ms) (Figure 8b), F(1,13) = 47.91,

p < .01. But the duration of the cue-target interval had no effect on reaction time (0 ms: 338 ms,

50 ms: 337 ms, 100 ms: 339 ms) (Figure 8b, left panel), F(2,26) = 1.11, p = .35. Overall, higher

accuracy was associated with shorter RTs (Figure 8a & 8b), suggesting that there was no speed-

accuracy trade-off.

Discussion.

In this experiment, the participant’s expectation of the target direction was induced by

presenting a direction cue before the stimulus appeared. The direction cue significantly affected

task performance, as both the accuracies and RTs in the valid and invalid conditions were

different (Figure 8). The difference between the valid and invalid conditions became greater

with increasing cue-target interval. Interestingly, with a longer interval, a valid cue did not

yield better performance; rather, an invalid cue further impaired performance. This may imply

that inhibition in unwanted directions-but not enhancement in the wanted direction-may have

played a major role during attentional processing, between 80 ms to 180 ms following the onset

of the cue.

Although the differences in performance between the valid and invalid conditions were

large, at least part of these differences was likely due to a guessing bias. As some participants

pointed out, when they were not sure about the direction of the target, they were more likely to

29

respond with the cued direction. They knew that choosing the cued direction had a greater

chance to be correct because the cue was highly valid. However, the advantage in the valid trials

over the invalid cannot completely be attributed to the guessing bias. The interaction between

cue validity and cue-target interval demonstrated that the performance difference between the

valid and invalid conditions was not merely due to the guessing bias. Otherwise, the interaction

would not have been significant, assuming that the benefit from biased guessing was the same

across different cue-target intervals.

The reaction times also provided an indication of guessing behaviour, particularly at

large eccentricities when the cue was invalid. An inverted V shape pattern was observed in the

RT means with increasing eccentricity in the invalid cue condition. The fastest response and

highest accuracy at 10o reflects the lowest difficulty level. However, at an eccentricity of 30

o,

the reaction times decreased compared to an eccentricity of 20o. Participants probably found

determining the direction of the target very difficult and started to rely on guessing to complete

the task.

Experiment 2B.

The AVF task was changed from reporting the direction of the target to reporting its

identity. There were two visually different targets and participants reported which target had

been presented. Because no relationship existed between the identity of the target and the cued

direction, participants were not able to improve performance by guessing. This methodology

can examine the pure effect of covert orienting of attention on the distribution of attention.

Method.

Participants. 20 undergraduates at the University of Toronto (eight males, 12 females;

age range: 17-23 years) participated for course credit.

30

Task. The paradigm was very similar to Experiment 2A except participants reported the

identity of the target (the target was now bisected by either a horizontal or a vertical line) rather

than the direction of the target (Figure 9). Given that the effect of cue-target interval was

demonstrated in Experiment 2A, only two cue-target intervals (0/80 ms) were used to limit the

number of trials.

Stimuli. All task settings were the same as in Experiment 2A, except the target was a

filled dark-gray circle (2.2o×2.2

o) with a dark-gray line (0.8

o×3.6

o) bisecting the circle either

horizontally or vertically. After a blank interval of 0 or 80 ms, the stimulus display appeared for

40 ms, followed by a mask of 200 ms duration. Participants responded after the mask

disappeared.

Procedure. This remained the same as in Experiment 2A except that participants

reported the identity of the target, pressing ‘Z’ for targets with the horizontal line and ‘/’ for

targets with the vertical line. Participants pressed ‘Z’ with the left hand, and ‘/’ with the right

hand. Participants were instructed to respond both accurately and quickly. The choices of target

identity and the reaction times were recorded. In the experimental session, 720 trials were

presented in a random sequence with the target appearing 30 times in each of the 12 locations,

for each of the two cue-target interval conditions. 80% of the trials were valid.

Results.

Accuracy. Participants were less accurate with increasing eccentricity (10o: 87%, 20

o:

71%, 30o: 69%) (Figure 10a), F(2,38) = 68.53, p < .01. Performance was better on valid trials

than invalid trials (valid: 77%, invalid: 74%), F(1,19) = 10.20, p < .01. This advantage on valid

trials was greater at the longer cue-target interval (0 ms: 1%, 80 ms: 6%) (Figure 10a), F(1,19) =

10.06, p < .01. When the cue-target interval was zero, there was little difference in accuracy

between valid and invalid conditions (valid: 77%, invalid: 76%) (Figure 10a, left panel), F(1,19)

31

= 1.40, p = .25. However, with an interval of 80 ms, the difference was large (valid: 78%,

invalid: 72%) (Figure 10a, right panel), F(1,19) = 18.88, p < .01. Although target detection in

valid trials was not enhanced at longer intervals (0 ms: 77%, 80 ms: 78%) (Figure 10a, left

panel), F(1,19) = 1.46, p = .24, accuracy was impaired in invalid trials at longer intervals (0 ms:

76%, 80 ms: 72%) (Figure 10a, left panel), F(1,19) = 6.67, p < .05.

RT . Participants took longer to respond with increasing eccentricity (10o: 449 ms, 20

o:

494 ms, 30o: 545 ms) (Figure 10b), F(2,38) = 16.62, p < .01. Participants also responded faster

with valid cues than with invalid cues (valid: 463 ms, invalid: 529 ms) (Figure 10b), F(1,19) =

44.91, p < .01. When the cue-target interval was zero, the mean reaction time was shorter with

valid cues than with invalid cues (valid: 476 ms, invalid: 534 ms) (Figure 10b, left panel),

F(1,19) = 20.41, p < .01. With an interval of 80 ms, participants also responded faster with valid

cues than with invalid cues (valid: 450 ms, invalid: 525 ms) (Figure 10b, right panel), F(1,19) =

34.39, p < .01. Higher accuracies were generally accompanied by faster RTs (Figure 10a & 10b),

suggesting that there was no speed-accuracy trade-off.

Discussion.

Expectation of the direction of the target significantly changed the distribution of

attention. The data suggest that more attentional resource was deployed in the expected

direction. As a result, when the cued direction was correct, target identification was more

accurate than when the cued direction was incorrect. With a cue-target interval of 0 ms, the

mean difference in accuracy was not significant, although it was in the right direction. However,

with a cue-target interval of 80 ms, the effect was clear. The mean difference in accuracy was

small (yet still significant) at an eccentricity of 10o, probably because of the low difficulty of the

task at this eccentricity. The mean of this difference became larger (a trend) at 20o and 30

o

(Figure 10a, right panel).

32

General discussion.

Experiments 2A and 2B demonstrated that expectation of the direction of the target is an

effective form of top-down influence on selective attention across the visual field. In both

experiments, the expectation was induced by presenting a cue indicating the likely direction of

the target before the stimulus. When the target appeared in the expected direction, it was easier

to identify the target among distractors, indicating that participants covertly oriented attention

toward the expected direction. In addition, the expectation of the direction of the target was

more effective at larger eccentricities. This increasing effectiveness of endogenous orienting of

attention at higher eccentricities implies that the spatial attributes of stimuli become more

important for attentional processing in the periphery.

An increase of the cue-target interval produced a more pronounced effect with

endogenous covert orienting of attention. This is evident in both Experiments 2A and 2B.

Particularly, in Experiment 2B, with a cue-target interval of 0 ms, there was a trend (but not

significant) of higher accuracy when the target appeared in the expected direction (valid

conditions) compared to when the target appeared in unexpected directions (invalid conditions).

As the cue-target interval was increased to 80 ms, the difference in accuracy became significant.

Presumably, with a longer cue-target interval, participants had more time to form their

expectation of the likely direction (Posner, 1980; Shepherd & Müller, 1989). Thus the

difference in accuracy between the valid and invalid conditions became greater. In both

Experiments 2A and 2B, this greater difference with a prolonged interval was caused by further

impairment in accuracy by an invalid cue, rather than by an enhancement in accuracy by a valid

cue. This may imply that, when discrimination and identification of a target among distractors

are necessary, expectation of the direction of the target improves performance mostly by

33

inhibiting the unexpected directions between 80 ms to 180 ms (160 ms in Experiment 2B)

following the onset of the cue.

Experiment 3

It is known that the attentional intensity decreases with increasing eccentricity from

fixation (Ball et al., 1988; Feng et al., 2007). What still remains unknown is whether the amount

of attentional resource deployed to locations is the same in the left and the right visual fields, or

in the upper and lower visual fields. In other words, whether the distribution of attention is

spatially symmetrical. Spatial bias has frequently been reported in perceptual tasks (e.g. Jewell

& McCourt, 2000) and a memory task (e.g. Dickinson & Intraub, 2009). Since attention is

involved in many perceptual and memory tasks (Intraub, Daniels, Horowitz & Wolfe, 2008;

Johnson & Proctor, 2004), it would not be surprising to find evidence of asymmetry in attention.

Experiment 3 was designed to investigate whether biases exist in the distribution of attention in

the horizontal (left vs. right) and vertical (upper vs. lower) directions.

The magnitude and direction of the spatial bias in some perceptual tasks, such as line

bisection, are modulated by the characteristics of the stimulus (e.g. stimulus saliency and spatial

location) and the participant (e.g. gender, age) (as reviewed by Jewell & McCourt, 2000). In

addition to the usual factors that are varied in the AVF task (target eccentricity and target

direction), three other factors were varied in this experiment: gender, eye dominance and

stimulus exposure. A gender difference has been demonstrated in AVF task performance (Feng

et al., 2007) and in spatial bias in some perceptual tasks (Brodie, 2010; McCourt & Olafson,

1997; Varnava & Halligan, 2006). Therefore, it is possible that a gender difference may be

present in the distribution of attention. Additionally, since eye dominance is an indication of

lateralization in the visual system (Crovitz & Zener, 1962), and attentional bias may be

associated with visual lateralization (Roth, Lora & Heilman, 2002), the effect of eye dominance

34

was examined. Moreover, because stimulus exposure affects performance on the AVF task (Ball

et al., 1988), this factor was also varied.

Method.

Participants. 52 undergraduates at the University of Toronto (26 males: 7 left eye-

dominant, 19 right eye-dominant; 26 females: 11 left eye-dominant, 15 right eye-dominant; age

range: 17-30 years) participated for course credit.

