Citation
The Role of the Quiet-Eye Period and the Bereitschaftspotential in Arousal Regulation and Motor Preparation for Performance of a Self-Paced Motor Skill

Material Information

Title:
The Role of the Quiet-Eye Period and the Bereitschaftspotential in Arousal Regulation and Motor Preparation for Performance of a Self-Paced Motor Skill
Creator:
Mann, Derek Thomas Yonge
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (190 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Health and Human Performance
Applied Physiology and Kinesiology
Committee Chair:
Janelle, Christophe M.
Committee Members:
Cauraugh, James H.
Tillman, Mark D.
Linderholm, Tracy A.
Graduation Date:
12/14/2007

Subjects

Subjects / Keywords:
Anticipation ( jstor )
Anxiety ( jstor )
Cognitive psychology ( jstor )
Hemispheres ( jstor )
Paradigms ( jstor )
Performing artists ( jstor )
Soccer ( jstor )
Sports psychology ( jstor )
Tennis ( jstor )
Visual fixation ( jstor )
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
bereitschaftspotential, expertise, eye, golf, quiet, sport
Genre:
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Health and Human Performance thesis, Ph.D.

Notes

Abstract:
Given the robust empirical support for and practical implications of the quiet-eye (QE) period, it was my objective to assess the role of the QE period in emotion regulation and motor preparation. The concurrent exploration of the BP and QE period under varying levels of anxiety was designed to assess the principal mechanisms responsible for the psychomotor differences between expert and near-expert performers. Twenty golfers were classified by their USGA handicap rating as either a high handicap (HH; near-expert) or low handicap (LH; expert) to permit skill-based inferences. Participants completed 45 trials in both low and high anxiety conditions during which cognitive anxiety, somatic anxiety, heart rate, QE duration, BP activity, and putting performance were recorded. Results indicated that the LH golfers are more accurate and less variable in their performance than the HH group, as revealed by measures of radial error, bivariate variable error, and group centroid radial error. Systematic differences in QE duration and BP were also observed, with experts exhibiting a prolonged quiet eye period and greater cortical activation in the right-central region compared to non-experts. A significant association between cortical activation and QE duration was also noted. Despite performing under high and low anxiety conditions, QE duration and cortical activation did not fluctuate across conditions. Taken together, the results of this investigation lend primary support to the motor programming/motor preparation function of the QE period. Practical and theoretical implications are presented and suggestions for empirical work provided. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2007.
Local:
Adviser: Janelle, Christophe M.
Statement of Responsibility:
by Derek Thomas Yonge Mann.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Mann, Derek Thomas Yonge. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
662634953 ( OCLC )
Classification:
LD1780 2007 ( lcc )

Downloads

This item has the following downloads:

mann_d.pdf

mann_d_Page_087.txt

mann_d_Page_075.txt

mann_d_Page_081.txt

mann_d_Page_073.txt

mann_d_Page_151.txt

mann_d_Page_190.txt

mann_d_Page_014.txt

mann_d_Page_113.txt

mann_d_Page_023.txt

mann_d_Page_137.txt

mann_d_Page_171.txt

mann_d_Page_052.txt

mann_d_Page_111.txt

mann_d_Page_160.txt

mann_d_Page_065.txt

mann_d_Page_046.txt

mann_d_Page_031.txt

mann_d_Page_157.txt

mann_d_Page_185.txt

mann_d_Page_130.txt

mann_d_Page_036.txt

mann_d_Page_123.txt

mann_d_Page_041.txt

mann_d_Page_057.txt

mann_d_Page_156.txt

mann_d_Page_181.txt

mann_d_Page_187.txt

mann_d_Page_121.txt

mann_d_Page_043.txt

mann_d_Page_093.txt

mann_d_Page_136.txt

mann_d_Page_001.txt

mann_d_Page_152.txt

mann_d_Page_147.txt

mann_d_Page_044.txt

mann_d_Page_011.txt

mann_d_Page_102.txt

mann_d_Page_058.txt

mann_d_Page_189.txt

mann_d_Page_038.txt

mann_d_Page_144.txt

mann_d_Page_077.txt

mann_d_Page_165.txt

mann_d_Page_069.txt

mann_d_Page_028.txt

mann_d_Page_032.txt

mann_d_Page_167.txt

mann_d_Page_094.txt

mann_d_Page_163.txt

mann_d_Page_115.txt

mann_d_Page_020.txt

mann_d_Page_143.txt

mann_d_Page_168.txt

mann_d_Page_100.txt

mann_d_Page_010.txt

mann_d_Page_169.txt

mann_d_Page_016.txt

mann_d_Page_120.txt

mann_d_Page_053.txt

mann_d_Page_066.txt

mann_d_Page_139.txt

mann_d_Page_019.txt

mann_d_Page_008.txt

mann_d_Page_173.txt

mann_d_Page_029.txt

mann_d_Page_015.txt

mann_d_Page_040.txt

mann_d_Page_176.txt

mann_d_Page_116.txt

mann_d_Page_155.txt

mann_d_Page_090.txt

mann_d_Page_034.txt

mann_d_Page_112.txt

mann_d_Page_158.txt

mann_d_Page_009.txt

mann_d_Page_072.txt

mann_d_Page_086.txt

mann_d_Page_178.txt

mann_d_Page_107.txt

mann_d_Page_174.txt

mann_d_Page_079.txt

mann_d_Page_180.txt

mann_d_Page_054.txt

mann_d_Page_061.txt

mann_d_Page_146.txt

mann_d_Page_042.txt

mann_d_Page_083.txt

mann_d_Page_026.txt

mann_d_Page_082.txt

mann_d_Page_084.txt

mann_d_Page_003.txt

mann_d_pdf.txt

mann_d_Page_048.txt

mann_d_Page_074.txt

mann_d_Page_140.txt

mann_d_Page_051.txt

mann_d_Page_070.txt

mann_d_Page_078.txt

mann_d_Page_037.txt

mann_d_Page_005.txt

mann_d_Page_108.txt

mann_d_Page_035.txt

mann_d_Page_060.txt

mann_d_Page_055.txt

mann_d_Page_110.txt

mann_d_Page_161.txt

mann_d_Page_109.txt

mann_d_Page_007.txt

mann_d_Page_148.txt

mann_d_Page_085.txt

mann_d_Page_101.txt

mann_d_Page_045.txt

mann_d_Page_022.txt

mann_d_Page_145.txt

mann_d_Page_063.txt

mann_d_Page_150.txt

mann_d_Page_018.txt

mann_d_Page_182.txt

mann_d_Page_049.txt

mann_d_Page_097.txt

mann_d_Page_128.txt

mann_d_Page_088.txt

mann_d_Page_017.txt

mann_d_Page_177.txt

mann_d_Page_068.txt

mann_d_Page_104.txt

mann_d_Page_135.txt

mann_d_Page_134.txt

mann_d_Page_062.txt

mann_d_Page_030.txt

mann_d_Page_064.txt

mann_d_Page_184.txt

mann_d_Page_133.txt

mann_d_Page_067.txt

mann_d_Page_076.txt

mann_d_Page_096.txt

mann_d_Page_186.txt

mann_d_Page_129.txt

mann_d_Page_080.txt

mann_d_Page_132.txt

mann_d_Page_154.txt

mann_d_Page_059.txt

mann_d_Page_172.txt

mann_d_Page_025.txt

mann_d_Page_188.txt

mann_d_Page_166.txt

mann_d_Page_089.txt

mann_d_Page_105.txt

mann_d_Page_114.txt

mann_d_Page_179.txt

mann_d_Page_131.txt

mann_d_Page_047.txt

mann_d_Page_124.txt

mann_d_Page_039.txt

mann_d_Page_013.txt

mann_d_Page_175.txt

mann_d_Page_004.txt

mann_d_Page_092.txt

mann_d_Page_021.txt

mann_d_Page_149.txt

mann_d_Page_095.txt

mann_d_Page_162.txt

mann_d_Page_024.txt

mann_d_Page_118.txt

mann_d_Page_125.txt

mann_d_Page_126.txt

mann_d_Page_056.txt

mann_d_Page_106.txt

mann_d_Page_159.txt

mann_d_Page_002.txt

mann_d_Page_117.txt

mann_d_Page_138.txt

mann_d_Page_103.txt

mann_d_Page_091.txt

mann_d_Page_012.txt

mann_d_Page_119.txt

mann_d_Page_142.txt

mann_d_Page_127.txt

mann_d_Page_027.txt

mann_d_Page_170.txt

mann_d_Page_153.txt

mann_d_Page_122.txt

mann_d_Page_183.txt

mann_d_Page_071.txt

mann_d_Page_164.txt

mann_d_Page_141.txt

mann_d_Page_006.txt

mann_d_Page_050.txt

mann_d_Page_098.txt

mann_d_Page_099.txt

mann_d_Page_033.txt


Full Text





Definition of Terms


Anxiety (State). State anxiety is defined as the subjective feelings of tension, apprehension,
nervousness, and worry, coupled with the activation of the autonomic nervous system
(Spielberger, 1983). Furthermore, anxiety is believed to be comprised of cognitive (i.e., thoughts
and worries) and somatic (i.e., perception of physiological arousal such as increased heart rate)
components (Liebert & Morris, 1967).

Bereitschaftspotential (BP). The BP, first described by Kornhuber and Deecke (1964), is a
negative cortical potential that develops approximately 1-1.5 seconds prior to the onset of a self-
paced task, suggesting that the BP has both motor preparation and attention components.

Closed Task. This is a skill that is performed in a stable or predictable environment in which the
participant is in control of movement onset (Magill, 1998).

Cortical Generators. Cortical generators are known as the distinct regions of the cortex (e.g.,
Supplemental Motor Cortex, Motor Cortex) known to have topographical representation of the
Bereitschaftspotential (e.g., Frontal, Parietal, Temporal).

Electroencephalography (EEG). The EEG is a process for recording the electrical potential at the
scalp associated with the cortical activity of the underlying structures.

Event Related Potential (ERP). The ERP is a time-locked recording of cortical activation in
accord with the International 10-20 system (Jasper, 1958). More specifically, the ERP is the
result of an averaging of samples recorded from a continuous EEG that is time-locked to a
specific event. As such, the averaging technique draws on the signal-to-noise ratio, such that the
components of the waveform not deemed related to a specific event are assumed to randomly
vary across samples, thereby rendering only a representation of systematic variation or
components of the ERP (i.e., slope, amplitude; Fabiani et al., 2000).

Eye-Movement Registration. Eye-movement registration refers to the process of recording the
visual search characteristics during a perceptual-cognitive, or perceptual-motor task. The most
common procedure is the corneal-reflection method, which records the movement of the eye
under the assumption that the central portion of the cornea corresponds with a point of visual
interest (Duchowski, 2002; Williams, Davids, & Williams, 1999).

Fixations. Fixations are characterized by the tiny eye movements associated with tremors, drifts,
and microsaccades (Duchowski, 2002) that are necessary to stabilize the fovea on a specific
target, enabling comprehensive stimulus extraction and information-processing (Williams et al.,
1999).

Gaze Behaviors. A gaze is the "absolute position of the eyes in space and depends on both the
eye position in orbit and the head position in space" (Schmid & Zambarbieri, 1991, p.229).
Therefore, a gaze behavior refers to a specific coordinated action of the eyes and head, which
include a saccade, fixation, smooth-pursuit, vestibular ocular reflex, and the QE period.









with previous research, the experienced players demonstrated superior anticipatory skills.

Furthermore, Williams and Davids (1998) examined differences in anticipation and visual search

strategies in 3-on-3 and 1-on-l soccer situations. Twelve experienced and 12 inexperienced

soccer players were presented with 20 offensive soccer sequences and asked to anticipate final

pass location. The results demonstrated that the experienced players were more adept at

anticipating final pass destination and did so more quickly as compared to the less skilled

participants. Additionally, when aspects of the visual environment were occluded the

performance of the advanced group was significantly hindered while the less skilled participants

performance remained unaffected, further suggesting that advanced visual cues are requisite

components of experienced performers swift and accurate decision-making.

Recent studies in baseball perceptual decision-making have supported expert novice

differences. Radlo, Janelle, Barba, and Frehlich (2001) compared groups of baseball players (i.e.,

varsity players and college students) on a baseball pitch discrimination task that required the

participants to identify the type of pitch seen (fastball or curveball) as quickly as possible by

pressing one of two buttons. The findings demonstrated that elite performers were quicker and

more accurate at identifying the type of pitch, supporting the notion that advanced cues, in

addition to extensive knowledge structures facilitate accurate and expeditious decision-making.

Similarly, Abernethy and Russell (1987a, 1987b) demonstrated enhanced anticipatory

behaviors of expert performers in badminton. Twenty expert and 35 novice badminton players

were required to predict the landing position of a badminton shuttle-cock in response to varying

levels of temporal occlusion (see later section on Occlusion Paradigms). Systematic expert-

novice differences were apparent with the expert performers demonstrating a prolific ability to

use cues presented earlier in the action sequence to predict stroke outcome. The novice









have been impossible to draw any conclusions regarding the role that the QE period may have in

emotion regulation without a successful anxiety manipulation. Therefore, the changes in anxiety

and arousal permit a direct comparison of the effects of anxiety on the cortical and visual

mechanisms deemed characteristic of expert performers.

Skill Based Putting Performance and Quiet-Eye Differences Across Anxiety Conditions

Since the seminal work of Vickers (1996a), a growing body of evidence has been

compiled supporting the notion that fixations of relatively longer duration are evident during the

preparatory stages of a self-paced task, and can differentiate the expert from the near-expert

performer (Mann et al., 2006). Given that many of the hypotheses outlined in this investigation

are dependent on overt and measurable skill based differences, several performance variables

were of interest. Putting performance was measured as a function of "hits" and "misses," and

missed putts were additionally assessed for radial error (RE) and bivariate variable error (BVE).

I hypothesized that the LH group would make more putts than the HH group, with reduced

variability on missed putts and in turn, a longer QE period than the HH group across anxiety

conditions.

Results confirmed the hypothesized effects of .\ki// but not Anxiety. That is, the LH group

successfully holed more putts under both the low and high anxiety conditions while exhibiting a

longer QE duration than the HH group. As expected, the LH group was also more consistent

(i.e., less error) with "nearer misses" than the HH group. Anxiety failed to significantly alter

putting accuracy when considered absolutely (i.e., hit, miss) and qualitatively (i.e., RE, BVE).

However, for both the HH and LH groups, absolute and qualitative performance was superior in

the high anxiety condition, although not statistically significant. In retrospect, the anxiety

manipulation may have actually facilitated motivation and concentration in both groups. In other

words, the high anxiety condition may not have been severe enough to promote performance









CHAPTER 2
REVIEW OF LITERATURE

Capturing Expertise

The performance of experts across domains has intrigued scientists for centuries. More

recently, scientific enquiry has attempted to put forth a theoretical framework from which the

study of expertise can be undertaken in an attempt to deduce the mechanisms that set apart the

expert from the novice. Since the principle work of Chase and Simon (1973), researchers have

demonstrated that expertise is characterized by an extensive knowledge base that facilitates both

stimulus recognition and subsequent procedural execution (Richman, Gobet, Stazewski, &

Simon, 1996). To effectively capture the essence of the expert, the many traits comprising such

individuals must be understood. As such, the systematic observation and manipulation of both

the overt and covert psychological processes of expert and novice performers can provide

detailed insight into those mechanisms separating the expert from the novice. To begin, a clear

operational definition of what constitutes an expert is warranted.