Task. Participants completed the AVF task without cues at several stimulus exposures

(10/20/30/40/50/80 ms).

Design. The experiment was a between-within design, with three within-subject factors:

target eccentricity (10o/20

o/30

o), target direction (directions 1, 2, 3, 4, 6, 7, 8, 9, as illustrated in

Figure 11), and stimulus exposure (10/20/30/40/50/80 ms). There were also two between-

participant factors: gender (male/female) and eye dominance (left-dominant/right-dominant).

When comparing performance on the left and right areas of the visual field, left included

directions 1, 4, and 7; and right included directions 3, 6, and 9 (Figure 11; the same arrangement

as on the number key pad of a computer keyboard). Similarly, in the comparison of the upper

and lower halves of the visual field, upper included directions 7, 8, and 9; and lower included

directions 1, 2, and 3 (Figure 11).

Stimuli. In the AVF task, the stimuli were presented in a circular area (63.1o in diameter)

centered on a uniform light-gray screen. Each trial began with a centered, unfilled fixation

square with a dark-gray border (3o×3

o). The fixation square was presented for 800 ms and then

augmented by the stimulus display, which consisted of 23 identical distractors and one target,

each uniquely localized at an eccentricity of 10o, 20

o, or 30

o in one of eight equally spaced

directions. The location of the target was randomly selected on each trial, subject to the

restriction that the target appeared an equal number of times in each possible location. The

35

target was a dark-gray filled square (1.5o×1.5

o) surrounded by an unfilled circle with a dark-gray

circumference (3o×3

o). The distractors were unfilled squares with dark-gray borders (3

o×3

o),

identical to the fixation square. The stimulus display was presented for an exposure that varied

among 10/20/30/40/50/80 ms at random, with each exposure being selected with equal

frequency. Participants indicated direction of the target after the presentation of a mask for 500

ms. The next trial started 800 ms after a response was made.

Procedure. Participants were first asked to report their gender and age. Then they

completed the Miles Eye Dominance Test (Miles, 1930) under the supervision of the

experimenter. During the eye dominance test, participants were instructed to look through an

opening formed by both hands with arms extended. They first looked with both eyes and then

with each eye in turn. If an object that could be seen through the opening with both eyes was

also seen with one eye (but not the other), that eye was judged to be the dominant eye.

Participants then completed the AVF task. Participants were required to position their head on a

chin rest at a distance of 35cm from the centre of the screen. A practice session, consisting of 72

trials, was mandatory before the experimental session in order to ensure that participants

understood the task. In the experimental session, 720 trials were divided into 5 blocks of 144

trials. Trials in each block were presented in random sequence with the target appearing six

times in each of the 24 locations. The exposure condition varied constantly within each block

with the constraint that each exposure was equally represented over each block of 144 trials.

Participants were allowed a two-minute rest after each block. Responses indicating the direction

of the target and the reaction times were recorded.

36

Results.

Horizontal bias: Left vs. Right

Accuracy. Mean correct detection of the direction of the target decreased with an

increase in eccentricity (10o: 77%, 20

o: 76%, 30

o: 66%) (Figure 12a, upper panels), F(2,96) =

30.27, p < .01, and shorter exposures (10 ms: 51%, 20 ms: 65%, 30 ms: 73%, 40 ms: 79%, 50

ms: 82%, 80 ms: 88%), F(5,240) = 199.95, p < .01; there was a linear trend, F(1,48) = 329.64, p

< .01. Decreasing the exposure increased task difficulty particularly at large eccentricities

(Figure 13, left panel), F(10,480) = 11.92, p < .01. These effects of eccentricity and exposure

were similar in both the left and right attentional visual fields. This is supported by lack of

interaction between eccentricity and horizontal direction (left/right), F(2,96) = 1.38, p = .26;

between exposure and horizontal direction (left/right), F(5,240) = .977, p = .43; and among

eccentricity, exposure and horizontal direction (left/right), F(10,480) = .62, p = .80. However,

there was a bias toward either the left or the right visual field and men and women differed in

the direction of bias, F(1,48) = 14.27, p < .01. Men were more accurate in the right visual field

compared to the left (left: 78%, right: 81%) (Figure 12a, upper left panel), F(1,24) = 4.83, p <

.05; while women were more accurate in the left visual field compared to the right (left: 72%,

right: 65%) (Figure 12a, upper right panel), F(1,24) = 10.36, p < .01. Such gender difference

was evident when comparing each pair of corresponding directions, including the upper

diagonals (direction 7 vs. direction 9) (men: left 76%, right 80%; women: left 66%, right 70%),

F(1,48) = 7.74, p < .01, the horizontal directions (direction 4 vs. direction 6) (men: left 81%,

right 85%; women: left 74%, right 76%), F(1,48) = 5.10, p < .05, and the lower diagonals

(direction 1 vs. direction 3) (men: left 76%, right 79%; women: left 66%, right 62%), F(1,48) =

5.45, p < .05.

37

In general, men were more accurate on the AVF task than women (men: 79%, women:

68%), F(1,48) = 9.32, p < .01. Eye dominance was not associated with either overall accuracy

(left dominant: 73%, right dominant: 73%), F(1,48) = .03, p = .87, or the direction of bias (left

dominant: left 73%, right 72%; right dominant: left 74%, right 73%), F(1,48) = .07, p = .80. The

finding that eye dominance did not affect the direction of bias was valid for both men and

women, F(1,48) = .08, p = .78.

RT. Speeds of responses differed among eccentricity (10o: 360 ms, 20

o: 345 ms, 30

o: 377

ms) (Figure 12b, upper panels), F(2,96) = 8.48, p < .01. A contrast revealed significantly slower

responses at an eccentricities of 30o than at an eccentricity of 20

o, F(1,48) = 31.24, p <

.01.Particpants also became slower at shorter exposures (10 ms: 410 ms, 20 ms: 370 ms, 30 ms:

367 ms, 40 ms: 345 ms, 50 ms: 334 ms, 80 ms: 337 ms), F(5,240) = 12.90, p < .01; there was a

linear trend, F(1,48) = 18.75, p < .01. Decreasing the exposure increased reaction time,

particularly at large eccentricities (Figure 13, right panel), F(10,480) = 1.99, p < .05. Men and

women did not differ significantly in speed (men: 343 ms, women: 378 ms), F(1,48) = .36, p =

.55. But men were faster on the right side (left: 356 ms, right: 330 ms) (Figure 12b, upper left

panel), F(1,24) = 12.49, p < .01; while women’s speed on the two sides did not differ (left: 379

ms, right: 378 ms) (Figure 12b, upper right panel), F(1,24) = .02, p = .89. Eye dominance was

not associated with speed of responding F(1,48) = 1.39, p = .24. Quicker responses were

generally associated with higher accuracies (Figure 12a & 12b, upper panels), suggesting that

there was no speed-accuracy trade off.

Vertical bias: Upper vs. Lower

Accuracy. Accuracy decreased with an increase in eccentricity (10o: 76%, 20

o: 73%, 30

o:

55%) (Figure 12a, lower panels), F(2,96) = 82.90, p < .01, and shorter exposures (10 ms: 46%,

20 ms: 60%, 30 ms: 68%, 40 ms: 73%, 50 ms: 77%, 80 ms: 84%), F(5,240) = 232.32, p < .01;

38

there was a linear trend, F(1,48) = 405.41, p < .01. Decreasing the exposure resulted in

decreased accuracy at large eccentricities (Figure 14, left panel), F(10,480) = 9.12, p < .01.

These effects of eccentricity and exposure were similar in both the upper and lower visual

fields. This is supported by the lack of interaction between eccentricity and vertical direction

(upper/lower), F(2,96) = .90, p = .41; between exposure and vertical direction (upper/lower),

F(5,240) = 1.88, p = .10; and among eccentricity, exposure and vertical direction (upper/lower),

F(10,480) = 1.24, p = .27. Accuracy was higher in the top half of the visual field (upper: 70%,

lower: 66%), F(1,48) = 7.25, p < .05; and there was no gender difference in this bias (Figure

12a, lower panels), F(1,48) = 1.05, p = .31. This lack of difference among men and women was

also evident when comparing each pair of corresponding directions, including the left diagonals

(direction 1 vs. direction 7) (men: upper 76%, lower 76%; women: upper 66%, lower 70%),

F(1,48) = 2.15, p = .15, the vertical directions (direction 2 vs. direction 8) (men: upper 72%,

lower 65%; women: upper 65%, lower 54%), F(1,48) = .33, p = .57, and the right diagonals

(direction 3 vs. direction 9) (men: upper 80%, lower 79%; women: upper 66%, lower 62%),

F(1,48) = 1.01, p = .32.

In general, men were more accurate on the AVF task than women (men: 74%, women:

61%), F(1,48) = 9.96, p < .01. The magnitude of the bias differed among eccentricities (10o: 6%,

20o: 8%, 30

o: 3%), F(2,96) = 3.62, p < .05. In particular, the magnitude of bias was much greater

at an eccentricity of 20o than at 30

o, F(1,48) = 9.16, p < .01. The magnitude of bias was also

smaller with increase of exposure (10 ms: 7%, 20 ms: 8%, 30 ms: 6%, 40 ms: 5%, 50 ms: 4%,

80 ms: 3%), F(5,240) = 2.27, p < .05; there was a linear trend, F(1,48) = 4.66, P < .05. Eye

dominance was not associated with accuracy (left dominant: 68%, right dominant: 68%),

F(1,48) = .01, p = .93.

39

RT. Speeds of responses differed among eccentricities (10o: 359 ms, 20

o: 349 ms, 30

o:

407 ms) (Figure 12b, lower panels), F(2,96) = 20.86, p < .01. Contrasts among the eccentricities

revealed that response was slower at an eccentricity of 30o compared to that of 10

o, F(1,48) =

16.13, p < .01, and 20o, F(1,48) = 41.94, p < .01. Responding was also slower with a decrease

in exposure (10 ms: 420 ms, 20 ms: 384 ms, 30 ms: 380 ms, 40 ms: 356 ms, 50 ms: 348 ms, 80

ms: 342 ms) (Figure 12b, lower panels), F(5,240) = 17.30, p < .01; there was a linear trend:

F(1,48) = 31.04, p < .01. Speed did not differ between the upper and lower halves of the visual

field (upper: 377 ms, lower: 366 ms), F(1,48) = 1.37, p = .25, nor between men and women

(men: 353 ms, women: 391 ms) (Figure 12b, lower panels), F(1,48) = .39, p = .54. Faster

responses were, in general, associated with higher accuracies (Figure 12a & 12b, lower panels),

suggesting that there were no speed-accuracy trade-off effects.