Definition of Expertise

Sport expertise has been defined as the ability to consistently demonstrate superior athletic

performance (Janelle & Hillman, 2003; Starkes, 1993). Although performance competency and

mastery are requisite components of expertise, unquestionably other equally prominent

components exist. Siedentop and Eldar (1989) concur that a primary requisite of expertise is the

ability to demonstrate technical proficiency that is reliable and consistent. However, they also

emphasize the importance of the athlete's aptitude to modify performance skills to meet varying

contextual conditions. More precisely, experts possess an extensive procedural and declarative

knowledge base (McPherson, 1993,1994, 1999, 2000; French, Spurgeon, & Nevett, 1995; French

& Thomas, 1987) that enables them to extrapolate important information from relevant









An extensive body of evidence suggests that the visual system is the dominant perceptual

system by which all other perceptual systems are attuned (Abernethy, 1996; Janelle, Hillman, &

Hatfield, 2000; Posner, Nissen, & Klein, 1976; Van Wynsberghe, Noback, & Carola, 1995). The

inability to coordinate the visual and motor systems while regulating affective states may

confound the mechanical elements of the golf putt, rendering it a difficult and often frustrating

task. As such, the ability to attain, master, and demonstrate performance proficiency of

motorically simple tasks under varying contextual conditions (e.g., high pressure) can prove

demanding even for the most skilled athletes (Singer, 2000), suggesting that optimizing

attentional processes during the preparatory period immediately preceding task execution for a

self-paced task is of paramount importance (Hillman, Apparies, Janelle, & Hatfield, 2000).

Given that the visual system is the dominant perceptual system, researchers have dedicated

considerable effort to addressing the visual search characteristics and gaze behaviors accounting

for the attentional factors that preclude the expert advantage. Mann, Williams, Ward, and Janelle

(2006) conducted a meta-analysis encompassing nearly three decades of work, examining the

many performance metrics and indices of attentional allocation differences of experts and non-

experts. The results provide further support for the role of visual attention in the expert

advantage, revealing that experts consistently exhibit fewer fixations ( rpb = 0.26) of longer

duration ( rpb = 0.23). Such visual search characteristics index an individual's point of interest

and relative attention allocation. The longer the eye remains fixated on a given target, the more

information is thought to be extracted from the display, permitting detailed information

processing. Additionally, the number of visual fixations during a given period provides an index

of the search characteristics representative of the most pertinent cues extracted from the

environment, thereby facilitating the decision making process. Given the typically dynamic









Perceptual-Cognitive ,N/ill The ability to identify and acquire environmental information for
integration with existing knowledge such that appropriate responses may be selected and
executed (Marteniuk, 1976).

Quiet-Eye Period. The QE period, a component of the gaze behavior, is a measure defined as the
elapsed time between the last visual fixation to a target and the initiation of the motor response
(Vickers, 1996a).

Self-Paced Task. This is a skill that is performed free from external temporal constraints or
prompts, while under the volition of the participant.

SpectralAnalysis. Spectral analysis is the process of separating an EEG time series into its
constituent frequencies (e.g., Alpha 8-12Hz, Beta 13-36Hz) by subjecting the raw EEG data to
Fast Fourier Transform (FFT).

Spectral Power. The decomposition of the cortical levels expressed by means of a spectral
analysis is inversely related to activation (Pfurstcheller, 1992). That is, spectral power refers to
the decrease in activation otherwise known as the cortical quieting of the structures related to a
perceptual-motor task. For example, an increase in left hemisphere Alpha power (8-13Hz),
suggests a decrease in the amount of activation that is experienced.

Sub-Cortical Generators. Sub-cortical generators represent the deep structures of the brain (e.g.,
Basal Ganglia, Thalamus, Cerebellum) believed to significantly contribute to the development of
the motor program and in turn the cortical structures known to have a topographical
representation of the Bereitschaftspotential (e.g., Frontal, Parietal, Temporal).

Stimpmeter. The stimpmeter, designed by Edward S. Stimpson and revised and implemented by
the USGA in 1978 (USGA, 2006), is a device used to quantify the speed of a putting surface.
The stimpmeter is a V-shaped aluminum bar measuring 36in long and 1/2in wide that is designed
to channel the golf ball and reduce unnecessary lateral movement as the ball is released.
Positioned 6 inches from the top of the stimpmeter is a precisely milled ball-release notch
designed to discharge a ball when the stimpmeter is raised approximately 20 degrees from the
ground. The bottom of the stimpmeter is leveled to facilitate the desired angle of release. Three
balls should be released independently from the same starting point and measured for the
distance each ball travels from the end of the stimpmeter. The average distance traveled is
referenced as the speed of the green. For example, if the average distance traveled of three golf
balls released from the same position is 9.5ft then the speed of the green is a 9.5.

Visual Search. A two-stage process conducted to identify relevant cues in which a memory
dependent systematic search (i.e., serial search) or a memory-less random search of the
environment is coupled with a decision process confirming relevant stimulus identification.

Volition. Volition refers to the conscious will to act; a voluntary act that is endogenously driven
and free of externally imposed restrictions (Libet, 2003).

Assumptions

The current investigation was conducted under the following assumptions:









visual search behavior in karate kumite. The viewing patterns of both groups' (Expert and

Novice) were altered under the high anxiety condition as compared to the low anxiety condition,

with a marked increase in attention to peripheral cues. Furthermore, search rate increased along

with greater anxiety among novices. More specifically, experts in the high anxiety condition

demonstrated an increase in the mean duration of their fixations while the mean fixation duration

of the novice group decreased. The corresponding increase in search activity in the novice group

can be explained by a comparative decrease in processing efficiency and ineffective cue

utilization (Williams & Elliot, 1999).

In a related study, Janelle, Singer, and Williams (1999) examined changes in gaze

behavior under varying levels of anxiety during a simulated racing task. Results indicated that as

anxiety increased, processing efficiency and task performance decreased, while the

corresponding gaze behaviors became more variable (i.e., more fixations of shorter duration).

Janelle et al. (1999) concluded that anxiety increases attentional narrowing, resulting in the

ineffective search and use of cues. Furthermore, Williams et al. (2002) assessed table-tennis

performance under combinations of high and low working memory with corresponding changes

in anxiety (high and low). As expected, results indicated increased effort, delayed reaction times,

and increased search rates while performing under the high working memory and high anxiety

condition as compared to the low working memory and low anxiety condition, a pattern

indicative of less efficiency. As mentioned, Murray and Janelle (2003) used a dual-task paradigm

to assess the effects of increased anxiety on a simulated driving task. Consistent with previous

research (Janelle et al., 1999; Williams & Elliot, 1999; Williams et al., 2002), driving

performance decreased as anxiety increased concomitantly with search rate, indicative of a

decrease in processing efficiency.









Lyytinen, 1993, 2000). Conversely, the execution of the golf putt requires the facilitation of a

pre-planned motor act, a task that is more conducive to the use of the BP as means for

investigating the motor program. As such, it is plausible that the level of physical involvement

(i.e., prolonged and variable muscular activation prior to trigger pull) in the preparatory period

leading up to the point of trigger pull in marksmen, may overshadow the slow potential nature

recordings of the BP.

Cortical Activity and the Preparatory Period Summary and Review

The systematic observation of electro-cortical activity permits a real-time look into the

underlying cortical structures contributing to the psychological processes accounting for

psychomotor skill differences. Self-paced tasks demanding visuomotor coordination such as golf

putting, archery, and marksmanship place great demand on the psychomotor systems (i.e.,

attention, emotion, motor control and preparation) conducive to psychophysiological recording

(Hatfield & Hillman, 2001).

The early work of Hatfield and colleagues (1982) sparked a series of investigations

(Hatfield, Landers, & Ray, 1984, 1987) which examined the covert cognitive processes

associated with skilled psychomotor performance of elite marksmen and other self-paced

activities. To date, the majority of research in sport has relied on spectral analysis techniques to

address issues of hemispheric asymmetry consistently reporting skill automaticity and cerebral

efficiency in favor of the expert, concluding that the reported decrease in left hemisphere

activation with the concomitant increase right hemisphere activation may represent a reduction in

verbal-analytic processing coupled with vigilant visual-spatial processing as the time to

performance execution nears; arguably an optimal preparatory set for self-paced precision sports

(Hatfield et al.,1984).









Eye Movement Research

The seminal years: 1976-1989

One of the first attempts to determine the visual search behaviors of basketball players and

the first in sport to empirically validate the purported expert-novice visual search differences

when solving strategic problems was conducted by Bard and Fleury (1976) using a NAC eye

movement recorder. Expert (n=5,) and novice (n=5) basketball players were presented with 84

slides depicting one of 28 different typical offensive basketball schematics. At the onset of the

stimulus, each participant was required to identify the type of solution required by the presented

offensive strategy, selecting one of seven possible answer choices (i.e., shoot, dribble, four

passing options, and stay). Two dependent variables were assessed, decision-time and number of

visual fixations. Experts demonstrated fewer visual fixations per trial (3.3 vs. 4.9) than novice

players. No differences between groups were noted for decision time, suggesting that the visual

search patterns of experts were more efficient (i.e., fewer fixations of longer duration) than

novice players. Moreover, distinct cognitive strategies were apparent, with experts concentrating

fixations around the pairings of offensive-to-defensive players, while the novice players

maintained more frequent fixations to teammates, neglecting the opposition. However, the

efficacy of this difference is unknown, since no attempt was made to evaluate performance

outcome.

As a follow-up to the previous basketball study (i.e., Bard & Fleury, 1976), a series of

experiments with basketball and ice-hockey players (Bard & Fleury, 1987) was conducted using

a NAC eye movement recorder to assess the number of fixations and fixation durations across

tasks of varying complexity. As in the earlier study, basketball players were presented a series of

schematics; this time contextual complexity was varied. The results demonstrated that experts









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S ......................................... ............................... 8

LIST OF FIGURES ................................. .. ..... ..... ................. .9

A B S T R A C T ............ ................... ............................................................ 1 1

CHAPTER

1 INTRODUCTION ............... ................. ........... .............................. 13

Expertise Related Differences in Cortical Activity ..................................... .................16
Expertise, Visual Search and Emotion Regulation...................................... ............... 21
L im itatio n s ................... .......................................................... ................ 2 4
State ent of the Problem ..................................................................... 26
T h e C u rren t S tu d y ............................................................................................................. 2 6
H ypotheses.......... ..........................................................27
Definition of Terms ........................................... ........................30
A ssu m p tio n s ............................................................................. .. 3 1
Significance of the Study .................. ................................ .. ........ .. .......... 32

2 R EV IEW O F LITER A TU R E .................................................................... ...... .................34

Capturing Expertise ................................. .. ... .... ................ ... 34
D definition of Expertise ..................................... .................. ............ ...... ....... 34
Expert Performance .......................................... ..... ....... ......... .............. 36
Experts Are Faster and More Accurate at Recognizing Patterns ..................................36
Experts Have Superior Procedural, Declarative, and Strategic Knowledge ...................38
Experts Are Superior at Anticipating Opponent's Actions...........................................40
Experts Maintain Superior Perception of Relevant Kinematic Information .................43
Experts Maintain More Efficient and Effective Visual Search Patterns .......................46
Experts Demonstrate Physiological Patterns Indicative of Sensorimotor Efficiency .....46
V isual P perception in Sport .............................................................................. ............ 48
O cclusion Paradigm s ....................................... .................... .. ...... .... 49
O cclusion Research ................................... .. .. ........ .. ............50
S u m m ary ............................................................................... 6 9
V isu a l S e a rc h .................................................................................................................... 7 1
Eye M ovem ent Registration ........................................................................ 72
E ye M ovem ent R research ........................................................................... .............73
T he sem inal years: 1976-1989 ................................................. ............... .... 73
Empirical and methodological advancements: 1990-1998 ....................................78
Contem porary investigations: 1999-2002 ..................................... .................92
Visual search sum m ary and review .................................. ........................ 100


































2007 Derek T.Y. Mann









Lidor (1996) used a temporal occlusion paradigm. Eight different tennis strokes were video

taped (1. Cross-court slice, 2. Forehand down the line, 3. Backhand down the line, 4. Backhand

(winner) down the line, 5. Forehand volley (winner) near the net, 6. Serve, 7. Backhand drop

volley cross-court, 8. Forehand cross-court) and were presented in the laboratory using the

following six occlusion conditions:

1. 12 frames (-480 msec) prior to ball-racquet contact.
2. 8 frames (-320 msec) prior to ball-racquet contact.
3. 4 frames (-160 msec) prior to ball-racquet contact.
4. At the point of ball-racquet contact.
5. 4 frames (+160 msec) subsequent to ball-racquet contact.
6. 8 frames (+320 msec) subsequent to ball-racquet contact.

After viewing each occluded stroke participants responded as quickly as possible to the

landing position of the ball. A response sheet with a scaled replica of the tennis court was used to

record the predicted landing locations. Radial error for response accuracy was assessed across

occlusion conditions.

The results of this investigation mirrored those of previous research, with the expert and

intermediate groups outperforming the novice group in the advance occlusion conditions (i.e., -

480, -320, and -160 msec prior to ball-racquet contact). No differences were noted between the

expert and intermediate groups across occlusion conditions, suggesting that the benefits of

advance perceptual skill may reach an asymptotic point when a level of performance competency

has been attained. In conclusion, early temporal cues (i.e., cues up to the point of ball-racquet

contact) appear to be detected and relied upon by skilled players, while remaining unprocessed

by less skilled performers.

Interceptive sports such as the racquet sports discussed here (i.e., tennis, badminton, and

squash) are predominant across sport science research on cue utilization. One of the few

exceptions to this rule is the work of Paul and Glencross (1997) with expert (n=15) and novice









differences beginning as early as 167 ms prior to racquet-shuttle contact with the largest

differences evident 83 ms prior to contact, confirming Abernethy and Russell's (1987a) previous

findings with expert and novice players. The results of the spatial occlusion task also illustrate

intermediate-novice differences in the prediction of stroke direction. Most notably, the error rate

for occlusion conditions el (arm and racquet) and e2 (racquet only) differed significantly for

both groups as compared to all other occlusions conditions. However, inconsistent with previous

findings (Abernethy & Russell, 1987a), the intermediate group did not "statistically" perform

better than the novice group during condition el, yet it should be noted that performance trends

were similar to the expert-novice differences of Abernethy and Russell (1987a). These results

signify that perceptual advantages are evident with elevated levels of performance, even though

performance may not be elite (Isaac & Finch, 1983).

In another study of tennis players, Buckolz, Prapavesis, and Fairs (1988) attempted to

delineate the specific advance perceptual cues used by advanced (n=21) and intermediate (n=23)

tennis players during passing shots. A series of tennis strokes (i.e., down-the-line, cross-court,

and lob passing shots, for both forehand and backhand) were filmed at two different speeds (i.e.,

24 frames/second and 48 frames/second) allowing for greater experimental control over the

duration of cue exposure. A temporal occlusion paradigm was used to facilitate the identification

of the cues used by and those that discriminated between the advanced and intermediate players.

The authors unconventionally ordered the sequence of film clips, beginning with the most

occluded condition (168 ms at 24frames/second and 84 ms at 48frames/second prior to ball

contact) for a given stroke through to the least occluded condition (168 ms at 24frames/second

and 84 ms at 48frames/second post ball contact) for that stroke. Once each occlusion condition

had been delivered for a specific stroke, the next stroke was introduced. To further clarify, each









In an attempt to extend the previous research, Abernethy and Russell (1987a) conducted

two experiments to independently establish and compare the temporal and spatial characteristics

of the advance cues used by expert and novice sport performers. In Experiment 1 five temporal

occlusion conditions were used (i.e., 167 msec prior to racquet-shuttle contact; 83 msec prior to

racquet-shuttle contact; occlusion at the point of racquet-shuttle contact; 83 msec subsequent to

racquet-shuttle contact; and no occlusion). Participants were 20 expert and 35 novice badminton

players who were required to predict the probable landing position of a badminton shuttle.

Immediately following each trial, all responses were recorded on a scaled representation of the

receiver's court. The findings were consistent with previous results, in that differences in

prediction accuracy were notable between expert and novice performers across occlusion

conditions. That is, from 83 msec prior to racquet-shuttle contact through the no-occlusion

condition, expert and novice players differed in their performance accuracy. The authors also

concluded that the cues essential for successful directional perception are present during the

interval between 83 msec prior to racquet-shuttle contact and 83 msec post racquet-shuttle

contact, while depth perception cues are present in the final 83 msec prior to racquet-shuttle

contact. The collective findings suggest advance cues are apparently critical for depth perception

whereas a greater window of opportunity presides for directional detection (Abernethy &

Russell, 1987a).

In an effort to isolate expert's attention to the most salient of perceptual cues, Abernethy

and Russell (1987a) adopted a spatial occlusion paradigm in Experiment 2, allowing for the

constant temporal display of perceptual information while controlling the occlusion of explicit

cues. The same population of badminton player's participated in this study as in study one. In a

similar fashion, 32 different badminton strokes were included and subject to occlusion. The









utility of the QE's temporal window for organizing relevant cognitive and perceptual cues for

improving performance. Third, because both the BP and QE duration are believed to have motor

programming implications and were significantly related in the current study, further support for

the motor programming hypothesis is evident. Furthermore, the lack of relationship between the

QE period and anxiety suggests that the QE duration may not serve an emotion regulation

function, but that such regulation may take place in the complex pathways linking the cortical

and subcortical generators of the motor program. Although speculative, the BP has been shown

to reflect changes in arousal and motivation (Andreassi, 1980; McAdam & Seales, 1969), which

may circumvent the need for additional regulatory processes. As a result, the QE period may

simply regulate the type and amount of visual information needed to evoke the requisite motor

program, rather than modulate the effects of anxiety itself.

Inter- and Intra-Group Performance Variability on Quiet-Eye Duration

Research with golfers (Vickers, 1992), marksmen (Janelle et al, 2000), basketball players

(Vickers, 1996b), biathletes (Vickers et al., 1999), and volleyball players (Vickers, 1997, 1998)

indicates that experts not only demonstrate a longer QE when compared to near-expert

performers, but that the QE is associated with within group performance differences. I therefore

hypothesized that variability in the duration of the QE period would account for inter and intra-

group performance variability. The LH group was hypothesized to exhibit a longer QE period

relative to HH group, and the QE duration of both the LH and HH groups was predicted to

exceed the QE duration for missed putts compared to successful ones.

In accord with expectation, the LH group not only performed better on the putting task, but

also engaged a longer QE duration relative to the HH group. Contrary to previous research

however, QE duration did not account for within group performance differences across skill

levels. Caution however, must be taken when interpreting these statistical findings. Despite the









movements by approximately 1 to 1.5 second. The slow negative cortical waves, of which the BP

is one, are believed to be associated with arousal, the recruitment and augmenting of responses,

and the general facilitation of processes required for task completion (Deecke, 1973). That is, an

increase in negativity consistent with the BP is representative of cortical activation or excitation.

By deductive reasoning, Brunia (1993) reports that all functions, including perception, attention,

and preparation are realized by complex patterns of facilitation and inhibition in specific

neuronal circuits which must be depolarized in order to produce an action. Therefore, it becomes

possible to study, through electroencephalography, the cortical activity of the pre-motor period

given that the depolarization process characterized by an increase in negativity at the surface of

the cortex must precipitate an action (Brunia, 1993). Accordingly, readiness to act is depicted by

a cortical signature manifested in motor and attentional processes during the preparatory period

(Brunia, 1993; Deecke, 1973).

The BP is a visually distinct waveform (see Figure 2-6), denoted by three critical

components that are temporally and cortically distinct (Deecke, Scheid, & Kornhuber, 1969): the

BPearly, BPlate, and BPpeak. The slow rising negativity of the BPearly starts approximately 1500 ms

prior to movement onset and although demonstrated to have a wide spread scalp distribution with

maximal potentials recorded at the vertex, the early onset and pronounced amplitude implicates

the supplementary motor area (SMA). The relative increase in cortical activity associated with

the mid-frontal (Cz) region is not altogether surprising given its role in working memory,

inhibition, planning, and executive functioning, including the integration and regulation of

emotion and its corollaries. Specifically, the SMA is reportedly central to the planning and

initiation of voluntary movement (Parent, 1996). For example, the BP is absent in movements

exogenously driven (i.e., reflexive or passive movements) but is further exaggerated in those









(Konttinen & Lyytinen, 1993) and golfers (Crews & Landers, 1993) suggests that an increase in

BP negativity corresponds with an increased readiness to perform and performance excellence.

As such, psychophysiological research has validated a cerebral efficiency hypothesis and further

indicates that the expert maintains a well-developed sensorimotor program necessary for

performance.

The early work of cognitive psychologists and the relentless inquiry into human

information processing have yielded much of the information from which the above conclusions

have been inferred. As such, the following section will proceed with a cursory review of the

pivotal developments leading up to and significantly contributing to the current understanding of

information-processing and cortical changes as applied to sport. Furthermore, an attempt to

isolate the gaps in the current understanding with respect to the role of the QE period in motor

preparation and/or emotion regulation for successful sport performance will be included.

Visual Perception in Sport

Expert performance is mediated by a number of factors including cognitive and perceptual

motor skills, as well as task specific anatomical and physiological adaptations (Ericsson &

Lehmann, 1996). Moreover, Janelle and Hillman (2003) postulate that in order "to obtain expert

status, athletes must excel in no less than four domains: physiological, technical, cognitive

(tactical/strategic; perceptual/ decision-making), and emotional (regulation/coping;

psychological)" (p.21). It is commonly understood that expert performance is a product of the

delicate balance between innate talents and the amount of practice/training (Ericsson, Krampe, &

Tesch-Romer, 1993). Of particular importance is the latter, in which the athletes' declarative and

procedural knowledge base develops and becomes both extensive and accessible. The extent of

this knowledge can be inferred from indirect perception/action coupling and more specifically by

means of the visual search behaviors and advance cue utilization abilities of exceptional










execution of a self-paced task. As such, a Pearson Product Moment correlation was conducted to

explore the relationship between QE duration and BP. Although emphasis is placed on the BPpeak

- Quiet-Eye duration relationship, the association between each component of the BP and QE

was assessed. Results indicated a significant correlation between QE duration and C3peak (r =

.3096, p = .026, d= .65), C4peak (r = .2874, p = .036, d= .60), and Czpeak (r = .2901, p= .035, d=

.61), suggesting that as QE duration increased so too did BP negativity within the specified

regions. No other significant correlations were found (P3peak (r = .1696, p = .148, d= .34); P4peak

(r =.1574,p= .166, d= .32).

Table 4-1 Pearson Product Moment correlations demonstrating the regional specificity associated
with the relationship between anxiety and cortical activation.

C3peak r = 0.28 p = 0.04* d = 0.58
Left-Central C31ate r = 0.354 p = 0.013* d = 0.76
C3early r = 0.339 p = 0.016* d = 0.72
C4peak r = 0.306 p = 0.027* d = 0.64
Right-Central C41ate r = 0.466 p < 0.001* d = 1.05
C4early r = 0.343 p = 0.015* d = 0.73
CZpeak r = 0.211 p = 0.096 d = 0.43
Midline Cziate r = 0.288 p = 0.036* d = 0.6
CZearly r = 0.326 p = 0.02* d = 0.69
P3peak r = 0.392 p = 0.006* d = 0.85
Left-Parietal P31ate r = 0.11 p = 0.249 d = 0.22
P3early r = 0.005 p = 0.487 d = 0.01
P4peak r = 0.482 p < 0.001* d = 1.1
Right Parietal P41ate r = 0.112 p = 0.246 d = 0.23
_P4early r = -0.073 p = 0.328 d = -0.15
*Denotes significant correlation p<.05.



Quiet-Eye Duration and Anxiety

It was hypothesized that both the LH and HH group would exhibit a longer QE duration

across both the low and high anxiety conditions. That is, an increase in the QE duration is

believed to circumvent the deleterious effects of anxiety while maintaining performance. Results

indicated that no significant differences were noted for Anxiety, (F(, 18) = .002, p =.963, d= .02).

Figure 4-3 shows the null differences in which QE duration was stable for each skill group across

anxiety conditions.









Upon providing consent, participants were prepared for electro-ocular (EOG),

electromyographic (EMG), heart rate (ECG) and electroencephalographic (EEG) measurement in

accord with the guidelines put forth by the Society for Psychophysiological Research (Pivik,

Broughton, Coppola, Davidson, Fox, & Nuwer, 1993). Vertical (VEOG) and Horizontal (HEOG)

bipolar electro-oculograhpic movements were collected to assess QE duration and to control for

ocular artifact in the EEG waveform. Four 4mm Biopac silver/silver chloride (Ag/AgC1)

electrodes (EL204) were positioned above and below the right eye, and lateral to each eye,

adjacent to the left and right orbital fossi.

EMG activity was recorded using two 10mm silver/silver chloride (Ag/AgC1) electrodes

placed 3 cm apart over the muscle belly of the extensor carpi ulnaris (ECU) of the right arm.

EMG activity was sampled at 1000Hz and amplified (x 5000) using the Biopac EMG amplifier

(EMG 100B).

ECG activity was recorded using two pre-gelled snap electrodes placed over the radial

artery of the left and right forearm. ECG activity was sampled at 1000Hz and amplified (x 5000)

using the Biopac ECG amplifier (ECG 100C).

The continuous EEG was recorded with an array of 6 silver/silver chloride (Ag/AgC1)

electrodes in accord with the International 10-20 system (Jasper, 1958) using a lycra electrode

cap manufactured by Electrode-Cap International, Inc. (ECI, Eaton, OH). A central cluster of

electrodes was positioned over the left, mid-line, and right central (primary motor cortex: C3, Cz,

C4) sites to concentrate on those cortical regions known to have implications in motor planning

and execution, as well as source generators of the BP (Orgogozo et al. 1979; Roland et al. 1980;

Ikeda et al. 1993; Rektor et al. 1994; Huckabee, et al. 2003). Furthermore, an additional cluster

of electrodes was positioned over the parietal cortex (P3, P4), a region associated with visuo-









1. The Mental Readiness Form Likert (Krane, 1994) is an appropriate measure for assessing
levels of cognitive and somatic anxiety prior to and during the course of the experimental
session.

2. The use of a platform covered with a nylon NP 50 artificial putting surface (Synthetic Turf
International, STI, Jupiter, FL) outfitted with a 4.25in regulation size golf hole adequately
simulates an actual putting surface, thereby promoting ecological validity.

3. The putting task (i.e., 12 ft flat putt) and the corresponding dependent measure of putting
accuracy (i.e., made vs. missed putts, bias, and consistency) are appropriate measures of
performance.

4. Vertical and horizontal electro-oculogram measures used to derive QE duration are both
appropriate and accurate.

5. The extensor carpi ulnaris (ECU) of the right arm is an appropriate muscle to demarcate the
fiducial time point necessary for the off-line reduction and analysis of the
Bereitschaftspotential and the corresponding cessation of the QE period.

Significance of the Study

The importance of understanding the complex integration of systems associated with

expert performance is essential for the advancement of both theory and practice. To date,

technological and methodological advances have permitted a rapid increase in research related to

expert performance, which has lead to a greater conceptual understanding of the expert

performer while executing self-paced, closed tasks. The work of Hatfield et al. (1984)

precipitated an increase in theory driven research examining the information processing, cortical

adaptations, and central nervous system differences of experts performing sensorimotor tasks.

However, it was not until recently that researchers have begun to examine the various

mechanisms of perceptual-cognitive expertise in an integrative manner (e.g., Hillman et al.,

2000; Janelle, Hillman, & Hatfield., 2000). Of primary importance are theoretical and

mechanistic developments stemming for the use of electroencephalographic and visual search

technology within an expert-novice paradigm. For example, the use of continuous EEG and

spectral analysis have reliably demonstrated the role of the alpha band, and to a lesser extent the









Tremayne, P., & Barry, R.J. (2001). Elite pistol shooters: Physiological patterning of best vs.
worst shots. International Journal ofPsychophysiology, 41, 19-29.

Van Wynsberghe D.M., Noback, C., & Carola, R. (1995). Human anatomy and physiology (3rd
ed.). New York: McGraw-Hill, Inc.

Vickers, J.N. (1988). Knowledge structures of expert-novice gymnasts. Human Movement
Science, 7, 47-72.

Vickers, J.N. (1992). Gaze control in putting. Perception, 21, 117-132.

Vickers, J.N. (1996a). Visual control while aiming at a far target. Journal ofExperimental
Psychology: Human Perception and Performance, 22, 342-354.

Vickers, J.N. (1996b). Control of visual attention during the basketball free throw. The American
Journal of Sports Medicine 24(6), S93-S96.

Vickers, J.N. (2004). The quiet-eye. It's the difference between a good putter and a poor one.
Here's proof. GolfDigest, 1, 96-101

Vickers, J.N., & Adolphe, R.M. (1997). Gaze behavior during a ball tracking and aiming skill.
International Journal of Sports Vision, 4(1), 18-27.

Vickers, J.N., Williams, A.M., Rodrigues, S., Hillis, F., & Coyne, G. (1999). Eye movements of
elite biathlon shooters during rested and fatigues states. Journal of Sport and Exercise
Psychology, 21, s116.

Ward, P., & Williams, A. M. (2003). Perceptual and cognitive skill development in soccer: The
multidimensional nature of expert performance. Journal of Sport & Exercise Psychology,
25(1), 93-111.

Ward, P., Williams, A.M., & Bennett, S.J., (2002). Visual search and biological motion
perception in tennis. Research Quarterly for Exercise and Sport, 73(1), 107-112.

Warren, C., & Karrer, R. (1984). Movement-related potentials in children: A replication
of waveforms, and their relationships to age, performance and cognitive development.
Annals of the New York Academy of Sciences, 425, 489-495.

Williams, A.M. (2000). Perceptual skill in soccer: Implications for talent identification and
development. Journal of Sports Sciences, 18(9), 737-750.

Williams, A.M., & Davids, K. (1995). Declarative knowledge in sport: A by-product of
experience or a characteristic of expertise? Journal of Sport and Exercise Psychology, 17,
259-275.

Williams, A.M., & Davids, K. (1998). Visual search strategy, selective attentive, and expertise in
soccer. Research Quarterly for Exercise and Sport, 69, 111-128.









setter's body, followed by hands and ball flight. Conversely, the novice group identified the

overwhelming importance of ball flight, followed by the 'spikers' move, and setter's body as the

most pertinent cues. These variations in perceived locus of attention are exclusive to this

investigation, as the study by Abernethy (1990b) found no differences in self-reported direction

of attention between experts and novice squash players. Based on the unique characteristics of

each sport one can only speculate as to the root of these differences. Regardless, the work of

Wright et al. (1990) extended beyond racquet sport research to volleyball signifying the

importance and effective use of advance visual cues by experienced athletes across multiple

sports.

Houlston and Lowes (1993) further addressed the depth and direction prediction while

assessing the cue-utilization processes of expert (n=6) and non-expert (n=6) wicketkeepers. A

film occlusion technique with four distinct temporal conditions including; (tl), ball release; (t2),

156 msec post release; (t3), 234, sec post release; (t4), 390 msec post release) was used.

Wicketkeepers were required to view a series of distinct deliveries and anticipate the landing

position of the ball. A target mat was positioned in front of the batting wicket that was used to

assist in the scoring of pitches. All predictions were recorded on a scaled version of the target

mat and were evaluated for radial error as well as depth and lateral error. The results of this

investigation were atypical to those previously noted. That is no differences between group

prediction accuracy were noted across occlusion conditions. However, significant prediction

differences were evident for lateral versus depth error; both the expert and novice groups

predicted lateral error more accurately across occlusion conditions as compared to depth

predictions with a significant improvement in depth perception occurring under (t4). The results

suggest that lateral estimations can reliably occur early in the perceptual process, while depth









the position of the joystick intercepted the coordinates of the flight path of the soccer ball at the

time it crossed the goal-line a 'save' was declared. Therefore, accuracy was determined by the

(a) number of total saves; (b) correct side (i.e., correct lateral prediction, but incorrect vertical

prediction); (c) correct height (i.e. incorrect lateral prediction, but correct vertical prediction);

and (d) proportion of corrections (i.e., percentage of corrective movements). The visual search

data were assessed for search rate, and percentage of viewing time distributed to the following

fixation locations: head, trunk, shoulders, arms, hips, kicking leg, non-kicking leg, ball, and

unclassified.