Discussion.

The distribution of attention was not symmetrical across the visual field. Spatial biases

were evident when comparing the left half of the visual field to the right half, as well as the

upper half to the lower half.

In the left vs. right comparison, the distribution of attention was in general biased to the

left in women and to the right in men. This was true when comparing each pair of corresponding

directions (the horizontal pair and the upper and lower diagonal pairs). Similar to the large

individual variability in the direction of bias found in a line bisection task (Manning, Halligan &

Marshall, 1990), individual difference also exists in the direction of the horizontal bias in the

spatial distribution of attention.

In the upper vs. lower comparison, an upward bias in the spatial distribution of attention

was found in both men and women. This was true when comparing each pair of corresponding

directions (the vertical pair and the left and right diagonal pairs). This supports the notion that

40

an upward bias in attention exists, and this bias may have played a role in the upward bias

observed in line bisection tasks (e.g. Bradshaw et al., 1985; Van Vugt et al., 2000). No gender

difference was found in the direction of this bias. This implies that the underlying mechanism

for the vertical bias in the spatial distribution of attention may be different from the mechanism

for the horizontal bias. Further discussion of possible mechanisms that may be responsible for

the spatial attentional biases observed in this experiment is included in Chapter 5.

41

Chapter 5: Discussion

Top-down Influence in the Attentional Visual Field

Three forms of top-down influence on selective attention across the visual field were examined.

These influences included 1) the expectation of the size of the area in which the target was

likely to occur; 2) the expectation of the likely direction of the target; and 3) existing biases in

the spatial distribution of attention. Results from the experiments suggested that a participant

can consciously modify the distribution of attention according to the expectation (e.g. varying

the size of the attended area, emphasizing particular directions or locations). These conscious

influences vary as a function of the time available for preparation (cue-target interval). In

addition to conscious biases, a participant may also give preference to certain spatial areas

(unconscious bias). Unconscious bias affects the distribution of attention across the visual field.

Expectation of the size of the area.

The attentional visual field has been defined as the area “in which useful information can

be acquired without eye or head movement” (Ball et al., 1988, p.2210). Experiment 1

demonstrated that the expectation of the size of the area in which the target is likely to occur

influences the size of the attentional visual field and the distribution of attention over this field.

Experiment 1B found that when the size of the cued area was increased, average accuracy

declined, implying that the average attentional intensity within the attended area had decreased.

This inverse relationship between the size of the attended area and the attentional intensity was

first documented by Wolff (1738, 1740, translated and interpreted in Hatfield, 1998) in his

Psychologia Empirica (1738) and Psychologia Rationalis (1740). Later, similar ideas were

implied in Titchener’s Law of Two Levels (1908) and also the zoom-lens model of selective

attention (Eriksen & Yeh, 1985; Eriksen & St. James, 1986). In the zoom-lens model, it is

assumed that attention is distributed evenly across the selected area except there is a gradual

42

rather than rapid decrease of the attentional intensity near the boundary (Eriksen & St. James,

1986); whereas performance on the AVF task suggests that distribution of attention is more like

a unimodal probability distribution (such as a bivariate Gaussian), with lower attentional

intensity at locations further from fixation. Also, when highly focused on a primary central task,

participants were less capable of noticing stimuli presented outside the area of the primary task

(Ikeda & Takeuchi, 1975; Williams, 1995). Such experiments suggest that the size of the

attended area can be modified according to the demands of the task.

The ability to modify the distribution of attention should help ensure optimal

performance, since the attentional resource is limited (Cavanagh & Alvarez, 2005; Pylyshyn &

Storm, 1988; Sears & Pylyshyn, 2000; Trick & Pylyshyn, 1993, 1994). This ability to

consciously modify the size of the attended area attests to the great flexibility with which

attention may be deployed. This flexibility allows us to optimize the distribution of attention

according to the predicted importance of events that are likely to occur. When we make a highly

confident prediction about the location of a target, attention is preferentially deployed to that

location to maximize the chance of quickly and accurately identifying the target. In contrast,

when we are not certain about the specific location, we tend to distribute attention over a much

larger area, to ensure that a target is less likely to be missed no matter where it occurs.

Modifying the size of the attended area takes time to complete. Modification of the

distribution of attention when the cue was given before each trial was not as effective as when

the cue was given at the beginning of each block. When the area size cue was given at the

beginning of each block, the participant had plenty of time to modify the expectation of the area

in which the target was likely to appear. In contrast, when the cue varied from trial to trial, the

participant had only a short time to change the expectation. The data from Experiment 1 are

consistent with Eriksen and St. James (1986), who found that the performance benefit from

43

attention becoming more concentrated on the location of a target increased with a longer cue-

target onset interval; presumably, the longer interval allows the modification of the distribution

of attention to be fully completed. The difference in the results of Experiments 1A and 1B also

implies that a clear sense of the boundaries of the to-be-attended area plays an important role.

When the size of the expected area did not have to change over a block of trials (72 or 80 trials

per block), it was easier to form an unambiguous sense of boundaries, as reported by some

participants and implied by the results.

Results from Experiment 1 were consistent with a unimodal bivariate probability model

by demonstrating that the spatial extent and intensity of the distribution of attention is modified

according to expectation. As illustrated in Figure 2a, when participants expected a small area for

the target, the spread of the distribution of attention was narrowed. The attentional intensity

increased in the central area and decreases at locations distant from fixation. However, as in

Figure 2b, when participants expected a large area for the target, the spread of the distribution

was much wider and the attentional intensity dropped around the center and increased in the

periphery. These modifications of the distribution of attention were more obvious with a longer

time for preparation of the expectation of the size of the to-be attended area, and when a clearer

sense of the boundary of the to-be-attended area could be established.

Expectation of the direction of the target.

Expectation of the direction of the target induces endogenous covert orienting of

attention (Wright & Ward, 2008). This type of attentional orienting refers to orienting of

attention without eye or head movement (Posner, 1978). Attention is directed to a location other

than the fixation without changing fixation. When an endogenous cue is used to induce orienting,

the process involves top-down modulation of attention, since the spatial information contained

in the endogenous cue must be interpreted to guide the orienting of attention (Wright & Ward,

44

2008). Experiment 2 showed that expectation of the direction of the target is an effective form

of top-down influence.

The influence of the expectation of the likely direction of the target on attention

increases with time. This is evident in both Experiment 2A and 2B. In Experiment 2A, the

advantage on valid trials, compared to invalid trials, was greater with a longer cue-target

interval (comparing among intervals of 0 ms, 50 ms and 100 ms). In Experiment 2B, the

advantage on valid trials increased considerably when the interval was increased from 0 ms to

80 ms (from a non-significant difference in accuracy and RT, to a much higher accuracy and

faster RT on valid trials). Improved performance with a valid endogenous cue and a longer cue-

target interval has been demonstrated at eccentricities of 10o and 20

o (Shepherd & Müller, 1989)

but in their study, the target was presented without distractors, thus only target detection (no

discrimination or identification) was necessary. Experiments 2A and 2B suggested that this

effect holds when discrimination, localization and identification (finding the target in the

presence of distractors) were also involved. And the effect holds not only at eccentricities of 10o

and 20o, but also at an even more extreme eccentricity of 30

o. Moreover, the benefit from a valid

cue is progressively greater at locations further from fixation.

In both Experiments 2A and 2B, with a longer cue-target interval, a valid cue did not

further facilitate identification of the target, but an invalid cue further impaired the identification.

This differs from the findings in Shepherd & Müller (1989). In Shepherd & Müller (1989), with

a longer time following the onset of the cue (increased from 50 ms to 150 ms), the accuracy on

target detection was further enhanced with a valid cue, and further impaired with an invalid cue.

Notably, in Shepherd & Müller (1989), there was no distractor presented together with the target;

in contrast, in Experiment 2A and 2B, the stimuli consisted of a target and eleven distractors.

This may imply that, when only detection is involved, both enhancement in the expected

45

direction and inhibition in the unexpected directions occur (between 50 ms to 150 ms following

the onset of the cue). However, if discrimination and identification are also necessary, inhibition

in the unexpected directions may have played a major role (between 80 ms to 180 ms following

the onset of the cue).

Endogenous covert orienting of attention induced by an expectation of the target is an

efficient mechanism for enhancing attentional performance when the prediction is highly

accurate. A significant advantage of endogenous covert orienting of attention across the

extended attentional visual field would be faster detection of a target that appears some distance

from fixation. Endogenously orienting attention to a new location can be faster (≈ 120 ms,

Johnson & Proctor, 2004) than making an eye movement (≈ 200 ms, Johnson & Proctor, 2004).

Increased attentional intensity at a location is more quickly achieved by covert orienting of

attention than by making an eye movement to that location.

Results from Experiment 2 provided further support for the simple statistical model of

the distribution of attention by demonstrating that the distribution of attention is modified by the

expectation of the likely direction of the target. Similar to the illustration in Figure 2c, when

participants expected the target to appear in a particular direction, the attentional intensity

increased in that direction, and decreased in other directions. This modification of the

distribution of attention was further enhanced with a longer time for preparation (cue-target

interval).

Spatial bias in the distribution of attention.

The distribution of attention across the visual field is asymmetrical, even when a

participant focuses on the central fixation point without any expectation of the likely location of

the target (Chan & So, 2007). Experiment 3 demonstrated that the distribution of attention is

biased toward certain areas in the visual field. Individual performance on the AVF task was

46

superior in either the left or right hemifield; and, in general, performance in the upper half of the

visual field was better than that in the lower half.

Although not consciously initiated by the observer, this bias in the distribution of

attention is a significant source of top-down influence. Because attention facilitates encoding of

information into memory (Johnson & Proctor, 2004), a stimulus that receives more attention

may be better memorized. As a result, the bias in the spatial distribution of attention may

contribute to the bias observed in some cognitive tasks like visual memory tasks and visual

search tasks. In one example, Dickinson & Intraub (2009) found that participants’ first eye

movement was more often to the left rather than to the right when viewing a picture.