The anticipation tests revealed no significant skill-based differences for save percentage

(Savelsbergh et al., 2002). However, it is possible that the secondary measures of accuracy may

be more indicative of skill-based differences. Specifically, the expert goalkeepers predicted shot

height and direction more accurately, committed fewer corrective movements, and began their

responses closer to ball-foot contact than the novice goalkeepers. The nature of the contrived

task may therefore, have contributed to the lack of significant accuracy differences. Simply, the

novice performers committed more corrective movements throughout the sequence, suggesting

that more information was required for these athletes to confirm their decision, a luxury

unavailable in a real-world condition.

Secondly, experts were better able to anticipate the height and direction of shots more

accurately and initiate their responses just prior (300 ms) to ball-foot contact, suggestive of the

effective use of advance cues (Savelsbergh et al., 2002). In contrast, the novice group initiated

their response 500 ms prior to ball-foot contact. Although such a brief differential existed

between the expert and novice players, the early response time by the novice players may be

more suited to a guess, whereas the expert players are able to match the advance cues with an

































B D D
C


Expectancy Developed
E F

Anticipation Selective Preparation

Incorrect Correct Incorrect Correct
I I I I
G H I J
Increase Eliminate RT, Increased Decreased
Movement Reduce MT Reaction Reaction
Time (MT) Time (RT) Time



Figure 2-1 An information-processing account of the advantages of advance cue usage. (Adapted
from E. Buckolz, H. Prapavesis, and J. Fairs (1988). Advance cues and their use in
predicting tennis passing shots. Canadian Journal of Sport Science, 13(1), 20-30).









1993; Lawton, Hung, Saarela, & Hatfield, 1998). Consistent with the notion that the QE period is

associated with the coordination and production of the requisite resources to execute a task

successfully while filtering out irrelevant environmental cues, it was postulated that expert

shooters would engage in prolonged QE periods when compared to novice shooters. Confirming

this hypothesis, Janelle and colleagues (2002) reported performance and QE differences between

groups, with the expert group outperforming the novices and exhibiting relatively longer QE

periods.

Quiet Eye Summary and Review

Advances in research technology, specifically advances in visual search equipment, have

provided a unique opportunity for expertise researchers to probe previously unattainable

questions from a mechanistic perspective. The resulting gaze behavior research in sport has

provided valuable insight into the mechanisms that may differentiate expert from novice

performers. In this case, the QE period, has indexed a mechanism that appears to be temporally

linked with the organization of visual information and necessary for the coordination of the

motor pathways required for successful execution of a desired task. To this point, research has

reliably demonstrated relatively prolonged QE periods as an effective marker for differentiating

skilled and lesser skilled athletes. Moreover, these findings have shown consistency across

domains as diverse as a rifle shooting and billiards and across tasks that require aiming at a target

(e.g., billiards and shooting) to those that require the individual to receive a projectile

momentarily while aiming and releasing it to a designated target (e.g., volleyball).

Expertise, Visual Search and Emotion Regulation

The interaction between the visual and emotional systems has been robustly demonstrated

across a variety of tasks (e.g., martial arts, driving, billiards, biathlon shooting), suggesting that

as anxiety increases, a corresponding reduction in visual attention efficiency and psychomotor









LIST OF REFERENCES


Aalto, H., Pyyko, I., Ilmarinen, R., Kahkonen, E., & Starck, J. (1990). Postural stability in
shooters. Journal of Oto-Rhino-Larynology and its Related Specialties, 52, 232-238.

Abernethy, B. (1988). The effects of age and expertise upon perceptual skill development in a
racquet sport. Research Quarterly for Exercise and Sport, 59, 210-221.

Abernethy, B. (1989). Expert-novice differences in perception: How expert does the expert have
to be? Canadian Journal of Sports Sciences, 14(1), 27-30.

Abernethy, B. (1990a). Anticipation in squash: Differences in advance cue utilization between
expert and novice players. Journal of Sports Sciences, 8, 17-34.

Abernethy, B. (1990b). Expertise, visual search, and information pick-up in squash. Perception,
19, 63-77.

Abernethy, B. (1991). Visual search strategies and decision-making in sport. International
Journal of Sport Psychology, 22, 189-210.

Abernethy, B. (1996). Training the visual-perceptual skills of athletes: Insights from the study of
motor expertise. The American Journal of Sports Medicine, 24(6), S-89-S-92.

Abernethy, B., Burgess-Limerick, R., & Parks, S. (1994). Contrasting approaches to the study
of motor expertise. Quest, 46, 186-198.

Abernethy, B., Neal, R.J., & Koning, P. (1994). Visual-perceptual and cognitive differences
between expert, intermediate, and novice snooker players. Applied Cognitive Psychology,
18, 185-211.

Abernethy, B., & Russell, D.G. (1987a). Expert-novice differences in an applied selective
attention task. Journal of Sport Psychology, 9, 326-345.

Abernethy, B., & Russell, D.G. (1987b). The relationship between expertise and visual search
strategy in a racquet sport. Human Movement Science, 6, 283-319.

Abernethy, B., Thomas, K.T., & Thomas, J.T. (1993). Strategies for improving understanding of
motor expertise [or mistakes we have made and things we have learned!!]. In J.L. Starkes
& F. Allard (Eds.), Cognitive issues in motor expertise (pp.317-356). Amsterdam: North-
Holland.

Adolphe R., Vickers, J. N., & LaPlante, G. (1998). The effects of training visual attention on
gaze behaviour and accuracy: A pilot study. International Journal of Sports Vision, 4(1),
28-33.

Allard, F., & Burnett, N. (1985). Skill in sport. Canadian Journal ofPsychology, 39(2), 294-312.









LIST OF TABLES


Table page

4-1 Pearson Product Moment correlations demonstrating the regional specificity
associated with the relationship between anxiety and cortical activation......................160









preparation and execution of a self-paced task. As such, a significant and positive correlation
between BP and QE period was predicted. More specifically, I expected that as the duration
of the QE period increases, relative increases in the amplitude of the BPpeak would be evident.
Furthermore, as anxiety levels increase, a corresponding increase in both QE duration and
BPcomponents were expected across skill levels.

8. Increases in anxiety result in a corresponding increase in cognitive demand, as evidenced by
attentional narrowing (Eysenck & Calvo, 1992) and the inefficient use of perceptual cues
(Easterbrook, 1959). Therefore, I hypothesized that both the LH and HH groups would
exhibit a longer QE period in the high anxiety condition as compared to the low anxiety
condition. That is, an increase in the QE was hypothesized to circumvent the deleterious
effects of anxiety, preventing a decrement in performance (Vickers et al., 1999). As such, a
corresponding increase in QE duration would permit the allocation of attentional resources to
the information processing of the visual cues previously attended to, while suppressing the
processing of competing stimuli, thereby facilitating performance.

Although each of the eight aforementioned hypotheses were designed to address the covert

psychophysiological differences between expert and near-expert performers, the ability to isolate

the role of the QE duration during a self-paced task is principal to all others. Given the extent to

which QE differences have been reported in the extant literature, expertise differences were

expected. However, the degree to which the skill-based QE differences and BP changes are

related offers both pragmatic and theoretical support for the role of pre-performance vision

previously unaccounted for. Previous research supports the contention that changes in quiet-

duration are directly related to inter- and intra-group variability, including under high stress

conditions (Williams, et al., 2002), lending support to the arousal regulation hypothesis.

However, the degree to which the QE period is believed to serve a motor programming function

is speculative. Given that the BP is responsible for motor planning and pairing of environmental

cues to response specifications (Brunia & van Boxtel, 2000), confirmation of a relationship

between the QE period and the various BP components would lend empirical support to the

motor programming function of the QE period.









Consequently, Deeny et al. (2003) employed coherence procedures to assess whether alpha

(low 8-10Hz and High 10-13hz) and beta (13-22Hz) band coherence of left anterior-temporal

region (T3) and the motor planning regions (pre-motor cortex; Fz) of the cortex are inversely

related to skill level. Ten expert and 9 skilled marksmen performed 40 shots each using the

Noptel Shooter Training System (ST-2000) during an 80 minute testing session. Cortico-cortical

communication was assessed during four one-second epochs preceding trigger pull. Results

indicated that compared to the skilled group, the expert marksmen demonstrated lower coherence

between the anterior-temporal and motor regions of the cortex for low alpha and beta

frequencies. Subsequent analyses further indicated lower coherence between motor regions and

all left hemisphere reference sites for high-alpha activity and lower coherence for beta activity

between the anterior-temporal site and midline locations (i.e., Cz, Pz) in the expert group.

The findings of Deeny et al. (2003) support the notion of decreased cortical

communication between functionally diverse brain regions in the elite level performer. As

previously mentioned, much of the psychophysiological research in sport has ascribed to the

basic tenets of Fitts and Posner's (1967) notion of automaticity, such that elite level performance

is governed by the automatic processing of the planning and execution of movement. Moreover,

as skill develops, the cortical regions necessary for skill execution become more specialized,

therefore relying more on the cortical mechanisms responsible motor programming and

execution (i.e, supplementary motor cortex, primary motor cortex, and the corresponding sub-

cortical generators) as confirmed by the psychological differences of expert and skilled

marksmen.

Bereitschaftspotential

The Bereitschaftspotential (BP) (or Readiness Potential) first described by Korhuber and

Deecke (1964) is a negative moving cortical slow potential that precedes the onset of self-paced









location, or type. The nature of the scan-paths and the extent of the expert-novice differences

cannot be generalized to closed-tasks or those that require precise aiming at a far target. Vickers

(1992), interested in the systematic changes in visual search as aiming accuracy improved,

assessed low handicap (i.e., 0-8) and higher handicap (i.e., 10-16) tournament caliber golfers. To

determine eye movements and gaze behaviors, participants were fitted with a mobile ASL 3001H

Eye View recorder, while executing consecutive putts from a distance of 3m. Participants were

required to continuously putt in sets of twelve until ten successful (i.e., hit) and ten unsuccessful

(i.e., misses) putts were recorded. For coding and analysis of the putt, three distinct stroke

segments (i.e., preparation, backswing/foreswing, and contact phase) and six gaze locations (i.e.,

feet, ball, club head, cup, and putting surface) were analyzed for fixation duration (stabilized

gaze for a minimum of 99.99 ms), saccades (eye movement from one location to another with a

duration of 133.2 ms), express saccades (rapid shifts lasting 66.6-99.9 ms) and smooth pursuit

(tracked object for 99.9 ms or more) movements. The results of this investigation revealed that

the low handicap (LH) golfers used significantly fewer fixations (14.2) per putt than the higher

handicap (HH) golfers (19.4), while total putt time was indistinguishable. In-line with this

finding, LH golfers maintained fewer fixations of longer duration during the preparation and

backswing/foreswing phases. Furthermore these results revealed that better golfers employ

variable gaze behaviors, including more express saccades between fixation locations and

fixations of longer duration to the ball and target, while minimizing gaze behaviors to the club

head and putting surface. Conversely, the HH golfers dedicated significantly more gaze

behaviors (fixations, saccades, etc.) to the putting surface and feet and were more consistent in

their gaze behaviors across stroke phase and fixation location.









implementing a new condition. In addition to being tested on the recall of a typical game board,

participants were exposed to a game board with pieces unsystematically arranged. Although

Chase and Simon (1973) replicated the initial findings of de Groot, they also found that the

pattern recognition abilities of the expert group were context specific. To elaborate, when the

chess pieces were arranged in a fashion comparable to an actual chess match, experts recalled the

pieces with amazing accuracy. However, when the pieces were randomly placed, performance of

the experts declined dramatically, resembling the recall accuracy of the lesser skilled group.

Similarly in sport, researchers have made use of recall paradigms, in which participants are

exposed to domain specific slides depicting the strategic positioning of players in offensive or

defensive formations. For example, Allard, Graham, and Paarsalu (1980) assessed the speed and

accuracy of pattern recognition in basketball players and non-players. Participants were exposed

to a series of slides that included structured (i.e., offensive formation) and unstructured (i.e.,

rebound) game situations for brief intervals of four seconds. Following exposure, participants

demonstrated recall ability by identifying player locations on a reduced scale magnetic board.

Allard et al. (1980) confirmed the findings of de Groot (1978) and Chase and Simon (1973),

revealing that experts were more adept at recalling structured game knowledge compared to the

non-player group, recognizing critical strategic patterns that formulate during competition.

Modifying and extending the work of Allard et al. (1980), Starkes (1987) assessed the

expert knowledge structures and pattern recognition abilities of elite, skilled, and novice field

hockey players. Similar to the Allard et al. (1980) study, participants viewed structured and

unstructured slides depicting offensive attacks and turnovers. However, to accommodate the

complexity of the field hockey pitch (11-on-11), viewing time was increased from four seconds

to eight. Once again, response accuracy and recall ability was measured by identifying player









age, expertise, and occlusion conditions. These results signify the developmental changes

associated with skill proficiency. That is, high skill combined with maturation appears to

positively influence perceptual skill but maturation has little to no effect when performance

ability is low. Figure 2-2 depicts this relationship between age effects and expertise on

anticipatory cue use as a function of the occlusion condition.

The implementation of the temporal occlusion condition identified critical age and

expertise differences during the most significant viewing times for attaining advance perceptual

cues. However, the extent to which selective attention influences prediction accuracy as a

function of age and skill could not to be determined. Therefore, in an attempt to bridge this gap

Abernethy & Russell (1987a) conducted a spatial occlusion analysis and upheld the previous

findings reported earlier. Results indicated that experts, regardless of their age, are more adept at

extracting perceptual information from both the opponent's racquet and arm whereas novice

badminton players were inept at using the kinematic cues provided by the opposing player's arm,

relying on the racquet only for advance perceptual information.

Abernethy and Russell (1987b) examined the perceptual differences of two groups distinct

in performance ability (i.e., international level and novice undergraduate students), identifying

ostensible expert-novice differences. The comparison of such variable performers lends itself to

large effects. However, Abernethy (1989) probed further in an effort to determine the sensitivity

of these differences and whether or not the perceptual advantage was merely a product of

expertise. Using a temporal and spatial occlusion task (see Abernethy & Russell, 1987a)

intermediate (n=12, skilled but not elite) and novice (n=15, undergraduate students) badminton

players were assessed on their ability to predict, from the film presented, the probable direction

of their opponent's stroke. The results of the temporal condition show intermediate-novice









context of sport, researchers have interpreted visual search strategies involving fewer fixations of

longer duration as more efficient, permitting more time for more detailed information extraction

(Williams, Davids, & Williams, 1999).

Researchers (i.e., Janelle, Hillman, & Hatfield, 2000; Vickers, 1992, 1996a, 1996b) have

turned their attention to additional gaze behavior indices that may reveal expert-novice

differences. Of these indices, a promising and robust observation is that experts exhibit an

extended quiet-eye period relative to non-experts. According to Vickers (1996a), the quiet-eye

(QE) is a temporal period when task relevant environmental cues are processed and motor plans

are coordinated for the successful completion of an upcoming task. Specifically, the QE period is

defined as the elapsed time between the last visual fixation to a target and the initiation of the

motor response (Vickers, 1996a). As such, the QE appears to functionally represent the time

needed to organize the neural networks and visual parameters responsible for the orienting and

control of visual attention (Vickers, 1996a). Collective analysis of the extant literature reveals

that experts exhibit longer quiet eye periods ( rpb = 0.62) when compared to less skilled

performers (Mann et al., 2006). Furthermore, intragroup variability has also been reported,

suggesting that longer quiet eye periods correspond with increased accuracy (Harle & Vickers,

2001; Janelle et al., 2000; Vickers, 1996a, 1996b; Vickers & Adolphe, 1997).

Despite its promise, the underlying mechanisms) responsible for the robust QE findings

remain in question. From a pure visuo-motor perspective, the QE may serve to maximize

cerebral efficiency, as reflected in cortical patterns indicative of elite performance (Janelle,

Hillman, & Hatfield, 2000; Janelle, Hillman, Apparies, Murray, Meilli, Fallon, & Hatfield,

2000). That is, previous research has consistently reported cortical quieting in visuospatial and

motor coordination tasks in the left hemisphere as compared to the right hemisphere at temporal,









duration or number of fixations across skill levels. The results of Experiment 2, as in previous

work by Abernethy and Russell (1987b), failed to identify skill-based visual search differences.