Participants also remembered scene objects better in the left half of the scene compared to the

right half.

Bias in spatial attention has also been associated with bias in other perceptual tasks

(Heilman & Van Den Abell, 1980; Jewell & McCourt, 2000; Niemeier, Stojanoski & Greco,

2007; Niemeier, Singh, Keough & Akbar, 2008a; Niemeier, Stojanoski, Singh & Chu, 2008b).

In line bisection, for example, a participant’s response was often to the left of the veridical

center of a horizontal line (e.g. Bowers & Heilman, 1980; Bradshaw et al., 1983), and above the

center of a vertical line (e.g. Bradshaw et al., 1985; Mefferd, Wieland & Difilho, 1969). Similar

perceptual bias was also found when participants judged which of two grating scales had more

thinner or thicker stripes (Niemeier et al., 2007; Niemeier et al., 2008a, 2008b). This was

attributed to a leftward attentional bias resulting from a bias in inhibition of distractors during

attentional processing toward the left (Niemeier et al., 2008a).

Horizontal bias (left or right) in the spatial distribution of attention may be a function of

lateralization in the brain. The dominant role of the right hemisphere has been observed in many

spatial tasks (e.g. Kimura, 2000; Hillis et al., 2005). For example, healthy adults relied more on

47

the right hemisphere when performing a spatial cueing attentional task (Corbetta, Kincade,

Ollinger, McAvoy & Shulman, 2000), or a line-bisection task (Foxe, McCourt, & Javitt, 2003).

Similarly, patients with a right hemispheric brain lesion demonstrated severe impairment when

performing tests of hemispatial neglect, including copying a scene, line bisection, line

cancellation, sentence reading (Hillis et al., 2005), and sentence letter cancellation (Karnath,

Ferber & Himmelbach, 2001). Since visual information is processed primarily in the

contralateral hemisphere (Heliman, Waston, & Valenstein, 1985; Robertson & Marshall, 1993),

the dominant role of the right hemisphere in spatial processing may contribute to a general

leftward bias in the distribution of attention.

However, in the left vs. right comparison, why did the direction of the bias differ in men

and women in Experiment 3? Some clues may be found in other experiments that have observed

reversed perceptual asymmetry. For example, large individual variation was found in both the

magnitude and direction of bias in line bisection tasks: Manning, Halligan, & Marshall (1990)

found that some participants consistently biased their responses toward the left, while others

were consistently biased toward the right. Gender differences in the magnitude of bias and

variation in the direction of bias have also been reported in several studies (Brodie, 2010; Jewell

& McCourt, 2010; McCourt & Olafson, 1997; Varnava & Halligan, 2006). Men were found to

display either greater bias (Brodie, 2010; Jewell & McCourt, 2010; McCourt & Olafson, 1997)

or greater variation in the direction of bias when bisecting lines (Varnava & Halligan, 2006).

It has been suggested that the direction of bias in a line bisection task might depend on

how we monitor the movement of an arm (Daini, Arduino, Menza, Vallar & Silveri, 2008).

According to Daini et al. (2008), a bias to the left may be related to visual monitoring of the

movement of an arm during a line bisection task, while a rightward bias is likely associated with

kinaesthetic monitoring of the movement. This implies that individual differences in the

48

direction of bias may be related to how an individual performs the task. It is not clear that if men

and women in Experiment 3 might have performed the task differently as noted in Daini et al.

(2008), although the influence of performing the task differently, even it existed, was likely to

be minimal. In an AVF task, participants responded by pressing a certain key on the keyboard.

In the trials with correct responses (excluding the trials during which participants guessed the

answer), the response is known by the participant before key pressing. Therefore, the accuracy

of a response is not linked to visual or kinaesthetic monitoring of the movement of an arm.

In contrast, it is more likely that the gender difference in the direction of bias in

Experiment 3 is a function of a gender difference in hemispheric asymmetry (Hellige, 2001) and

a gender difference in the brain structures involved in spatial processing (Frederikse et al., 1999;

Koscik et al., 2009). The volume of left inferior parietal lobule (IPL) is greater in men

(Frederikse et al., 1999) and men also often display a leftward lateralization in the structure of

the IPL (Frederikse et al., 1999), while women on average have no lateralization or slightly

lateralized toward the right (Marsh & Casper, 1998). The inferior parietal lobule (IPL), often

also referred to as the posterior parietal cortex (PPC, Mesulam, 1998) (IPL is part of PPC), is

heavily involved in visuospatial processing (Keating & Gooley, 1988; Peterson, Robinson &

Currie, 1989) and selective attention (Peterson et al., 1989; Heilman, Watson, Valenstein &

Damasio, 1993). Because information from one side of the visual field is primarily processed in

the contralateral hemisphere (Heliman et al., 1985; Robertson & Marshall, 1993), a

lateralization in the structure of the IPL may be associated with superior attentional performance

in the contralateral side of the visual field. As a result, a leftward lateralization in the structure

of the IPL in men may be related to a rightward bias in the spatial distribution of attention;

while a slightly rightward bias in the structure of the IPL in women may be related to a leftward

bias in the spatial distribution of attention.

49

In addition to bias in horizontal line bisection, bias favouring the upper half of a vertical

line has also been observed in vertical line bisection (Bradshaw et al., 1985; Drain & Reuter-

Lorenz, 1996; McCourt & Olafson, 1997; Van Vugt, Fransen, Creten & Paquier, 2000), and on

a task in which participants compared the upper and lower halves of a pair of greyscales

(presented vertically side by side) and judged which half was darker (Herber, Siebertz, Wolter,

Kuhlen & Fimm, 2010). In addition, superior performance has been shown when a target was

presented in the upper visual field (compared to in the lower visual field) when participants

searched for a target among distractors (Previc & Blume, 1993) and when participants had to

recognize an object (Chambers, McBeath, Schiano & Metz, 1999). These biases in the vertical

direction may be due to the upward bias in the distribution of attention. Arguing from an

evolutionary perspective, Jeerakathil and Kirk (1994) noted that there is a tendency to deploy

more attention to the upper part of an object and conjectured that this is because the upper part

of an object generally contains more important information (e.g. face in relation to the rest of

the body). According to Previc (1990, 1998), the upper visual field is mainly involved in visual

search and recognition, while the lower visual field is primarily engaged in visuomotor

coordination, and these specializations of the upper and lower visual fields may have occurred

during primate evolution. Given the abilities (e.g. discrimination, localization and identification)

required in the AVF task are also critical for visual search (Chan & So, 2007; Chan & Tang,

2007), it would not be surprising to find an upward bias in performance on the AVF task, if the

upper visual field is evolved to specialize in visual search.

The distribution of attention across the visual field is not symmetrical. Even when

attention is not consciously oriented away from the fixation, eccentricity is not the only

parameter associated with the attentional intensity at a particular location. Experiment 3

identified significant biases in the spatial distribution of attention and individual differences in

50

the biases. Women generally displayed a bias toward the left while men showed a bias toward

the opposite side. Both men and women demonstrated a bias toward the upper half of the visual

field. These individual differences in spatial asymmetries in the distribution of attention are

important when considering the performance of an individual or a particular population (e.g.

older adults, men and women) on perceptual, attentional and even memorial tasks.

New Findings from Investigating Attention across the Extended Visual Field

Three forms of top-down influence on selective attention across the visual field were

discussed in the above sections. The significant effects of the top-down influences demonstrated

in Experiments 1, 2 and 3 implied that selecting spatial locations is a very important mechanism

for attentional selection across the visual field. This is likely associated with declining visual

acuity and color vision at larger eccentricities (Anstis, 1998; Hansen et al., 2009). Therefore, a

model of attentional selection across the visual field should emphasize selection mechanisms

based on spatial location.

Most previous research on attention has been confined to a relatively small area around

the fixation. Normally, the visual angle of stimulus was no greater than 15o in these studies (see

para. 1, p.2, Chapter 1). Because of this restriction, investigation of the effect of eccentricity has

been very limited in previous studies. The present experiments reveal how eccentricity affects

the distribution of attention. First, performance on the AVF tasks in all three experiments

demonstrate that attention is more intensely distributed around the fixation and that the

attentional intensity decreases gradually with an increase of eccentricity from the fixation. Thus,

the distribution of attention across the visual field may be described by a unimodal probability

distribution (e.g. a bivariate Gaussian). Second, at larger eccentricities, the effect of top-down

influence on selective attention may be strengthened. Results from Experiment 2A showed that

the performance advantage in valid conditions over the invalid conditions became greater with

51

the increase in eccentricity. A similar trend was also found in Experiment 2B. This suggests that,

as the perceptual quality of features like color and shape degrades in the periphery, consciously

concentrating attention at expected target locations becomes increasingly important. Third, an

increase in task difficulty when the stimulus exposure is reduced is greater at larger

eccentricities. This was evident in Experiment 3. This implies that at a location with lower

attentional intensity (e.g. at a greater eccentricity), a sufficiently long exposure becomes more

critical to facilitate the discrimination of target among distractors.

Previously, unconscious bias in the spatial distribution of attention has not been directly

observed, although studies on perceptual and memorial asymmetries have suggested such bias in

the distribution of attention (e.g. Dickinson & Intraub, 2009; Jewell & McCourt, 2000). Results

from Experiment 3 demonstrated unconscious bias in the horizontal and vertical directions. In

addition, a significant gender difference was found in the direction of horizontal bias. These

findings provided evidence supporting the notion that bias in the distribution of attention in the

visual field exists and such bias might have caused the perceptual and memorial asymmetries

observed by others.

Neuropsychological Basis of Top-down Influence

Conscious top-down influence on selective attention across the visual field modifies the

distribution of attention according to the expectation of the priority for processing of the

information at particular locations. More attentional resources are assigned to locations that are

expected to contain important information, resulting in greater attentional intensity at these

locations. At the neuropsychological level, the neural response is enhanced when processing the

information at these locations (Connor et al., 1997; Heinze et al., 1990; Luck et al., 1997;

Motter, 1993).