That is, the search patterns associated with fixation distribution, order, and duration remained

stable across skill levels. Abernethy concluded that the notable performance differences

exemplified in Experiment 1 were not the result of more efficient search strategies but rather the

ability to make strategic inference from the information extracted. Again, the experts were better

able to make use of the advance cues, while the novice players relied more on the ball-racquet

contact zone and subsequent ball flight information.

The majority of research reviewed has relied on interceptive racquet sports and advance

cue use for the prediction of stroke type and location. Extending the literature beyond racquet

sports Helsen and Pauwels (1990) examined the behaviors of soccer players. Fifteen expert and

15 novice soccer players were presented with 90 slides of typical offensive soccer situations. The

participants were required to view the slide and as quickly as possible verbalize the most correct

decision (i.e., shoot, dribble, or pass). Visual search behavior was recorded using a NAC-V

Eyemark recorder. The findings from this investigation indicate that experts were faster and

more accurate in finding tactical solutions to offensive soccer situations. In accord with previous

findings (e.g., Abernethy 1990a; Abernethy & Russell, 1987a), there was little difference amid

the locations of fixations between skill groups. However, in contrast to previous work (e.g.,

Abernethy 1990a; Abernethy & Russell, 1987a), analysis of the visual search data for number of

fixations revealed expert-novice differences. Experts displayed fewer fixations than novices.

These results are a sign of the improved ability of experts to make use of the information they

possess, while novices continue to seek out validation. As endorsed by Helsen and Pauwels

(1990), fewer fixations allow for faster information-processing time and result in shorter









McAdam, D.W., & Rubin, E.H. (1971). Readiness potential and vertex positive wave and
contingent negative variation and accuracy of perception. Electroencephalography &
Clinical Neurophysiology, 30, 511-517.

McAdam, D.W., & Seales, D.M. (1969). Bereitschaftspotential enhancement with increased
level of motivation. Electroencephalography & Clinical Neurophysiology, 27, 73-75.

McPherson, S.L. (1993). The influence of player experience on problem solving during batting
preparation in baseball. Journal of Sport & Exercise Psychology, 15, 304-325.

McPherson, S.L. (1994). The development of sport expertise: Mapping the tactical domain.
Quest, 46, 223-240.

McPherson, S.L. (1999). Tactical differences in problem representations and solutions in
collegiate varsity and beginner women tennis players. Research Quarterlyfor Exercise and
Sport, 70, 369-284.

McPherson, S.L. (2000). Expert-novice differences in planning strategies during collegiate
singles tennis competition. Journal of Sport & Exercise Psychology, 22, 39-62.

Milton, J., Solodkin, A., Hlusk, P., & Small, S. (2007). The mind of expert motor performance is
cool and focused. Neurolmage, 35, 804-813.

Molander, B., & Backman, L. (1989). Age differences in heart rate patterns during concentration
in a precision sport: Implications for attentional functioning. Journal of Gerontology and
Psychological Science, 44, 80-87.

Murray, N.M., & Janelle, C.M. (2003). Anxiety and performance: A visual search examination
of the processing efficiency theory. Journal of Sport and Exercise Psychology, 25, 171-
187.

Neisser, U. (1967). Cognitive psychology. New York: Appleton, Century, Crofts.

Niinimaa, V., & McAvoy, T. (1983). Influence of exercise on body sway in the standing rifle
shooting position. Canadian Journal of Applied Sport Sciences, 8, 30-33.

Nunez, P.L. (1995). Neuromodulation of neurocortical dynamics. In P.L. Nunez (Ed.),
Neurocortical dynamics and human EEG rhythms (pp. 591-627). New York: Oxford
University Press.

Orgogozo, J.M., & Larsen, B. (1979). Activation of the supplementary motor area during
voluntary movement suggests it works as a supramotor area. Science, 206, 847-850.

Papakostopoulos, D. (1978). Electrical activity of the brain associated with skilled performance.
In D.A. Otto (ed.), Multi-disciplinary perspectives in event-related brain potentials
research (pp.134-137). Washington, DC: U.S. Environmental Protection Agency, Office of
Research and Development.









(Roland, 1984). This finding supports the contention that the BP may play a role in the detection

and pairing of task relevant information with the necessary components of movement (Brunia &

van Boxtel, 2000), while reflecting the activation of a neural path linking perception to action

(Toni & Passingham, 2003). More recently, whole brain MRI data lend further support to the

results presented here, suggesting that experts develop a specialized motor program evidenced by

right brain activation that integrates visual information with the necessary motor commands for

performance (Milton et al., 2007).

The preparatory set of the performer may also be reflected in BPlate changes. For example,

Taylor (1978) reported increases in BPlate negativity in the hemisphere contralateral to the active

limb. In contrast, the cortical differences revealed here (i.e., C41ate) are ipsilateral to the dominant

limb. The golf putt is a bimanual task by nature, however, it is often argued that an expert right-

handed golfer will control the putt with his or her right hand, with the left hand simply providing

support. Although the ipsilateral cortical differences reported here apparently contradict those of

Taylor (1978), they may be explained by the bimanual coordination required for the golf putt.

Data regarding the significant difference in activation of the BPpeak component of the

right-central region (i.e., C4peak) for the LH group relative to the HH group is congruent with

previous research (Deecke, 1987; Shibasaki et al., 1980). Of the three BP components, the BPpeak

component is believed to reflect the coordinated activation of the SMA and MI. Combined,

activation of these structures play a critical role in the organization of complex motor sequences

that are rehearsed from memory and fit into a precise timing plan (Gerloff et al., 1988). As such,

this finding is not altogether surprising, given that the LH group engages in significantly more

practice and competition than the HH group. Indeed, the elite group should have a more refined

cortical representation of the task that facilitates the movement and timing patterns of the golf









In sum, research examining the many underlying attributes of expertise, including

perceptual, cognitive, and motor behavioral domains, has generally concluded that experts are

more efficient, effective, and accurate in recognizing task specific patterns, are more proficient at

making decisions, maintain superior procedural and declarative information, have a profound

reservoir of contextual cues that are systematically retrievable, and possess an unparalleled

ability to foreshadow events and outcomes (Holyoak, 1991; Starkes & Allard, 1993).

Expert Performance

Experts Are Faster and More Accurate at Recognizing Patterns

Skilled performers posses a superior ability to recognize and recall structured patterns

during performance within a specific domain as compared to their less skilled counterparts as a

result of their advanced knowledge structures that permit the chunking of small bits of

information into much larger and more meaningful constituents. Simply, the ability to recognize

and recall sport-specific cues is fundamental to memory structures in which previously presented

information is encoded into short-term memory and stored in a retrievable form in long-term

memory (Woo Sohn & Doane, 2003). It is known that experts acquire knowledge and skills to

rapidly program information in long-term memory to facilitate the retrieval of performance

information with efficiently processed contextual cues via long-term working memory (Ericsson

& Delaney, 1999).

The seminal work of deGroot (1978) with chess masters demonstrated the ability of expert

performers to rapidly perceive and store domain specific patterns into memory. Specifically,

chess players were exposed to a board configuration for brief intervals. Following exposure they

were requested to recite the location of each piece from memory. de Groot (1978) found that

chess masters were near flawless in their recall, however players of lesser caliber were

significantly less effective. Chase and Simon (1973) extended the work of de Groot (1978) by









BP Activity Across Skill Level and Cortical Region

To investigate the hypothesis that both increased negativity in mean BPiate and BPpeak

amplitude are characteristics of greater movement preparation and cerebral efficiency, cortical

activation levels in each of the BP components (i.e., early, late, peak) across skill level and

cortical region (i.e., C3, Cz, C4, P3, P4) was assessed. As anticipated, a non-significant .\kl// x

BPcomponent interaction was found (Pillai's Trace = .177, (F(o, 66) = .640, p = .774, r2 = .177). An

overall significant difference in cortical activity was evident for the main effect of ill//, Pillai's

Trace = .690, (F(5, 14) = 6.245,p = .003, r2 = .690) and BPcomponent, Pillai's Trace = .483, (F(o0, 66)

= 2.103, p = .036, r2 = .242). Combined, the significant .\1kil and significant BPcomponent main

effects suggest that not only did the LH group demonstrate more BP negativity relative to the HH

group, cortical negativity also increased from the BPearly to BPiate component for both groups,

reaching maximal negativity immediately prior to movement execution (i.e., BPpeak). Follow-up

univariate analyses of variance for the main effect of.\/k// revealed significant cortical region

differences for C4 (F(, 18) = 14.171, p = .001, d= 1.77) and P4 (F(, 18) = 8.304, p < .010, d=

1.36). Given the relative degree of relatedness among BP components, a multivariate analysis of

variance was used to examine the temporal location of the cortical region differences between

groups (Figure 4-5). Results indicated that for C4, the LH group exhibited greater negativity for

each BPcomponent (C4early, F(1, 18) = 6.023, p = .025, d= 1.16; C41ate, F(1, 18) = 17.519, p = .001, d=

1.97; and C4peak, F(, 18) = 27.425, p < .001, d= 2.47) compared to the HH group, while parietal

differences were only evident between the two groups during the early component, P4early (F(l, 18)

= 7.661, p = .013, d= 1.30).











Lastly, Pearson Product Moment correlations were conducted to explore the relationship

between QE duration and RE, and QE duration and anxiety. Results indicated a non-significant

correlation (r = .046, p = .389, d= .09) between QE and RE, and a non-significant correlation

between QE and anxiety (r = -.059, p = .359, d= -.12).

Inter- and Intra-Group Performance Variability in Quiet-Eye Duration

Assesssing both inter- and intra-group QE differences while controlling for anxiety, it

was hypothesized that the QE duration of both the LH and HH groups for successful putts would

exceed the QE duration for missed putts. Furthermore, after collapsing across skill, a prolonged

QE duration was expected for successful putts relative to missed putts. The results yielded a

significant main effect for .\kill (F(, 1793) = 51.989, p < .001, d = .34) but not for Accuracy (F(l,


1793) = 3.323, p = .068, d= .08), suggesting that expertise is reflected by a prolonged QE

duration. No other differences were noted (i.e., Accuracy x .\kill interaction, F(1, 1793)= .304, p =

.581, d= .03, Figure 4-4).



2900
2700
S2500
0
m 2300-
O 2100-
iU 1900-
1700-
o'
1500
1300
HH LH HH LH
Hit Miss


Figure 4-4 When controlling for anxiety, the LH participants demonstrate longer quiet eye
durations for successful putts as compared to missed putts. Conversely, minimal
variability in quiet eye duration is evident for the HH participant as a function of
performance, controlling for anxiety.









putt. Practice and experience may contribute to the elevated right-central cortical activation of

the LH group relative to the HH group, such that the preparatory period of the LH player reflects

an attentional process that permits the assessment, organization, and recall of the requisite motor

program from memory. The HH player likely has not developed such refined control, therefore

resulting in more deliberate cognitive processes.

The majority of sport psychophysiological research (Haufler et al., 2003) has adopted

spectral techniques for inferring the cortical role in psychomotor performance. Although this

investigation used a class of ERP (i.e., BP) to examine skilled-based differences, complimentary

findings attained from both approaches can be garnered to implicate common cortical

mechanisms. As indexed by the collective spectral findings to date, as individuals become more

skilled, the cognitive strategies used during the preparatory process and movement execution

stage become more routine, demanding fewer cognitive resources (Fits & Posner, 1967; Hatfield

et al., 1984). Direct comparisons of expert and near-expert performers have revealed cortical

asymmetry differences, such that the expert performer demonstrates a relative decrease in left

compared to right hemisphere cortical activity (i.e., increased alpha power or cortical quieting).

This finding suggests that near-expert performers require greater conscious processing of the task

and its demands. The BP data reflect greater cortical activation in the right hemisphere relative to

the left, and moreover, that the LH golfers maintain greater activation in the right hemisphere

relative to the HH golfers. Corroborating the extant spectral work, BP evidence gathered in the

current study suggests that the LH players allocate more resources to the visual-spatial

processing of the task and fewer resources to the conscious processing of the movement, linking

the visual-spatial area of the cortex to movement preparation and performance.









Chase, W.G., & Simon, H.A. (1973). Perception in chess. Cognitive Psychology, 4, 55-81.

Chi, M.T.H., Feltovich, P.J., & Glaser, R. (1981). Categorization and representation of physics
problems by experts and novices. Cognitive Science, 5, 121-125.

Craig, C.M., Delay, D., Grealy, M.A., and Lee, D.N. (2000). Guiding the swing in golf putting.
Nature, 405, 295-296.

Crews, D.L. (1989). The influence of attentive states on golfputting as indicated by cardiac and
electrocortical activity. Unpublished Doctoral Dissertation, Arizona State University,
Tempe, AZ.

Crews, D.J., & Landers, D.M. (1993). Electroencephalographic measures of attentional patterns
prior to the golf putt. Medicine and Science in Sports and Exercise, 25, 116-126.

Cui, Q.R., & Deecke, L. (1999). High-resolution DC-EEG analysis of the Bereitschaftspotential
and post movement onset potentials accompanying uni-or bilateral voluntary finger
movements. Brain Topography, 11, 233-248.

Davids, K.W., Rex Pe Palmer, D., & Savelsbergh, G.J.P. (1989). Skill level, peripheral vision
and tennis volleying performance. Journal ofHuman Movement Studies, 16, 191-202.

Davidson, R.J., Jackson, D.C., & Larson, C.L. (2000). Human electroencephalography. In J.T.
Cacioppo, L.G. Tassinary, & G.G. Berntson (Eds.), Handbook ofpsychophysiology (2nd
ed., pp. 27-52). New York: Cambridge University Press.

Deecke, L. (1973). Potential changes associated with motor action, reflex responses and
readiness. In W.C. McCallum & J.R. Knott, The responsive brain: The proceedings of the
third international congress on event-related slow potentials of the brain (pp.91-93).
Dorchester, UK: John Wright & Sons Ltd.

Deecke, L. (1987). Bereitschaftspotential as an indicator of movement preparation in
supplementary motor area and motor cortex. In Motor areas of the cerebral cortex (Ciba
Foundation Symposium, no.132, pp.231-250). Chichester, UK: Wiley.

Deecke, L., & Kornhuber, H.H. (1978). An electrical sign of participation of the mesial
'supplementary' motor cortex in human voluntary finger movement. Brain Research, 159,
473-476.

Deecke, L., & Kornhuber, H.H. (2003). Human freedom, reasoned will and the brain. In M.
Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential: Movement related cortical
potentials (pp.283-321). New York, NY: Kluwer Academic/Plenum Publishers.

Deecke, L., Kornhuber, H.H., Lang, W., Lang, M., & Schreiber, H. (1985). Timing functions of
the frontal cortex in sequential motor-and learning tasks. Human Neurobiology, 4, 143-154.









895) = 14.94, p < .001, d= .26). Lastly, elevated HR was evident among the HH group relative to

the LH group in both the low anxiety (F(1, 898) = 84.64, p < .001, d= .62) and high anxiety (F(1,

895) = 90.07, p < .001, d= .63) conditions. Figure 4-1 provides a graphic representation of these

findings.

Skill Based Putting Performance and Quiet-Eye Differences Across Anxiety Conditions

It was hypothesized that across both the low and high anxiety conditions, the LH group

would perform better than the HH group as measured by RE, BVE, and total number of putts

made. Furthermore, the expert participants were expected to exhibit a longer QE period

than the near-expert group. The LH group was more successful than the HH group (X2 = 33.59, p

< .001, d= .19) but did not differ in the magnitude of their bias (GRE). Pillai's Trace was used to

interpret the MANOVA results for putting performance and QE duration since it is deemed more

robust than the alternative test statistics (Liu, 2002). The omnibus test for putting performance

and QE duration between the LH and HH participants across anxiety conditions was significant,

Pillai's Trace = .089, (F(3, 893) = 28.90, p < .001, r2 = .089). Follow-up univariate analyses of

variance for QE (F(1, 895) = 41.509,p < .001, d= .43), RE (F(1, 895) = 36.305,p < .001, d= .40),

and BVE (F(1, 895) = 34.753, p < .001, d= .39) yielded significant differences between skill levels,

indicating that the LH group was more accurate and consistent, while demonstrating a longer QE

duration relative to the HH group (Figures 4-2 and 4-3). However, no significant differences

were noted for Anxiety, Pillai's Trace = .006, (F(3,893) =1.735, p < .158, r2 = .006). Figure 4-2

and 4-3 provide a graphic representation of the results.

Furthermore, the analysis of expertise, performance, anxiety differences, and QE duration

revealed a significant main effect for .\ki/ (F(1, 379) = 35.211, p < .001, d= .61) suggesting that

the LH group engaged in a longer QE period relative to the HH group across conditions. Despite









anxiety increased, so too would the number of fixations to the periphery of the visual display.

Specifically, they found that as cognitive anxiety increased, the viewing patterns of the expert

and novice groups' changed variably, with a marked increase in attention to peripheral cues.

Worth noting however, was a reported decrease in visual search efficiency in the novice group.

That is, while the experts in the high anxiety condition demonstrated an increase in their

respective mean fixation duration, the mean fixation duration of the novice group decreased.

Consistent with previous findings, the corresponding increase in search rate among the novice

group can be explained by a comparative decrease in processing efficiency and ineffective cue

utilization (Williams & Elliot, 1999).

Collectively, the aforementioned investigations lend support to the negative effects of

anxiety on performance, processing efficiency and cue utilization. Moreover, these findings can

be viewed as illustrating the increase in the cognitive/attentional demands which accompany

increases in anxiety and arousal. However, the increased fixation duration of the expert group in

the Williams and Elliott study, may suggest that a decrease in visual search rate permits the

allocation of additional cognitive resources to offset the attentional demands imposed by

heightened anxiety. As such, the QE period may serve an anxiety regulation function during self-

paced tasks.

The QE research has demonstrated a unique and robust relationship between performance

and gaze stability during the temporal period immediately preceding task performance. Given the

magnitude of this relationship, it is not out of the question to infer that the QE period serves a

motor planning function. However, in light of the well-established relationship between anxiety,

visual search, and performance across tasks of varying complexity, researchers have suggested

that the QE period may also reflect a temporal window for the regulation of emotion (Janelle et









in response to the film, as quickly and as accurately as possible. The results of Experiment 2

revealed that experts were faster and more accurate in responding to the various tactical

problems. Specifically, the experts demonstrated faster initiation time, ball/foot contact time, and

total response time. Of the three areas in which the experts demonstrated superiority, initiation

time offers an explanation for the ability of the expert group to make use of advance cues,

whereas the differences in ball/foot contact and total response time may be accounted for by

physical skill differences. The results of the visual search data mirrored that of Experiment 2.

Specifically, experts maintained fewer fixations than the intermediate players, but they did not

differ in the duration of fixations. Finally, during the dynamic task the expert players were more

inclined to make use of the defenders, free back, and free space as compared to the intermediate

performers, who fixated more on the ball and the goal.

The work of Helsen and Starkes (1999) supports the notion that experts process advance

cues, thereby reducing response time in both film and static soccer presentations. Furthermore,

visual search differences were evident by the fewer fixations performed by the experts, indicative

of an efficient search process, and providing support for the differential cue accessibility and

processing capabilities of expert performers.

Savelsbergh, Williams, Van Der Kamp, and Ward (2002) conducted a study examining the

skill-based differences in anticipation and visual search during the penalty kick in soccer. An

ASL 4000SU eye tracker and an Ascension Technologies magnetic head-tracker were used to

record the visual search behaviors of expert (n=7) and novice (n=7) goalkeepers while viewing a

presentation of 120 penalty-kick trials. The participants were situated behind a joystick that was

used to indicate and record both response time and accuracy. Accuracy was determined by the

placement of the joystick relative to the position of the incoming soccer ball. To further clarify, if









Experiment 3. Williams and Davids (1998) used the same participants and test procedures

as in Experiment 1 with the addition of obtaining concurrent verbal reports. Consistent with the

findings of Experiment 1, the highly-skilled players had quicker response times but were equaled

in response accuracy by the less-skilled players. Furthermore, both groups reported similar

patterns of attention allocation to the left and right sides of the box, yet the less-skilled players

spent significantly more time fixating within the box (i.e., ball and or ball carrier) than the

highly-skilled players. No further differences were apparent to search order, yet the experienced

players' fixations were more equally distributed across the different viewing areas (i.e., left,

right, and box), with the less skilled players spending more time within the box.

It can be concluded from the work of Williams and Davids (1998) that skilled players are

able to make sufficient use of advance performance cues, reducing the time necessary to derive a

conclusion and respond. More importantly, skilled players apparently rely less on information

provided by the ball and more on the information presented throughout the visual display as

evident by the frequency of fixations directed at the opposing player's hip region during the 1-

on-1 conditions. Additionally, the discrepancy between verbal reports of frequent alternation of

attention allocation and visual fixations of equal distribution during the 3-on-3 conditions

suggests that skilled players rely more on non-ball specific information gathered through the

peripheral system.

Contemporary investigations: 1999-2002

Task type and complexity have both been points of contention for eye movement

researchers (e.g., Ripoll et al., 1995; Williams & Davids, 1997). However, the extent to which

performance related anxiety mediates the visual search process has been neglected. Attending to

this gap in the literature, Williams and Elliot (1999) examined the effects of cognitive anxiety

and level of experience on anticipation and visual search behavior in karate kumite. In accord









French, K.E., & Thomas, J.R. (1987). The relation of knowledge development to children's
basketball performance. Journal of Sport Psychology, 9, 15-32.

Goulet, C., Bard, C., & Fleury, M. (1989). Expertise differences in preparing to return a tennis
serve: a visual information processing approach. Journal of Sport & Exercise Psychology
11, 382-398.

Hancock, G.H., Butler, M.S., & Fischman, M.G. (1995). On the problem of two-dimensional
error scores: Measures and analyses of accuracy, bias, and consistency. Journal of Motor
Behavior, 27(3), 241-250.

Huckabee, M.L., Deecke, L., Cannito, M.P., Gould, H.J., & Mayr, W. (2003). Cortical control
mechanisms in volitional swallowing: The Bereitschaftspotenial. Brain Topography, 16(1),
3-16.

Hardy, L., Mullen, R., & Jones, G. (1996). Knowledge and conscious control of motor actions
under stress. British Journal ofPsychology, 87, 621-636.

Harle, S.H., & Vickers, J.N. (2001). Training quiet eye improves accuracy in the basketball free
throw. The Sports Psychologist, 15, 289-305.

Hatfield, B.D., & Hillman, C.H. (2001). The psychophysiology of sport: A mechanistic
understanding of the psychology of superior performance. In R.N. Singer, H.A.
Hausenblas, & C.M. Janelle (Eds.), Handbook of sport psychology (2nd ed., pp.362-386).
New York, NY: John Wiley & Sons, Inc.

Hatfield, B.D., Hillman, C.H., Apparies, R.J., Janelle, C.M., & Vickers, J. (1999). Expert
performance states: Psychophysiological perspectives on efficiency. Journal of Sport and
Exercise Psychology, 21, s5.

Hatfield, B.D., Landers, D.M., & Ray, W.J. (1984). Cognitive processes during self-paced motor
performance: An electroencephalographic profile of skilled marksmen. Journal of Sport
Psychology, 6, 42-59.

Hatfield, B.D., Landers, D.M., & Ray, W.J. (1987). Cardiovascular-CNS interactions during a
self-paced, intentional attentive state. Elite marksmanship performance. Psychophysiology,
24, 542-549.

Hatfield, B.D., Landers, D.M., Ray, W.J., & Daniels, F.S. (1982). An electroencephalographic
study of elite rifle shooters. The American Marksman, 7, 6-8.

Haufler, A.J., Spalding, T.W., Santa Maria, D.L., & Hatfield, B.D. (2000). Neuro-cognitive
activity during a self-paced visuospatial task: Comparative EEG profiles in marksmen and
novice shooters. Biological Psychology, 53, 131-160.

Helsen, W., & Pauwels, J.M. (1990). Analysis of visual search activity in solving tactical game
problems. In D. Brogan (Ed.), Visual search (pp. 177-184) London: Taylor and Francis.









Richman, H.B., Gobet, F., Staszewski, J.J., & Simon, H.A. (1996). Perceptual and memory
processes in the acquisition of expert performance. In K.A. Ericsson (Ed.), The road to
excellence: The acquisition of expert performance in the arts and sciences, sports, and
games (pp. 167-187). Mahwah, NJ: Lawrence Erlbaum Associates.

Ripoll, H., Kerlirzin, Y., Stein, J.F. & Reine, B. (1995). Analysis of information processing,
decision making, and visual search strategies in complex problem solving sport situations.
Human Movement Science 14, 325-349.

Roland, P.E. (1984). Organization of motor control by the human brain. Human Neurobiology, 2,
205-216.

Roland, P.E., Larsen, B., Lassen, N.A., & Skinhoj, E. (1980). Supplementary motor area and
other cortical areas in organization of voluntary movements in man. Journal of
Neurophysiology, 43, 118-136.

Salazar, W., Landers, D.M., Petruzello, S.J., & Han, M. (1990). Hemispheric asymmetry, cardiac
response, and performance in elite archers. Research Quarterly for Exercise and Sport, 61,
351-359.

Savelsbergh, G.J.P., Williams, A.M., Van Der Kamp, J., & Ward, P. (2002). Visual search,
anticipation, and expertise in soccer goalkeepers. Journal of Sport Sciences, 20, 279-287.

Schmid, R., & Zambarbieri, D. (1991). Strategies of eye-head coordination. In R. Schmid & D
Zambarbieri (Eds.), Oculomotor control and cognitive processes (pp.229-248).
Amersterdam: North-Holland.

Shank, M.D., & Haywood, K.M. (1987). Eye movements while viewing a baseball pitch.
Perceptual and Motor .\kll, 64, 1191-1197.

Shibasaki, H., Barrett, G., Halliday, E., & Halliday, A.M. (1980). Components of the movement-
related cortical and their scalp topography. Electroencephalography and Clinical
Neurophysiology, 49, 213-226.

Shim, J., & Carlton, L.G. (1999). Anticipation of movement outcome through perception of
movement kinematics. Journal of Sport and Exercise Psychology, 21, S98.

Siedentop, D., & Eldar, E. (1989). Expertise, experience, and effectiveness. Journal of Teaching
in Physical Education, 8(3), 254-260.

Singer, R.N. (2000). Performance and human factors: Considerations about cognition and
attention for self-paced and externally paced events. Ergonomics, 43(10), 1661-1680.

Singer, R., Cauraugh, J., Chen, D., Sternberg, G.M., & Frehlich, S. G. (1996). Visual search,
anticipation, and reactive comparisons between highly-skilled and beginning tennis players.
Journal ofApplied Sport Psychology, 8(1), 9-26.









A anxiety and B P A activity ..................................................................... ....................158
Quiet-Eye Duration and BPcomponent Activation ........................................ .................159
Quiet-Eye Duration and Anxiety............................. .... ............... 160

5 DISCUSSION, APPLIED IMPLICATIONS, AND FUTURE DIRECTIONS ................162

D isc u ssio n .... ............... ... ......... ....... .................. .... ........ ... .......... ............... 1 6 3
Pre-Putt Levels of Cognitive Anxiety, Somatic Anxiety, and Heart Rate .................... 163
Skill Based Putting Performance and Quiet-Eye Differences Across Anxiety
Conditions ............. ......... .... ..... ........ ....... .. ... ............... 164
Inter- and Intra-Group Performance Variability on Quiet-Eye Duration......................166
BP Activity Across Skill Level and Cortical Region ...................................................167
BP Activation, Performance and Anxiety ........................................ ............... 170
Practical Im plications and Future Directions ............................................ ............... 172
S u m m a ry ................... .......................................................................... 1 7 4

L IST O F R E F E R E N C E S ..................................................................................... ..................176

BIOGRAPHICAL SKETCH ............................................................. ...........190









other maneuvers. In simple condition B, participants were to respond to openings while ignoring

the other two possibilities when present. The complex condition required the participants to react

to both attacks and openings and ignore feints. In Experiment 1, the simple conditions A and B

failed to reveal skill-based differences in response accuracy and response time. It was concluded

that expertise does not discriminate between skill levels during a simple task. However, when the

participants had to respond to attacks and openings (complex task), experts were more efficient

(i.e., time and accuracy) in their responses as compared to both the intermediate and novice

groups. Surprisingly, the novice group outperformed the intermediate group.

The findings of Ripoll et al. (1995) suggest that the expert boxers were able to recognize

the type of movement initiated by the opponent, signifying the potential for a more successful

counter. Unfortunately, this finding was not maintained in a comparison of the lesser skilled

groups. In conjunction with the efficiency differences between the experts and less skilled

boxers, skill-based differences for the accuracy of response were revealed. Again there were no

differences between the intermediate and novice boxers. It was postulated that the lack of

performance differences was due to the homogeneity of the intermediate and novice groups.

Although the novice group did not have any formal combat experience, their practice

experiences may have been equivalent to that of the intermediate group (Ripoll et al., 1995).

Nevertheless, these findings are suspect. Following Experiment 1 Ripoll et al. (1995) concluded

that expertise appears only in the presence of a complex problem space.

In an extension of the previous study, Experiment 2 explicitly assessed skill-based visual

search differences in complex conditions (Ripoll et al., 1995). For the purpose of off-line

analysis, the visual display was divided into six distinct regions, including: the (a) head, (b)

trunk, (c) arm/fist, (d) pelvis, (e) legs, and (f) other unidentified fixations. The results of









ACKNOWLEDGMENTS

The completion of this dissertation is not a reflection of one man's passion for developing

human potential, but rather a reflection of one man's potential realized. If not for the love,

support, and guidance, this realization would not have been possible. I would first like to thank

my mom and dad, Diane and Dan for their eternal support and unconditional love. My sister,

Darlene, for always believing in me, and my nephews, Joshua and Michael for their gentle

reminders of what life is really about! A special thanks to Dr. Harold Minden, Dr. Jonathan Eto,

and Dr. Peter Papadogiannis for their friendship and inspiration.

I would like to express my gratitude to my mentor, Dr. Christopher Janelle. Your patience,

wisdom, and willingness to challenge me in all facets of my academic and professional

development are greatly appreciated.

I would like to extend my sincere appreciation to the members of my dissertation

committee, Dr. James Cauraugh, Dr. Mark Tillman, and Dr. Tracy Linderholm, for their

continued support, insight, and flexibility, permitting a project that I can be proud of.

The completion of this project would not have been possible without the technical and

methodological support of Dr. Steve Coombes, Melanie Mousseau, and Rob Barnes. I am

indebted to you all for the countless hours spent in the developmental stages of this project.

Lastly, I would like to thank Melanie Mousseau. You are a day-to-day reminder of what

life has to offer. Your infectious smile, thoughtful insight, and your willingness to challenge my

emotional, intellectual, and spiritual self, has been a tremendous inspiration!









imposed (i.e., experimental control), appears to influence the relationship between fixation

duration and the total number of fixations (Abernethy, 1988, 1990b).

However, the practical significance of the systematic findings that experts perform much

quicker than novices is significant. Sport by nature requires dynamic thinking and split second

decision-making. Clearly the use of advance perceptual information aids in this process.

Although the visual search behaviors of expert athletes have been equivocal across tasks of

varying type and complexity, the issue may not be at all be related to information sought via

foveal fixations but rather a result of the information collected from the peripheral system

(Abernethy, 1988; Williams et al., 1993; Williams et al., 1999). More direct research is

warranted assessing the peripheral vision differences in expert and novice performers and their

respective abilities to gather, process, and translate that information into enhanced performance.

Moreover, experts may engage in an advanced level of preattentive processing as

suggested by Neisser (1967), therefore constructing apriori strategies to facilitate the decision-

making process. To further clarify, a common finding throughout the visual search literature was

comparable response accuracy between skill levels, however expertise was indicative of faster

decision-times suggesting that certain cues would appear to "pop-out" to the expert while

diligent pursuits were required by the less-skilled athletes.