52

The enhancement in neural response when processing the information at attended

locations is achieved in two ways. One possible mechanism is to increase the baseline activity

of neurons that code information from the particular location, and this increase in activity can

even take place before stimulus onset (Colby, Duhamel & Goldberg, 1996; Luck et al., 1997). In

Experiment 1, when the size of the expected area became larger, performance on the AVF task

around the fixation declined. This suggests that when the baseline activities of more neurons (as

associated with a much larger area) have to be increased, the magnitude of the increase is

smaller. Otherwise, the performance on the AVF task around the fixation would have remained

the same regardless of the size of the expected area. The other possible mechanism may be the

suppression of information in unexpected locations (e.g. outside the expected area, or directions

other than the expected direction) (Bisley, 2011; Casagrande, Sary, Royal & Ruiz, 2005;

Desimone, 1998; Desimone & Duncan, 1995; Gazzaley, 2011; McAlonan, Cavanaugh & Wurtz,

2008; Motter, 1993). This suppression may become more effective at larger eccentricities. In

Experiment 2b, with a cue-target interval of 80 ms, the performance advantage in the expected

direction over the unexpected directions was greater at an eccentricity of 30o than at an

eccentricity of 20o. In addition, further suppression of information processing in unexpected

directions (rather than enhancement in the expected direction) was seen with longer cue-target

intervals in Experiment 2. This implies that the suppression mechanism at unexpected locations

was more dominant than the activity-increase mechanism at expected locations during the

period of 80 ms to 160 ms following the onset of the cue.

When the top-down influence is under conscious control (e.g. by expecting where the

target is likely to occur), a distributed network of higher areas implements the control processes

(Kastner & Ungerleider, 2001). This network includes the frontal eye fields (FEF),

supplementary eye fields (SEF) and superior parietal lobule (SPL) (Kastner, De Weerd,

53

Elizondo, Desimone & Ungerleider, 1998; Kastner, Pinsk, De Weerd, Desimone & Ungerleider,

1999). The network is thought to provide signals which feed back to the visual cortex to

implement top-down modulation of the spatial distribution of attention (Kastner & Ungerleider,

2001). This may help explain why the cue-target interval had greater influence from expecting

the area/direction of the target in Experiment 1 and 2. A sufficiently long cue-target interval

allows this distributed network of higher areas to generate signals and implement the top-down

influence. Thus, a longer cue-target interval (until the interval is sufficient) benefits the

modulation of the distribution of attention.

Unconscious top-down influence (bias in the spatial distribution of attention) is likely

associated with lateralization of brain structures that are involved in spatial attentional

processing (e.g. inferior parietal lobule, Frederikse et al., 1999, Marsh & Casper, 1998). A

leftward lateralization in the inferior parietal lobule may be related to a rightward bias in

distribution of attention; and in contrast, a rightward lateralization in the structure may be linked

to a leftward bias in the distribution of attention. Given lateralization of brain structures cannot

be modified by conscious control, top-down influence associated with brain lateralization is

involuntary and unconscious.

The Distribution of Attention

In this thesis, the spatial distribution of attention is conceptualized as a bivariate

probability distribution over the visual field. Top-down influences are assumed to be able to

modify the current distribution of attention at any time. When a participant expects a target to

appear anywhere within a large area, the spread of the distribution of attention is expanded, if

attention had previously been more tightly focused. Consequently, since attention is assumed to

be a fixed resource, the attentional intensity decreases at locations close to fixation and increases

at locations further away. The distribution flattens as it spreads. In addition, when a participant

54

expects a target to appear in a particular direction, the distribution of attention is shifted in that

direction. The attentional intensity increases in the expected direction while decreasing in other

directions. These conscious top-down influences operate over time until the distribution

stabilizes. Modification of the distribution of attention by conscious influences takes time to

complete, changing gradually over time, and eventually arriving at a state that reflects the top-

down influence in full.

However, there are also pre-existing spatial biases in the resting distribution of attention.

The distribution of attention is not generally symmetrical (like a bivariate Gaussian, for

instance), and is often skewed in the horizontal and/or vertical directions. In general, the

distribution of attention in men is skewed to the right and upper halves of the visual field; in

contrast, the distribution of attention in women is likely skewed to the left and upper halves of

the visual field.

Note that the model proposed in this thesis is rudimentary and was intended to provide a

simple qualitative framework to guide the experiments. Further development of the model

should include quantitative elaboration by specifying particular probability distributions and

how these are modified in response to endogenous and exogenous events.

Practical Applications of Research on Attention across the Visual Field

Investigations of selective attention across the visual field not only advances our

theoretical understanding of attentional processes, but also provides valuable insights that may

help find solutions to many practical questions (Boot, Kramer & Becic, 2007). One example is

understanding driving behaviour. Awareness of hazards across the visual field is critical to safe

driving. The size of the attentional visual field, as measured by the AVF task, has been found to

be a good predictor of driving performance (Ball et al., 1988, 1993; Ball, Owsley & Beard, 1990;

Ball, Roenker & Bruni, 1990). Shrinkage of the size of the attentional visual field due to aging

55

has been suggested as one of the main reasons for the age-related decline in driving performance

among normal older adults (Ball et al., 1988, 1993). Fortunately, expansion of the size of the

attentional visual field is possible with appropriate training (Ball et al., 1988), suggesting that

the effect of aging on visual attention may be modifiable. Increased knowledge regarding how

selective attention operates across the visual field will help in the development of effective

training tools to improve driving skills. For example, Experiments 1 & 2 demonstrated that

conscious expectation of the location of the target can significantly alter the distribution of

attention. As a result, effective training should not only focus on increasing sensitivity to

information in the visual field, but also include improving comprehension of the overall

situation and accurate prediction of the location of critical events. In addition, Experiment 3

suggested that individual difference exists in the direction of bias in the distribution of attention.

The roundedness of the attentional field has been shown to be higher among experienced

industrial inspectors than inexperienced students (Chan & Chiu, 2010). Therefore, it is

important to further investigate if the direction (and also magnitude) of bias in distribution of

attention affects driving skills.

Another practical area of great interest is computer interface design for large displays.

Recent increases in the physical size and resolution of computer displays have presented a

growing challenge to interface design (Czerwinski, Horvitz & Wilhite, 2004; Czerwinski et al.,

2003; Robertson et al., 2005; Tan, Czerwinski & Robertson, 2006). Understanding how

attention is distributed across the visual field will help in the design of information displays,

suggesting which information should be presented in which location, and at what time, for the

participant to make the most efficient use of it (Feng & Spence, 2010). For example, the results

of Experiment 3 on biases in the spatial distribution of attention suggest a hypothesis for the

spatial arrangement of information on a large display. Important information may be presented

56

where attentional processing is faster and more accurate (e.g. upper rather than the lower half of

the visual field). Because individual differences exist in the direction of bias in the distribution

of attention (e.g. leftward bias vs. rightward bias), presentation of information may vary

according to the directional bias of a particular individual. One solution is to provide several

arrangements of icons and task windows with each arrangement suitable for a known directional

bias, and a user may chose the arrangement that is most comfortable to work with.

57

Chapter 6: Contributions

This thesis contributes new information in both theoretical and practical areas of

research. These areas include theoretical modelling of attention and visual search as well as

human-computer interaction.

First, the experiments reveal how top-down influences affect attentional selection across

the visual field. The distribution of attention can be consciously modulated by the expectations

of the location of target (the size of the area in which the target is likely to occur, the likely

direction of the target). The distribution of attention is also unconsciously biased toward certain

spatial areas. A descriptive model of the distribution of attention is proposed in this thesis.

Although still at a preliminary stage, the model provides a framework for future investigation on

the distribution of attention.

Second, accurate estimation of the spatial distribution of attention around each fixation

improves precision of the models for visual search (Chan & So, 2007; Chan & Tang, 2007). The

size and shape of the distribution of attention is highly relevant to the efficiency of visual search

(Chan & So, 2007). If the spatial distribution can be accurately plotted, improved prediction of

the efficiency of visual search behaviour may be possible. For example, Dickinson & Intraub

(2009) proposed that a leftward bias in the distribution of attention among participants had

likely led to a leftward bias in the direction of the first eye movement when the participants

viewed a scene. Without incorporating the unconscious bias into the estimation of the

distribution of attention, it would be difficult to predict and explain the observation that

participants often made their first eye movement toward the left when viewing a scene.

Third, knowledge of how attention is deployed across the visual field can guide best

practices in applied areas such as interface design for large information displays. For example,

because the distribution of attention is biased toward the upper half of the visual field, it follows

58

that, important new information should be placed in the upper area of a large computer display,

since it will be more likely to attract attention in that area of the screen, and will likely be

processed more rapidly and accurately. In contrast, presenting important notifications at the

bottom of the screen, as currently practiced in software, may not be effective.

Last but not least, the individual differences in attentional bias reported in this thesis can

assist interface design for specific populations. For example, some e-commerce websites target

mostly women customers (e.g. cosmetic vendors, jewellery vendors) and there is increasing

awareness of “women-centric” design for these websites (Huang, 2005, p.75). Consideration of

the characteristics in the spatial distribution of attention among women can result in improved

placement of information and thus lead to more effective design of the websites.

59

References

Baldwin, J. M. (1889). Handbook of psychology: senses and intellect. New York: Henry Holt.

Ball, K. K., Beard, B. L., Roenker, D. L., Miller, R. L., & Griggs, D. S. (1988). Age and visual

search: expanding the useful field of view. Journal of the Optical Society of America A,

5, 2210–2219.

Ball, K. K., Owsley, C., & Beard, B. (1990). Clinical visual field perimetry underestimates

visual field problems in older adults. Clinical Vision Science, 5, 113–135.

Ball, K. K., Owsley, C., Sloane, M. E., Roenker, D. L., & Eruni, J. R. (1993). Visual attention

problems as a predictor of vehicle crashes among older drivers. Investigative

Ophthalmology & Visual Science, 34, 3110–3123.

Ball, K. K., Roenker, D. L., & Bruni, J. R. (1990). Developmental changes in attention

throughout adulthood. In J. T. Enns (Ed.), The Development of Attention: Research and

Theory. Amsterdam: Elsevier.

Baylis, G. C., & Driver, J. (1992). Visual parsing and response competition: The effect of

grouping factors. Perception & Psychophysics, 51, 145–162.