Nevertheless, this narrative review has provided inclusive evidence as to the distinct visual

search tendencies unique to athletes of varied skill level. Caution should be taken however, when

attempting to infer even from the equivocal findings reported here. The very nature of the tasks

assessed coupled with the inconsistent use of "experts" and "novices" sufficiently impedes the

ability to decisively conclude skill-based visual search differences. In response to this limitation

a quantitative analysis of this literature base is warranted.









With direct consideration given to the potential confounds of the laboratory, Abernethy

(1990b) implemented a two experiment design. Experiment 1 was conducted inside the

laboratory and Experiment 2 on a squash court using a temporal occlusion technique

supplemented with the collection of eye movement data. The overall visual search findings of

Experiment 1 will be discussed in a later section.

In the first experiment Abernethy (1990b), squash players were required to predict the

stroke outcome from a filmed representation of an opposing player. Experimental conditions

included the same temporal conditions as reported in Abernethy and Russell (1987b). Analyses

of prediction accuracy for both stroke depth and direction across occlusion conditions revealed

significant expert-novice differences, confirming the utility of advance cues for skilled players.

In light this robust finding, improved prediction performance was found for stroke depth as

compared to stroke direction across skill levels. This finding was contrary to previous research

(i.e., Isaacs & Finch, 1983) that had reported increased difficulty for players to detect slight

racquet variations and kinematic cues responsible for changes in depth. Overall, the findings

support the rapid abilities of expert athletes to make use of more cues germane to prediction

accuracy (Abernethy & Russell, 1987a, 1987b; Jones & Miles, 1978; Isaac & Finch, 1983).

However, despite this finding, only slight visual search characteristics were evident, with the

expert allocating more fixations to their opponent's arm and head coupled with fewer fixations to

the contact zone as compared to the novice performer. Moreover, across the entire visual search

sequence no skill-based differences for fixation duration were evident.

In the second experiment, in which a field-based assessment was conducted, the same

basic conclusions were derived. Simply, although the experts outperformed the novice

comparison group, the visual search behaviors of the experts were virtually indistinguishable









Hypothesis 1

Cognitive and somatic anxiety (MRF-L) and heart rate (BPM) data were evaluated using

separate one-way repeated measure ANOVAs to determine if the presentation of a monetary

incentive, video camera, and written release would modulate anxiety levels under the high

anxiety condition.

Hypothesis 2

Given that the LH group was expected to outperform and exhibit a prolonged QE period as

compared to the HH group across anxiety conditions, both putting performance (RE, GRE, BVE)

and QE duration were evaluated using a repeated-measures multivariate analysis of variance

(RM MANOVA). Given the reported association of QE duration to performance, MANOVA

procedures were preferred over separate univariate ANOVAs. Putting performance and QE

duration were analyzed using a 2 (.\kill LH, HH) x 2 (Anxiety: High, Low) MANOVA with

repeated measures on the last factor. Post hoc procedures included univariate ANOVAs, each at

the .05 level (Stevens, 2002). The success to failure ratio between LH and HH groups was

assessed using the Chi square statistic. Furthermore, to specifically address expertise,

performance, and anxiety differences on QE duration a 2 (.\kill LH, HH) x 2 (Accuracy: Hit,

Miss) x 2 (Anxiety: High, Low) ANOVA with repeated-measures on the last two factors was

conducted. Lastly, a Pearson Product Moment Correlation was calculated to explore the

relationship between QE duration and RE.

Hypothesis 3

Quiet-eye duration has been demonstrated to account for both inter and intra-group

performance variability (Janelle et al, 2000; Vickers, 1992,1996a, 1997, 1998; Vickers et al.,

1997, 1998). Accordingly, both inter and intra-group differences in QE duration were analyzed

using a 2 (Accuracy: Hits, Misses) x 2 (.\kill LH, HH) ANOVA.









location on a reduced scale magnetic board. The results demonstrated the superior recall abilities

of experts, extending the claim that experts possess domain specific pattern recognition and

recall abilities to the sport context. More recently, research with snooker players and soccer

players has corroborated these findings (Abernethy, Neal & Koning, 1994; Williams, Davids,

Burwitz, & Williams, 1993; Williams & Davids, 1995). The notion that long-term working

memory is a requisite skill to meet the particular information-processing demands of domain

specific expertise (Ericsson & Delaney, 1999) supports these empirical findings while offering a

theoretical account for the expert pattern recognition advantage.

Experts Have Superior Procedural, Declarative, and Strategic Knowledge

Declarative knowledge is "composed of factual information regarding concepts and their

interrelationships"(Ennis, 1994, p. 167) including rules and definitions (Thomas, 1994). In

comparison, procedural knowledge is "knowledge about how to perform or use the information

[necessary to perform]" (Ennis, pg. 167). Consider that the link between a situation and the

appropriate action is known as a procedure. Therefore, procedural knowledge can be further

clarified by matching the appropriate movement given that a specific situation is coupled with

the correct motor execution. As such, response selection and motor execution are matched to the

stimulus (Thomas, 1994).

Researchers interested in determining the extent to which declarative knowledge is

responsible for distinguishing expert athletes from lesser skilled performers began by adopting a

cognitive research paradigm (see Chi, Feltovich, & Glaser, 1981) that required participants to

categorize and sort pictures of domain specific stimuli into relevant categories. For example,

Allard and Burnett (1985) assessed female basketball players from the Canadian National Team

and novice players for their ability to make associations among the contextual cues provided

within each picture. Participants were assessed on the extent (i.e., complex vs. simple) to which









performance is likely (Murray & Janelle, 2003; Williams & Elliott, 1999; Williams, Vickers, &

Rodrigues, 2002; Williams, Singer, & Frehlich, 2002). It is a commonly held belief that anxiety

leads to attentional narrowing, such that the breadth of attention is reduced and distractibility is

increased (Easterbrook, 1959), often resulting in an increase in search rate (Janelle, Singer, &

Williams, 1999; Murray & Janelle, 2003). Moreover, under periods of heightened anxiety, a

mobilization of additional attentional resources is likely to occur in an effort to offset the

deleterious effects of anxiety, ultimately rendering a less efficient performance (Eysenck &

Calvo, 1992; Janelle, 2002).

When the outcome of an ensuing event is uncertain, participants typically experience

elevated levels of stress, arousal, and anxiety (Jones & Swain, 1992). Consequently, the highly

complex, competitive, and unpredictable nature of sport often prompts heightened arousal and

anxiety. According to the Processing Efficiency Theory (PET; Eysenck & Calvo, 1992), the

interaction of individual state and trait levels of anxiety coupled with environmental constraints

(e.g., performance pressure and/or uncertainty) directly impact the functional capacity of

attention. Specifically, the processing efficiency theory implicates the capacity and resilience of

the short-term working memory system as an inherent limitation in human information

processing. Short-term working memory is thus susceptible to the adverse effects of increased

state anxiety, thereby comprising performance. That is, cognitive anxiety, which is characterized

by worry and an inability to concentrate, may redirect attention toward a preoccupation with

thoughts of evaluation and outcome expectations, and away from task relevant cues (Liebert &

Morris, 1967). However, a major tenet of the processing efficiency theory states that as anxiety

increases, performance decrements may be avoided via the recruitment and allocation of










Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE ROLE OF THE QUIET-EYE PERIOD AND THE BEREITSCHAFTSPOTENTIAL IN
AROUSAL REGULATION AND MOTOR PREPARATION FOR PERFORMANCE OF A
SELF-PACED MOTOR SKILL

By

Derek T.Y. Mann

December 2007

Chair: Christopher M. Janelle
Major Department: Health and Human Performance

Given the robust empirical support for and practical implications of the quiet-eye (QE)

period, it was my objective to assess the role of the QE period in emotion regulation and motor

preparation. The concurrent exploration of the BP and QE period under varying levels of anxiety

was designed to assess the principal mechanisms responsible for the psychomotor differences

between expert and near-expert performers. Twenty golfers were classified by their USGA

handicap rating as either a high handicap (HH; near-expert) or low handicap (LH; expert) to

permit skill-based inferences. Participants completed 45 trials in both low and high anxiety

conditions during which cognitive anxiety, somatic anxiety, heart rate, QE duration, BP activity,

and putting performance were recorded.

Results indicated that the LH golfers are more accurate and less variable in their

performance than the HH group, as revealed by measures of radial error, bivariate variable error,

and group centroid radial error. Systematic differences in QE duration and BP were also

observed, with experts exhibiting a prolonged quiet eye period and greater cortical activation in

the right-central region compared to non-experts. A significant association between cortical

activation and QE duration was also noted. Despite performing under high and low anxiety









Parent, A. (1996). Carpenters human neuroanatomy (9th ed). Baltimore, MD: Williams and
Wilkins.

Paul, G., & Glencross, D. (1997). Expert perception and decision making in baseball.
International Journal of Sport Psychology, 28, 35-56.

Pelz, D. (2000). Putting bible: The complete guide to mastering the green. New York:
Doubleday.

Pfurtscheller, G. (1992). Event-related synchronization (ERS): An electrophysiological correlate
of cortical areas at rest. Journal ofElectroencephalogry and Clinical Neurophysiology. 83,
62-69.

Pfurtschelle, G. & Neuper, C. (2003). Movement and ERD/ERS. In M. Jahanshahi & M. Hallett
(Eds.), The Bereitschaftspotential: Movement related cortical potentials (pp. 191-205). New
York, NY: Kluwer Academic/Plenum Publishers.

Pivik, R. T., Broughton, R.J., Coppola, R., Davidson, R. J., Fox, N., & Nuwer, M. R. (1993).
Guidelines for the recording and quantitative analysis of electroencephalographic activity
in research contexts. Psychophysiology, 30, 547-558.

Posner, M.I., Nissen, M.J., & Klein, R.M. (1976). Visual dominance: An information processing
account of its origins and significance. Psychological Review, 83, 157-171.

Posner, M.I., & Raichle, M. (1991). Images of mind. New York: Scientific American Books.

Radlo, S.J., Janelle, C.M., Barba, D.A., Frehlich, S.G. (2001). Perceptual decision making for
baseball pitch recognition: Using P300 latency and amplitude to index attentional
processing. Research Quarterly for Exercise and Sport, 72, 22-31.

Reagan, D. (1989). Human brain electrophysiology: Evoked potentials and evoked magnetic
fields in science and medicine. New York: Elsevier.

Reeve, T.G., Fischman, M.G., Christina, R.W., & Cauraugh, J.H. (1994). Using one-dimensional
task error measures to assess performance on two-dimensional tasks: Comments on
"attentional control, distractors, and motor performance." Human Performance, 7(4), 315-
319.

Rektor, I. (2003). Intracerebral recordings of the bereitschaftspotential. In M. Jahanshahi & M.
Hallett (Eds.), The Bereitschaftspotential: Movement related cortical potentials (pp.61-77).
New York, NY: Kluwer Academic/Plenum Publishers.

Renshaw, I., & Fairweather, M.M. (2000). Cricket bowling and the discrimination ability of
professional and amateur batters. Journal of Sports Sciences, 18, 951-957.










CHAPTER 5
DISCUSSION, APPLIED IMPLICATIONS, AND FUTURE DIRECTIONS

The quest to understand the nature of sport expertise has lead researchers to explore the

psychophysiological characteristics of superior performance. The resulting evidence suggests

that psychological efficiency underlies expert performance (Hatfield et al., 1999), which is

characterized by both regional cortical deactivation (i.e., increased alpha power, Haufler et al.,

2000) and effective and efficient cue utilization (i.e., fewer fixations of longer duration and

prolonged QE, Mann et al., 2006). However, with the exception of the early work of Janelle,

Hillman and Hatfield (2000) and Janelle et al., (2000) these two components of expertise have

not been simultaneously explored.

According to Vickers (1996a), the "quiet-eye" is a temporal period when task relevant

environmental cues are processed and motor plans are coordinated for the successful completion

of an ensuing task. As such, the QE period theoretically represents the time needed to organize

the neural networks and visual parameters responsible for the orienting and control of visual

attention to promote cerebral efficiency (Vickers, 1996a). In addition to its motor planning

function, researchers have suggested that the QE period may reflect an opportunity for emotion

regulation, thereby minimizing the deleterious effects of anxiety and/or arousal by permitting the

recruitment of task specific resources (Janelle, Hillman, & Hatfield, 2000; Janelle et al., 2000;

Vickers et al., 1999).

Performance varies with the length of the QE period, with prolonged periods generally

resulting in increased performance. The importance of understanding the complex integration of

systems associated with expert performance, however, is essential for the advancement of both

theory and practice. Therefore, I sought to better understand the mechanisms that underlie the

efficacy of the QE period by concurrently assessing QE duration, anxiety, and a premotor,









putting surface. The testing area was designed to accommodate a 12ft. putt while leaving

25.75in. of space behind the hole and approximately 22in. on either side of the hole to allow for

the measurement and analysis of accuracy, bias, and consistency. Figure 3-1 provides a graphical

depiction, detailing the dimensions of the putting platform.



192"_



48'


4"-

Figure 3-1 Putting green dimensions.

Putting Performance

The target (i.e., golf hole) was supplemented with an imposed grid used for assessing

accuracy, bias, and consistency (Reeve, Fischman, Christina, & Cauraugh, 1994; Hancock,

Butler, & Fischman, 1995). Specifically, a 30in x 40in matrix progressing in lin. increments on

both the vertical and horizontal axes was projected onto the putting surface with a Sharp

Notevision LCD Projector (Model XG-NV2U, Tokyo, Japan). The coordinate (0,0) indicates the

center of the golf hole. The image was projected after each stroke and removed prior to each

subsequent stroke to avoid latent visual assessment of performance or impairment of

performance potentially induced by the display of the grid.

Gaze Behavior

A BIOPAC electro-oculogram amplifier (EOG 100B; BIOPAC Systems, Inc., Santa

Barbara, CA), with a bandpass range from DC to 100Hz was used to record eye-movements;

specifically, QE duration. Analog data were sampled at 1000Hz using an MP 150 analog/digital

converter and recorded on-line with AcqKnowledge 7.0 (BIOPAC Systems, Inc., Santa Barbara,




























Early Late Peal
C3 BP Component

Figure 4-5a 21


1

0.5,


-0.5




-2.


Early

Figure 4-5c


U HH
0LH
-0II
-1
-1.5.
-2-
-2.5


Early


Late
C4 BP Component


Peak


Figure 4-5b


L P I 4 j T h i


Late
CZ BP Component

2-
1.5,
1-
S0.5


Peak


0 -


Late
P3 BP Component


SHH
1 |*LH


Peat


0 I I LH


Early

Figure 4-5e


Figure 4-5d


Figure 4-5 Skill based differences (i.e., mean, SE) across cortical regions and BP components.
Figure 4-5a displays a marked increase in left-central negativity across BP
components for the LH group. Figure 4-5b illustrates a significant increase in right-
central negativity across BP components for the LH group. Figure 4-5c displays a
marked increase in negativity at the vertex across BP components for the LH group.
Figure 4-5d illustrates minimal hemispheric differences in the left-parietal region
between skill. Figure 4-5e illustrates an increase in right-parietal cortical negativity
for the BPearly component (* represents p < .05).


Late
P4 BP Component


|1









strategies of an athlete. Adapted from Abernethy and Russell (1987b) the following sequence

depicts a typical approach to the implementation of a spatial occlusion paradigm and its

experimental conditions.

el: The player's racquet and arm holding the racquet were occluded;
e2: The player's racquet (but not the arm holding it) was occluded;
e3: The player's face and head were occluded;
e4: The player's lower body was occluded;
e5: Irrelevant background features were occluded.

Occlusion Research

In a seminal study, Jones and Miles (1978) adopted the temporal occlusion paradigm in

their investigation of advance cue use in lawn tennis. In support of the superior ability hypothesis

of experts to extract relevant information from advance cues, Jones and Miles concluded that

perceptual information germane to decision accuracy is readily available to athletes throughout

the flight path of a tennis serve, regardless of performance level. However, as pertinent cues

were occluded, level of expertise accounted for significant differences in prediction accuracy.