Bedard, M., Leonard, E., McAuliffe, J., Weaver, B., Gibbons, C., & Dubois, S. (2006). Visual

attention and older drivers: The contribution of inhibition of return to safe driving.

Experimental Aging Research, 32, 119–135.

Bisley, J. W. (2011). The neural basis of visual attention. Journal of Physiology, 589, 49–57.

Boot, W. R., Kramer, A. F., & Becic, E. (2007). Capturing attention in the laboratory and the

real world. In A. F. Kramer, A. D. Wiegmann, & A. Kirlik. (Eds.), Attention: from

theory to practice (pp. 27–44). New York, NY: Oxford University Press.

Bowers, D., & Heilman, K. M. (1980). Pseudoneglect: effects of hemispace on a tactile line

bisection task. Neuropsychologia, 18, 491–498.

60

Bradshaw, J. L., Nettleton, N. C., Nathan, G., & Wilson, L. E. (1983). Head and body space to

left and right, front and rear. II. Visuotactual and kinaesthetic studies and left-side under-

estimation. Neuropsychologia, 21, 475–486.

Bradshaw, J. L., Nettleton, N. C., Nathan, G., & Wilson, L. E. (1985). Bisecting rods and lines:

effects of horizontal and vertical posture on left-side underestimation by normal subjects.

Neuropsychologia, 23, 421–425.

Bradshaw, J. L., Nettleton, N. C., Wilson, L. E., & Bradshaw, C. S. (1987). Line bisection by

left-handed preschoolers: a phenomenon of symmetrical neglect. Brain and Cognition, 6,

377–385.

Broadbent, D. E. (1958). Perception and Communication. London: Pergamon Press.

Brodie, E. E. (2010). Reflecting upon line bisection: mirror reversal increases in the magnitude

of pseudoneglect. Neuropsychologia, 48, 1517–1520.

Broman, A. T., West, S. K., Muňoz, B., Bandeen-Roche, K., Rubin, G. S., & Turano, K. A.

(2004). Divided visual attention as a predictor of bumping while walking: the Salisbury

Eye Evaluation. Investigative Ophthalmology & Visual Science, 45, 2955–2960.

Bundesen, C. (1990). A theory of visual attention. Psychological Review, 97, 523–547.

Bundesen, C., Pedersen, L. F., & Larsen, A. (1984). Measuring efficiency of selection from

briefly exposed visual displays: A model for partial report. Journal of Experimental

Psychology: Human Perception and Performance, 10, 329–339.

Casagrande, V. A., Sary, G., Royal, D., & Ruiz, O. (2005). On the impact of attention and motor

planning on the lateral geniculate nucleus. Progress in Brain Research, 149, 11–29.

Cave, K. R., & Pashler, H. (1995). Visual selection mediated by location: Selecting successive

visual objects. Perception & Psychophysics, 57, 421–432.

61

Cavanagh, P., & Alvarez, G. A. (2005). Tracking multiple targets with multifocal attention.

Trends in Cognitive Sciences, 9, 349–354.

Cepeda, N., J., Cave, K. R., Bichot, N. P., & Kim, M. S. (1998). Spatial selection via feature-

driven inhibition of distractor locations. Perception and Psychophysics, 60, 727–746.

Chambers, K. W., McBeath, M. K., Schiano, D. J., & Metz, E. G. (1999). Tops are more salient

than bottoms. Perception and Psychophysics, 61, 625–635.

Chan, A. H. S., & Chiu, C. H. Y. (2010). Visual lobe shape characteristics of experienced

industrial inspectors and inexperienced subjects. Human Factors and Ergonomics in

Manufacturing & Service Industries, 20, 367–377.

Chan, A. H. S., Courtney, A. J., & Ma, C. W. (2002). Visual performance on detection tasks

with double-targets of the same and different difficulty. Ergonomics, 45, 934–948.

Chan, A. H. S., & So, D. K. T. (2007). Shape characteristics of useful field of view and visual

search time. Le travail humain, 70, 343–367.

Chan, A. H. S., & Tang, N. Y. (2007). Visual lobe shape and search performance for targets of

different difficulty. Ergonomics, 50, 289–318.

Chun, M. M., Golomb, J. D., & Turk-Brown, N. B. (2011). A taxonomy of external and internal

attention. Annual Review of Psychology, 62, 73–101.

Clay, O. J., Wadley, V. G., Edwards, J. D., Roth, D. L., Roenker, D. L., & Bail, K. K. (2005).

Cumulative meta-analysis of the relationship between useful field of view and driving

performance in older adults: current and future implications. Optometry and Vision

Science, 82, 724–731.

Colby, C. L., Dyhamel, J. R., & Goldberg, M. E. (1996). Visual, presaccadic, and cognitive

activation of single neurons in monkey lateral intraparietal area. Journal of

Neurophysiology, 76, 2841–2852.

62

Colvin, J., Tobler, N., & Anderson, J. A. (2004). Productivity and multi-screen computer

displays. Rocky Mountain Communication Review, 2, 31–53.

Connor, C. E., Gallant, J. L., Preddie, D. C., & Van Essen, D. C. (1996). Responses in area V4

depend on the spatial relationship between stimulus and attention. Journal of

Neurophysiology, 75, 1306–1308.

Connor, C. E., Preddie, D. C., Gallant, J. L., & Van Essen, D. C. (1997). Spatial attention effects

in macaque area V4. Journal of Neuroscience, 17, 3201–3214.

Corbetta, M., Kincade, J. M., Ollinger, J. M., McAvoy, M. P., & Shulman, G. L. (2000).

Voluntary orienting is dissociated from target detection in human posterior parietal

cortex. Nature Neuroscience, 3, 292–297.

Czerwinski, M., Horvitz, E., & Wilhite, S. (2004). A diary study of task switching and

interruption. In Proceedings of the SIGCHI Conference on Human Factors in

Computing Systems (pp. 175–182). Vienna, Austria.

Czerwinski, M., Smith, G., Regan, T., Meyers, B., Robertson, G., & Starkweather, G. (2003).

Toward characterizing the productivity benefits of very large displays. In Proceedings of

INTERACT 2003 (pp. 9–16). Amsterdam, Netherland: IOS Press.

Daini, R., Arduino, L. S., Menza, D. D., Vallar, G., & Silveri, M. C. (2008). Line bisection and

cerebellar damage. Neuropsychiatry, neuropsychology, and behavioral neurology, 21,

214–220.

Desimone, R. (1998). Visual attention mediated by biased competition in extrastriate visual

cortex. Philosophical Transactions of the Royal Society of London, 353, 1245–1255.

Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual attention. Annual

Review of Neuroscience, 8, 753–758.

63

Deutsch, J. A., & Deutsch, D. (1963). Attention: some theoretical considerations. Psychological

Review, 70, 80–90.

Dickinson, C. A., & Intraub, H. (2009). Spatial asymmetries in viewing and remembering

scenes: consequences of an attentional bias? Attention, Perception & Psychophysics, 71,

1251–1262.

Downing, C. J., & Pinker, S. (1985). The spatial structure of visual attention. In M. I. Posner &

O. S. M. Marin (Eds.), Attention & Performance (Vol.11, pp.171–188). Hillsdale, NJ:

Erlbaum.

Drain, M., & Reuter-Lorenz, P. A. (1996). Vertical orienting bias: evidence for attentional bias

and „neglect‟ in the intact brain. Journal of Experimental Psychology: General, 125,

139–158.

Eriksen, C. W., & Hoffman, J. E. (1973). The extent of processing of noise elements during

selective encoding from visual displays. Perception & Psychophysics, 14, 155–160.

Eriksen, C. W., & St. James, J. D. (1986). Visual attention within and around the field of focal

attention: A zoom lens model. Perception & Psychophysics, 40, 225–240.

Eriksen, C. W., & Yeh, Y. (1985). Allocation of attention in the visual field. Journal of

Experimental Psychology: Human Perception and Performance, 11, 583–597.

Feng, J., & Spence, I. (2010). Left or right? Spatial arrangement for information presentation on

large displays. In A. W. Kark (Eds.), IBM CASCON 2010 Proceedings, Toronto,

Canada.

Feng, J., Spence, I., & Pratt, J. (2007). Playing an action video game reduces gender differences

in spatial cognition. Psychological Science, 18, 850–855.

Fortenbaugh, F. C., Hicks, J. C., Hao, L., & Turano, K. A. (2006). High-speed navigators: using

more than what meets the eye. Journal of Vision, 6, 565–579.

64

Foxe, J. J., McCourt, M. E., & Javitt, D. C. (2003). Right hemisphere control of visuospatial

attention: line-bisection judgements evaluated with high-density electrical mapping and

source analysis. Neuroimage, 19,710–726.

Frederikse, M. E., Lu, A., Aylward, E., Barta, P., & Pearlson, G. (1999). Sex differences in the

inferior parietal lobule. Cerebral Cortex, 9, 896–901.

Gazzaley, A. (2011). Influence of early attentional modulation on working memory.

Neuropsychologia, 49, 1410–1424.

Gazzaley, A., Clapp, W., McEvoy, K., Knight, R., & D‟Esposito, M. (2008). Age-related top-

down suppression deficit in the early stages of cortical visual memory processing.

Proceedings of the National Academy of Sciences, 105, 13122–13126.

Gibson, J. J. (1941). A critical review of the concept of set in contemporary psychology.

Psychological Bulletin, 38, 781–817.

Green, C. S., & Bavelier, D. (2003). Action video game modifies visual selective attention.

Nature, 423, 534–537.

Hansen, T., Pracejus, L., & Gegenfurtner, K. R. (2009). Color perception in the intermediate

periphery of the visual field. Journal of Vision, 9, 1–12.

Hatfield, G. (1998). Attention in early scientific psychology. In R. D. Wright (Ed.), Visual

Attention (pp.3–25). New York: Oxford University Press.

Heilman, K. M., & Van Den Abell, T. (1980). Right hemisphere dominance for attention: the

mechanism underlying hemispheric asymmetries of inattention (neglect). Neurology, 30,

327–330.

Heilman, K. M., Waston, R. T., Valenstein, E., & Damasio, A. R. (1993). Localization of

lesions in neglect. In A. Kertez (Ed.), Localization in neuropsychology (pp. 471–492).

New York: Academic Books.

65

Heinze, H. J., Luck, S. J., Mangun, G. R., & Hillyard, S. A. (1990). Visual event-related

potentials index focused attention within bilateral arrays. I. Evidence for early selection.

Electroencephalography and Clinical Neurophysiology, 75, 511–527.

Hellige, J. B. (2001). Hemispheric asymmetry. Cambridge, MA: Harvard University Press.

Herber, I. A., Siebertz, S., Wolter, M., Kuhlen, T., & Fimm, B. (2010). Horizontal and vertical

pseudoneglect in peri- and extrapersonal space. Brain and Cognition, 73, 160–166.

Hillis, A. E., Newhart, M., Heidler, J., Barker, P. B., Herskovits, E. H., & Degaonkar, M. (2005).

Anatomy of spatial attention: Insights from perfusion imaging and hemispatial neglect in

acute stroke. Journal of Neuroscience, 25, 3161–3167.

Huang, E. (2005). The acceptance of women-centric websites. Journal of Computer Information

Systems, 45, 75–83.

Ikeda, M., & Takeuchi, T. (1975). Influence of foveal load on the functional visual field.

Perception & Psychophysics, 18, 255–260.

Intraub, H., Daniels, K. K., Horowitz, T. S., & Wolfe, J. M. (2008). Looking at scenes while

searching for numbers: dividing attention multiplies space. Perception & Psychophysics,

70, 1337–1349.

Itti, L., & Koch, C. (2001). Computational modelling of visual attention. Nature Reviews

Neuroscience, 2, 194–203.

James, W. (1890). Principle of psychology (Vols. 1-2). New York: Dover (reprinted in 1950).

Jeerakathil, T. J., & Kirk, A. (1994). A representational vertical bias. Neurology, 44, 703–706.

Jewell, G., & McCourt, M. E. (2000). Pseudoneglect: a review and meta-analysis of

performance factors in line bisection tasks. Neuropsychologia, 38, 93–110.

66

Jonides, J. (1981). Voluntary versus automatic control over the mind‟s eye‟s movement. In J. B.

Long & A. D. Baddeley (Eds.), Attention & Performance (Vol.9, pp. 187–203).

Hillsdale, NJ: Erlbaum.

Johnson, A., & Proctor, R. W. (2004). Attention: Theory and Practice. Thousand Oaks, CA:

Sage Publications.

Karnath, H. -O., Ferber, S., & Himmelbach, M. (2001). Spatial awareness is a function of the

temporal not the posterior parietal lobe. Nature, 411, 950–953.

Kastner, S., De Weerd, P., Elizondo, I., Desimone, R., & Ungerleider, L. G. (1998).

Mechanisms of spatial attention in human extrastriate cortex as revealed by functional

MRI. Society of Neuroscience Abstracts, 24, 1249.

Kastner, S., Pinsk, M. A., De Weerd, p., Desimone, R., & Ungerleider, L. G. (1999). Increased

activity in human visual cortex during directed attention in the absence of visual

stimulation. Neuron, 22, 751–761.

Kastner, S., & Ungerleider, L. G. (2001). The neural basis of biased competition in human

visual cortex. Neuropsychologia, 39, 1263–1276.

Keating, E. G., & Gooley, S. G. (1988). Disconnection of parietal and occipital access to the

saccadic oculomotor system. Experimental Brain Research, 70, 385–398.

Kimberg, D. Y., & Farah, M. J. (2000). Is there an inhibitory module in the prefrontal cortex?

Working memory and the mechanisms underlying cognitive control. In S. Monsell & J.

Driver (Eds.), Control of cognitive processes: Attention and performance XVII (pp. 739–

751). Cambridge, MA: MIT Press.

Kimura, D. (2000). Sex and Cognition. Cambridge, MA: MIT Press.

Koch, C., & Ullman, S. (1985). Shifts in selective visual attention: towards underlying neural

circuitry. Human Neurobiology, 4, 219–227.

67

Koscik, T., O‟Leary, D., Moser, D. J., Andreasen, N. C., & Nopoulos, P. (2009). Sex

differences in parietal lobe morphology: Relationship to mental rotation performance.

Brain and Cognition, 69, 451–459.

Kröse, B. J. A., & Julesz, B. (1989). The control and speed of shifts of attention. Vision

Research, 29, 1607–1619.

LaBerge, D., & Brown, V. (1989). Theory of attentional operations in shape identification.

Psychological Review, 96, 101–124.

LaBerge, D., Carlson, R. L., Williams, J. K., & Bunny, B. G. (1997). Shifting attention in visual

space: Tests of moving-spotlight models versus an activity-distribution model. Journal

of Experimental Psychology: Human Perception and Performance, 23, 1380–1392.

Lajoie, Y., Teasdale, N., Bard, C., & Fleury, M. (1996). Attentional demands for walking: Age-

related changes. In A. M. Ferrandez, & N. Teasdale (Eds.), Changes in Sensory Motor

Behavior in Aging (pp. 235–256). Amsterdam: Elsevier.

Lambert, A., Spencer, E., & Mohindra, N. (1987). Automaticity and the capture of attention by

a peripheral display change. Current Psychology, 6, 136–147.

Luck, S. J., Chelazzi, L., Hillyard, S. A., & Desimone, R. (1997). Neural mechanisms of spatial

selective attention in areas V1, V2, and V4 of macaque visual cortex. Journal of

Neurophysiology, 77, 24–42.

Luck, S. J., Fan, S., & Hillyards, S. A. (1993). Attention-related modulation of sensory evoked

brain activity in a visual search task. Journal of Cognitive Neuroscience, 5, 188–195.

Mangun, G. R., Buonocore, M. H., Girelli, M., & Jha, A. P. (1998). ERP and fMRI measures of

visual spatial selective attention. Human Brain Mapping, 6, 383–389.

68

Manning, L., Halligan, P. W., Marshall, J. C. (1990). Individual variation in line bisection: a

study of normal subjects with application to the interpretation of visual neglect.

Neuropsychologia, 28, 647–655.

Marsh, L., Casper, R. (1998). Gender differences in brain morphology and in psychiatric

disorder. In R. C. Capser (Ed.), Women’s health: hormones, emotions and behavior

(pp.53–82). Cambridge: Cambridge University Press.

McAlonan, K., Cavanaugh, J., & Wurtz, R. H. (2008). Guarding the gateway to cortex with

attention in visual thalamus. Nature, 456, 391–394.

McCourt, M. E., & Jewell, G. (1999). Visuospatial attention in line bisection: stimulus

modulation of pseudoneglect. Neuropsychologia, 37, 843–855.

McCourt, M. E., & Olafson, C. (1997). Cognitive and perceptual influences on visual line

bisection: psychophysical and chronometric analyses of pseudoneglect.

Neuropsychologia, 35, 369–380.

Mefferd, R. B., Wieland, B. A., & Difilho, L. P. (1969). Systematic alternations of the apparent

centers of lines. Perceptual and Motor Skills, 28, 803–825.

Mesulam, M. M. (1998). From sensation to cognition. Brain, 121, 1013–1052.

Miles, W.R. (1930). Ocular dominance in human adults. Journal of General Psychology, 3,

412–429.

Moore, C. M., Yantis, S., & Vaughan, B. (1998). Object-based visual selection: evidence from

perceptual completion. Psychological Science, 9, 104–110.

Motter, B. C. (1993). Focal attention produces spatially selective processing in visual cortical

areas V1, V2, and V4 in the presence of competing stimuli. Journal of Neurophysiology,

70, 909–919.

69

Müller, H. J., & Findlay, J. M. (1988). The effect of visual attention on peripheral

discrimination thresholds in single and multiple element displays. Acta Psychologia, 69,

129–155.

Mülle, N. G., Mollenhauer, M., Rösler, A., & Kleinschmidt, A. (2005). The attentional field has

a Mexican hat distribution. Vision Research, 45, 1129–1137.

Myers, R. S., Ball, K. K., Kalina, T. D., Roth, D. L., & Goode, K. T. (2000). Relation of useful

field of view and other screening tests to on-road driving performance. Perception &

Motor Skills, 91, 279–290.

Neumann, O. (1987). Beyond capacity: A functional view of attention. In H. Heuer & A. F.

Sanders (Eds.), Perspectives on perception and action (pp.361–394). Hillsdale, NJ:

Erlbaum.

Niemeier, M., Singh, V. V. W., Keough, M., & Akbar, N. (2008a). The perceptual

consequences of the attentional bias: evidence for distractor removal. Experimental

Brain Research, 189, 411–420.

Niemeier M., Stojanoski, B., & Greco, A. L. (2007). Influences of time and spatial frequency on

the perceptual bias: Evidence for competition between hemispheres. Neuropsychologia,

45, 1029–1040.

Niemeier, M., Stojanoski, B., Singh, V. W. A., & Chu, E. (2008b). Paradoxical cross-over due

to attention to high or low spatial frequencies. Brain and Cognition, 67, 115–125.

Norman, D. A. (1968). Towards a theory of memory and attention. Psychological Review, 75,

522–536.

Owsley, C., & McGwin, G. (2004). Association between visual attention and mobility in older

adults. Journal of the American Geriatrics Society, 52, 1901–1906.

70

Owsley, C., McGwin, G., Sloane, M. E., Stalvey, B. T., & Wells, J. (2001). Timed instrumental

activities of daily living (TIADL) tasks: Relationship to visual function in older adults.

Optometry & Vision Science, 78, 350–359.

Palmer, J. (1990). Attentional limits on the perception and memory of visual information.

Journal of Experimental Psychology: Human Perception and Performance, 16, 332–

350.

Palmer, S. M., & Rosa, M. G. (2006). A distinct anatomical network of cortical areas for

analysis of motion in far peripheral vision. European Journal of Neuroscience, 24,

2389–2405.

Peterson, S. E., Robinson, D. L., & Currie, J. N. (1989). Influences of lesions of parietal cortex

on visual spatial attention in humans. Experimental Brain Research, 76, 267–280.

Posner, M. I. (1978). Chronometric exploration of mind. Hillsdale, NJ: Erlbaum.

Posner, M. I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32,

3–25.

Ponser, M. I., & Cohen, Y. (1984). Components of visual attention. In H. Bouma & D. G.

Bouhuis (Eds.), Attention & Performance (Vol.10, pp.531–556). Hillsdale, HJ: Erlbaum.

Posner, M. I., Snyder, C. R. R., & Davidson, B. J. (1980). Attention and the detection of signals.

Journal of Experimental Psychology: General, 109, 106–174.

Previc, F. H. (1990). Functional specialization in the lower and upper visual fields in humans:

its ecological origins and neurophysiological implications. Behavioral and Brain

Sciences, 13, 519–575.

Previc, F. H. (1998). The neuropsychology of 3-D space. Psychological Bulletin, 124, 123–164.

71

Previc, F. H., Beer, J., Liotti, M., Blakemore, C., & Fox, P. (2000). Is “ambient vision”

distributed in the brain? Effects of wide-field-of-view visual yaw motion on PET

activation. Journal of Vestibular Research, 10, 221–225.

Previc, F. H., & Blume, J. L. (1993). Visual search asymmetries in three-dimensional space.

Vision Research, 33, 2697–2704.

Pylyshyn, Z. (1999). Is vision continuous with cognition? The case for cognitive impenetrability

of vision perception. Behavioral and Brain Sciences, 22, 341–423.

Pylyshyn, Z. W., & Storm, R. W. (1988). Tracking multiple independent targets: Evidence for a

parallel tracking mechanism. Spatial Vision, 3, 179–197.

Raichle, M. E. (2010, March 1). The brain‟s dark energy. Scientific American, 44–49.

Remington, R., & Pierce, L. (1984). Moving attention: evidence for time-irrelevant shifts of

visual selective attention. Perception & Psychophysics, 35, 393–399.

Robertson, G., Czerwinski, M., Baudisch, P., Meyers, B., Robbins, D., Smith, G., et al. (2005).

The large-display user experience. IEEE Computer Graphics and Applications:

Applications of Large Displays, 25, 44–51.

Robertson, I. H., & Marshall, J. C. (1993). Unilateral neglect: Clinical and experimental studies.

Hove, UK: Erlbaum.

Roth, H. L., Lora, A. N., & Heilman, K. M. (2002). Effects of monocular viewing and eye

dominance on spatial attention. Brain, 125, 2023–2035.

Ruff, H. A., & Capozzoli, M. C. (2003). Development of attention and distractibility in the first

4 years of life. Developmental Psychology, 39, 877–890.

Ruff, H.A. & Lawson, K.R. (1990). Development of sustained, focused attention in young

children during free play. Developmental Psychology, 26, 85–93.

72

Sato, T. R., Watanabe, K., Thompson, K. G., & Schall, J. D. (2003). Effect of target-distractor

similarity on FEF visual selection in the absence of the target. Experimental Brain

Research, 151, 356–363.

Sears, C. R., & Pylyshyn, Z. W. (2000). Multiple object tracking and attentional processing.

Canadian Journal of Experimental Psychology, 54, 1–14.

Shapiro, K. L., & Lim, A. (1989). The impact of anxiety on visual attention to central and

peripheral visual stimuli. Behaviour Research and Therapy, 27, 345–351.

Shepherd, M., & Müller, H. J. (1989). Movement versus focusing of visual attention. Perception

& Psychophysics, 46, 146–154.

Shih, S. I., & Sperling, G. (1996). Is there feature-based attentional selection in visual search?

Journal of Experimental Psychology: Human Perception and Performance, 22, 758–779.

Sparks, D., Rohrer, W. H., & Zhang, Y. (2000). The role of the superior colliculus in saccade

initiation: a study of express saccades and the gap effect. Vision Research, 40, 2763–

2777.

Sperling, G., & Dosher, B. A. (1986). Strategy and optimization in human information

processing. In K. Boff, L. Kauffman, & J. Thomas (Eds.), Handbook of perception and

human performance (pp.1–85). New York: John Wiley.

Tan, D. S., Czerwinski, M., & Robertson, G. (2006). Large displays enhance optical flow cues

and narrow the gender gap in 3D virtual navigation. The Journal of the Human Factors

and Ergonomics Society, 48, 318–333.

Theeuwes, J. (1989). Effects of location and form cuing on the allocation of attention in the

visual field. Acta Psychologia, 72, 177–192.

Titchener, E. B. (1908). Lectures on the elementary psychology of feeling and attention. New

York: Macmillan.

73

Treisman, A. (1960). Contextual cues in selective listening. Quarterly Journal of Experimental

Psychology, 12, 242–248.

Treisman, A., & Riley, J. G. A. (1969). Is selective attention selective perception or selective

response? A further test. Journal of Experimental Psychology, 79, 27–34.

Treisman, A., & Gelade, G. (1980). A feature-integration theory of attention. Cognitive

Psychology, 12, 97–136.

Trick, L. M., & Pylyshyn, Z. W. (1993). What enumeration studies can show us about spatial

attention: evidence for limited capacity preattentive processing. Journal of Experimental

Psychology: Human Perception & Performance, 19, 331–351.

Trick, L. M., & Pylyshyn, Z. W. (1994). Why are small and large numbers enumerated

differently? A limited-capacity preattentive stage in vision. Psychological Review, 10,

80–102.

Tsal, Y., & Lamy, D. (2000). Attending to an object‟s color entails attending to its location:

support for location special views of visual attention. Perception & Psychophysics, 62,

960–968.

Tsal, Y., & Lavie, N. (1993). Location dominance in attending to color and shape. Journal of

Experimental Psychology: Human Perception and Performance, 19, 131–139.

Van der Heijden, A. H. C. (1993). The role of position in object selection in vision.

Psychological Research, 56, 44–58.

Van Vugt, P., Fransen, I., Creten, W., & Paquier, P. (2000). Line bisection performances of 650

normal children. Neuropsychologia, 38, 886–895.

Vandenberghe, R., Duncan, J., Dupont, P., Ward, R., Poline, J. –B., Bormans, G., et al. (1997).

Attention to one or two features in left or right visual field: A position emission

tomography study. Journal of Neuroscience, 17, 3739–3750.

74

Vandenberghe, R., Duncan, J., Arnell, K. M., Bishop, S. J., Herrod, N. J., Owen, A. M., et al.

(2000). Maintaining and shifting attention within right or left hemifield. Cerebral Cortex,

10, 706–713.

Varnava, A., & Halligan, P. W. (2007). Influence of age and sex on line bisection: A study of

normal performance with implications for visuospatial neglect. Aging, Neuropsychology,

and Cognition, 14, 571–585.

Wendelken, C., Baym, C. L., Gazzaley, A., & Bunge, S. A. (2011). Neural evidence of weaker

attentional modulation in children relative to young adults. Developmental Cognitive

Neuroscience, 1, 175–186.

Wickens, C. D., & Hollands, J. G. (1999). Engineering Psychology and Human Performance

(3rd

Ed.). Upper Saddle River, NJ: Prentice Hall.

Williams, L. J. (1995). Visual field tunneling in aviators induced by memory demands. The

Journal of General Psychology, 122, 225–235.

Wolff, C. (1738). Psychologia empirica. Frankfurt & Leipzig: Officina Libraria Rengeriana.

Wolff, C. (1740). Psychologia rationalis. Frankfurt & Leipzig: Officina Libraria Rengeriana.

Wood, J. M., & Troutbeck, R. (1995). Elderly drivers and simulated visual impairment.

Optometry & Vision Science, 72, 115–124.

Wright, R. D., & Ward, L. M. (2008). Orienting of Attention. New York, NY: Oxford

University Press.

75

Figures

Figure 1. An illustration of the distribution of attention across the visual field as a bivariate

Gaussian.

76

Figure 2. Illustrations of top-down influences on the distribution of attention across the visual

field. a. a concentrated distribution with a small attended area; b. a spreaded distribution with a

large attended area; c. the bimodal distribution during endogenous covert orienting of attention;

d. the skewed distribution when a unconscious spatial bias is present.

77

Figure 3. A sample trial of the Attentional Visual Field (AVF) task with an endogenous cue

indicating the size of the area in which the target was likely to appear in Experiment 1A. A cue

was given immediately following the fixation in each trial.

78

Figure 4. Accuracies (left panel) and reaction times (RTs) (right panel) on the AVF task with an

area size cue in Experiment 1A.

79

Figure 5. A sample trial of the AVF task with a cue indicating the size of the area in which the

target was likely to appear in Experiment 1B. A area size cue was given in the beginning of each

block before the trials start.

80

Figure 6. Accuracies (left panel) and reaction times (RTs) (right panel) on the AVF task with an

area size cue in Experiment 1B.

81

Figure 7. A sample trial of the AVF task with an endogenous cue indicating the likely direction

of the target in Experiment 2A.

82

Figure 8. Accuracies and reaction times (RTs) on the attention orienting AVF task in

Experiment 2A. a. accuracies, including overall accuracies across intervals (left panel) and

accuracies at each interval across eccentricities (three right panels); b. RT, including overall RTs

across intervals (left panel) and RTs at each interval across eccentricities (three right panels).

83

Figure 9. A sample trial of the AVF identification task with an endogenous cue indicating the

direction of the target in Experiment 2B. Participants reported the identity of the target.

84

Figure 10. Accuracies and reaction times (RTs) on the attention orienting AVF identification

task in Experiment 2B. a. accuracies at each interval across eccentricities; b. RTs at each

interval across eccentricities.

85

Figure 11.A sample trial of the AVF task with various exposures in Experiment 3.

86

Figure 12. Accuracies and reaction times (RTs) on the AVF task in Experiment 3. a. accuracy

comparison between the left and right halves of the visual field, for men (upper left panel) and

for women (upper right panel); and between the upper and lower halves of the visual field, for

men (lower left panel) and for women (lower right panel); b. RT comparison between the left

and right halves of the visual field, for men (upper left panel) and for women (upper right panel),

and between the upper and lower halves of the visual field, for men (lower left panel) and for

women (lower right panel).

87

Figure 13. Accuracies (left panel) and reaction times (RTs) (right panel) across exposures by

eccentricities on the attention orienting AVF task (left vs. right analysis) in Experiment 3.

88

Figure 14. Accuracies (left panel) and reaction times (RTs) (right panel) across exposures by

eccentricities on the attention orienting AVF task (upper vs. lower analysis) in Experiment 3.