Specifically, when the serve was occluded shortly (126 ms) after impact, performance

differences were notable. Additionally, when occluded just (42 ms) prior to ball-racquet contact,

performance differences were pronounced in favor of the top-level performers, signifying the

ability of skilled tennis players to effectively make use of advance perceptual information in the

performance environment typically provided by the opposing player.

In a similar study, Isaacs and Finch (1983) assessed the anticipatory timing of beginning

(n= 34) and intermediate (n= 16) tennis players. Four temporal occlusion conditions (i.e., 10

msec before contact; 0 msec at contact; 15 msec post-contact; and 30 msec post-contact) were

implemented to examine differences between tennis proficiency level and the participant's

ability to accurately predict the placement of a tennis serve. Immediately following each viewing

condition participants indicated the anticipated landing position of the serve on a specially









Finally, it can be concluded that eye movement registration techniques cannot ensure an

accurate understanding of the information extracted from the visual display despite the ability to

index the location of the eye relative to the visual display. As Abernethy (1988) indicates the

capacity for athletes to shift their attention without additional eye movements fosters the need for

a distinction between "looking" and "seeing". To confound this problem further, little is known

of the function peripheral vision assumes for information extraction during the visual search of a

dynamic visual display.

In sum, eye movement registration techniques have furthered the understanding of the

visual search strategies of athletes. Moreover, as technology continues to push the boundaries of

applied research, more definitive ecologically valid conclusions may be drawn as to the

performance differences present during actual competition.

Quiet Eye

Although traditional eye-tracking research has been limited to the measurement of the

number, duration, and frequency of visual fixations, the location of fixations, visual pivots, and

saccades, more recent trends have focused on a measure that is believed to be associated with the

organization of visual attention and information processing, the QE period. Specifically, the

"quiet-eye period" is a measure defined as the elapsed time between the last visual fixation to a

target and the initiation of the motor response (Vickers, 1996a). For example, the QE period of a

marksman is identified as the elapsed time from the last recorded visual fixation to the

commencement of the trigger pull. Literature examining the QE period and whether or not the

QE period can mechanistically account for notable expert-novice performance differences is the

focus of this section of the review.

From a theoretical perspective, Vickers (1996a) proposed the location-suppression

hypothesis to account for the requisite behaviors necessary for aiming. For example, consider









optimal aiming period characterized by a prolonged period of rifle stability as compared to the

pre-elite performers. Furthermore, the cortical profile of the elite performers was expected to be

more negative than that of the pre-elite performers, a profile indicative of less effortful,

automatic processing.

Specifically, the BP of six elite and six pre-elite marksmen was recorded from the frontal

(Fz), centro-lateral (C3,C4), and occipital midline (Oz) cortical regions. The results indicated

that the elite performers exhibited more accurate and less variable performance than the pre-elite

performers and were significantly more stable in their respective rifle hold while preparing to

pull the trigger. However, contrary to the anticipated findings, the BP analyses revealed a greater

positive cortical shift in the elite performers as compared to the pre-elite performers. According

to Konttinen et al. (2000) the increased positivity is likely due to the sustained gross motor

activation, which is not related to the central timing mechanism (p. 175). As previously

mentioned, the work of Karrer et al. (1978) indicates that a positive shift may be the result of an

insufficient motor plan necessary to inhibit irrelevant body movements. As such, the positive

deflection reported by Konttinnen and colleagues (2000) may be the result of an increased effort

of elite marksmen to prevent extraneous movements and not at all an indication of the

preparatory set. Alternatively, the physical stress imparted by the rifle hold may have in fact

masked the BP negativity among the elite markmen (Konttinen et al., 2000).

Although Konttinen and colleagues (2000) failed to provide empirical evidence to support

the BP as a measure for assessing psychomotor differences among expert and near-expert

performers, the BP should not deemed inappropriate for the study of psychomotor performance

differences. As previously mentioned the main preparatory function of the marksmen is the

inhibition of extraneous motor activity during the stabilization of the rifle hold (Konttinen &









Previous researchers (e.g., Jones & Miles, 1978) have questioned the ecological validity of

laboratory-based investigations. Therefore, in response to such skepticism, Abernethy (1990a)

employed a subjective self-report index assessing the 'naturalness' (i.e., ability to replicate and

provide a life-like simulation) of the visual display characteristics and temporal stress associated

with squash performance. Results revealed no differences between experts and novices with a

mean group rating of 4.53 out of 7 for display characteristics and 3.47 out of 7 for temporal

stress. This finding suggests that participants were relatively engaged by the film task but

thought the trial-to-trial period was too great, failing to capture the tempo of competition.

Abernethy (1990a) also collected subjective data to isolate the importance of specific

perceptual cues that may account for the robust expert-novice differences. Both the expert and

novice groups identified the racquet and the arm holding the racquet as the most important cue, a

finding consistent with Abernethy and Russell (1987a). Empirically, novice performers appear

unable to utilize advance kinematic information in a manner similar to the experts, yet no

subjective differences identifying the most pertinent sources of information were noted. As such,

experts apparently not only know where to look but they also maintain a wealth of declarative

information, information that is seemingly unavailable to the novice performer.

The laboratory basis of the previous occlusion studies, although sound methodologically,

have been questioned for their ecological validity. For example, Jones and Miles (1978) discuss

the inherent sterility of the laboratory and the inability of a laboratory setting/task to accurately

elicit comparable performance states. Such limitations may confound the empirical estimates of

perceived expert-novice differences. Additionally, Abernethy and Russell (1987b) questioned the

validity of the film presentation used in previous research and suggested that the use of film may

neutralize any notable expert-novice differences.









arm, non-dominant arm, and court position). Results from the temporal occlusion condition for

lateral error again supported previous findings, indicating that experts extracted information

earlier in the visual display (i.e., 160-80 ms prior to ball/racquet contact) and more information

(i.e., kinematic cues) than novice players, enhancing the prediction accuracy for stroke outcome.

Additionally, experts were better able to predict stroke force (depth) as compared to novice

squash players across occlusions conditions. Surprisingly, both the expert and novice groups

continually improved prediction accuracy from (tl) through (t5). Previous findings have noted

that novice performers typically do not improve performance until much later in the display (t3),

however this does not appear to be the case here. According to Abernethy (1990a), squash

players may have a tendency to reveal more accessible information about stroke depth earlier in

their preparation than badminton players, implying that the existing occlusion condition failed to

suppress a sufficient amount of advance information.

Results from the spatial occlusion condition for lateral error mirrored those of the

temporal condition, in that the experts were more accurate in their predictions across conditions.

However, contrary to previous investigations (i.e., Abernethy, 1988; Abernethy & Russell,

1987a) that have reported differences under occlusion condition (el)-(e2) for experts and as early

as (e2)-(e3) for novice performers, no within group performance differences were noted across

occlusion conditions. Again, expertise differences were noted for depth error, with the skilled

players predicting more accurately across all occlusion conditions. In accord with previous

findings, both experts and novices demonstrated performance decrements when both the racquet

and arm were occluded. Yet, unique to squash, the opponent's head was also a significant source

of advance information as illustrated by systematic performance decrements in prediction

accuracy across groups while the head was occluded.









concurrent investigation of these variables may provide greater insight into the underlying

psychological processes of both phenomena.

In summary, researchers have provided evidence, although inferential, in support of the

potential mechanisms of superior visuo-motor performance as evidenced by the QE. However, to

date, researchers have failed to offer a concrete theoretical or mechanistic account of the QE. As

such, the following investigation attempts to explicitly assess two mechanisms, motor

programming and/or emotion regulation that may account for the robust relationship between QE

duration and performance.

Statement of the Problem

Sport psychophysiologists interested in expertise have provided evidence for a unified

theme of psychological efficiency, offering descriptions from the cortical and visual search

domains (Hatfield, Hillman, Apparies, Janelle, & Vickers, 1999). The extant body of literature

permits the conclusion that expert performance is characterized by regional cortical deactivation

(i.e., increased alpha power) coupled with efficient cue utilization (i.e., fewer fixations of longer

duration and prolonged QE). Although a relationship between the QE period and cerebral

efficiency has been reported (Janelle, Hillman, Apparies, Murray, Meili, Fallon, & Hatfield,

2000), the extent to which such a relationship is evident is tenuous. Therefore, of primary interest

in this investigation is the assessment of the underlying mechanisms of the QE period

responsible for the psychomotor superiority of the expert (i.e., emotion regulation and/or motor

planning).

The Current Study

To assess the extent to which the proposed mechanisms correspond with modulations of

the QE period, the BP, which has been proposed as an electrophysiological index of cerebral

efficiency during the premotor period (Papakostopoulos, 1978), was assessed under high and low









response time, equating to better performance. Essentially, fewer fixations place less demand on

long-term working memory and experts can retrieve an appropriate response to the problem-

space more proficiently with the use of their extensive knowledge base.

Returning to racquet sports, Cauraugh, Singer, and Chen (1993) investigated the visual

search patterns and anticipation strategies used to predict an opponent's tennis stroke. Expert

(n=30) and novice (n=30) tennis players viewed two different types of tennis serves (i.e., spin or

flat) and had to identify its direction (i.e., left, right, or center court) as quickly and accurately as

possible. An Eye-Trac, Model 210 was used for recording eye fixations. A fixation was

operationally defined as a stationary eye for a minimum of 133 ms. The visual search data

revealed significant expert-novice differences. Specifically, of the nine predetermined areas for

fixations, the novice players demonstrated a preference for the head and left shoulder, with no

other notable differences with experts present. Additionally, experts dedicated more time per

location, that is, their fixations were of longer duration than the novice players. When

considering the visual search patterns across the three phases of the serve (i.e., preparation,

execution, and follow-through), the novice players fixated more frequently on the head during

the preparation phase. Similarly, skill-based differences were evident for ground strokes with the

novice players fixating more frequently on their opponent's hips. No other visual search

differences between groups were found. However, consistent with previous findings, the experts

were both quicker and more accurate in predicting the type and location of their opponent's

shots.

Despite the promising results of these studies in terms of identifying expert-novice visual

search differences, with the exception of Helsen and Pauwels (1990), research reviewed to this

point has involved interceptive racquet sports and the correct identification of stroke force,









latest point in the delivery, this process may confound early prediction accuracy in this

investigation as compared to the robust early expert-novice differences as seen elsewhere

(Abernethy & Russell, 1987a, 1987b).

Previous research has reported equivocal findings as to the proficiency for anticipation of

lateral and depth positioning among expert and novice athletes, with depth positioning being

more difficult to predict with relative accuracy. Such difficulties may prove troubling to sports

(i.e., cricket, baseball, and tennis) that require acute depth perception while anticipating lateral

direction. For example, a bowler in cricket will bounce the ball in front of the batter and

depending on the spin, trajectory, and speed, the ball will react differently with the ground,

forcing the batter to anticipate and react accordingly. The batters task in cricket is not unlike that

of baseball, requiring the use of advance perceptual cues including kinematic cues provided by

the bowler. Ball speed and rotation must also be processed effectively for the batter to make

contact.

Renshaw and Fairweather (2000) assessed the perceptual discrimination ability of national

(n=6), regional (n=6), and club (n=6) cricket batters exposed to 5 different types of bowling

deliveries (legspinner, toward batters feet; topspinner, drops short followed by high a bounce;

googly, away from batter; flipper, close to feet with low bounce; backspinner, similar to flipper

with lower bounce) from two temporal occlusion conditions. The first condition contained the

bowler's run-up and ball flight up to and including ground contact, while the second condition

included the bowler's run-up and the first 80 ms of ball flight. Immediately following each trial

the batter verbally reported the delivery type. The results revealed that overall the experts were

more accurate than the regional and club level players for delivery discrimination. Consistent

with previous findings, skill level was indicative of advance cue use (condition 1), however,









the neural networks responsible for increased visual-spatial processing and the recruitment of the

requisite motor program.

To better understand the covert psychological indices of the expert advantage, researchers

have made extensive use of electroencephalography (EEG) and spectral analysis techniques to

investigate cortical activation and hemispheric specialization during the preparatory period of

self-paced closed motor skills such as golf putting (Crews & Landers, 1993), archery (Salazar,

Landers, Petruzzello, & Han, 1990), and shooting (Deeny, Hillman, Janelle, & Hatfield, 2003;

Hatfield, Landers, & Ray, 1984, 1987; Hillman, Apparies, Janelle, & Hatfield, 2000; Janelle,

Hillman, & Hatfield, 2000). Analysis of EEG spectral power, in particular, has revealed that the

effectiveness and efficiency of expert performance has a cortical signature that differs from that

of non-experts (Deeny et al., 2003; Hatfield et al., 1984; Haufler et al., 2000; Janelle, Hillman, &

Hatfield, 2000; Landers et al., 1994). That is, as individuals progressively become more skilled,

the cognitive strategies employed during the planning and execution of movement become more

routine, demanding fewer cortical resources (Fitts & Posner, 1967; Smith, McEvoy, & Gevins,

1999), resulting in a demonstrable increase in left hemisphere alpha power (i.e., decrease in

cortical activity) and performance. The comparison of cortical activation across hemispheres at

corresponding reference sites permits an index of hemispheric asymmetry.

Within the psychomotor literature, researchers have demonstrated relatively stable cortical

activation across hemispheres in the novice performer, where as the expert reliably demonstrates

a pronounced asymmetrical ratio, characterized by a relative increase in left hemisphere to right

hemisphere alpha power (i.e., decreased cortical activity). Simply stated, the novice performer

requires greater conscious processing (i.e., verbal analytic processing) of the task demands

resulting in greater left hemisphere activation (Hatfield et al., 1984). Conversely, the expert









decrease in the preparatory period immediately preceding trigger pull. That is, when performing

a well-learned skill, the continued acquisition and analysis of visual information immediately

prior to performance may interfere with the formulation and execution of the requisite motor

program. As such, cortical idling or quieting in the occipital lobe would suggest a reduction in

visual attention. Therefore, Loze et al. (2001) examined the cortical activity of six expert air-

pistol shooters across best and worst shots to determine the mediating role of occipital EEG

alpha power on shooting performance.

Continuous electroencephalographic activity was collected from Oz, T3, and T4, and was

referenced to linked mastoids. Post acquisition analysis consisted of the reduction of pre-shot

data into three 2 sec EEG epochs for which spectral analysis was conducted, yielding absolute

alpha (8-13Hz) power values of best and worst shots. Results indicated that occipital alpha

power was greater in best shots, with the greatest difference in power noted in the third epoch

(i.e., the final 2-second period immediately prior to trigger pull). Moreover, not only did alpha

power increase during best shots, alpha power was found to decrease in the final epoch prior to

worst shots. Corroborating previous EEG work (e.g., Hatfield et al., 1984; Hatfield and Landers,

1987), hemispheric asymmetry was also reported. That is, significantly greater alpha power was

evident in the left temporal region (T3) as compared to the right temporal region (T4), a finding

consistent with the notion of cerebral efficiency.

In-line with the dominant theory of motor expertise (Fits & Posner, 1967), the work of

Loze et al. (2001) lends support to the notion that skilled performance is optimized when

conscious processing during skill execution is minimized. In this case, visual information

processing gives way to the formulation and execution of the necessary motor program,

increasing the probability of optimal performance. Although, compelling, such conclusions are









The anxiety manipulation used in this investigation successfully modulated both self-report

and physiological indices of anxiety and arousal. However, the lack of statistical support for the

emotion regulation function of the QE period may have been underestimated based on the

current results. For example, data trends suggest that as anxiety increased so too did QE, yet the

reported differences between high and low anxiety conditions were not significant. Perhaps

future studies should consider including multiple manipulations varying in relative anxiety. For

example, the current manipulation included a monetary reward for performing well, coupled with

the anxiety provoking experience of performing in front of a camera. In either case, there was

very little at risk for the participants. Considering that most people are more adversely affected

by the thought of losing money and welcome the chance to win money, future research may

consider implementing a monetary penalty for unsuccessful or poor performance.

The expert-near-expert paradigm that was used here has proven successful, providing

insightful information into the relative skill based differences. However, as previously alluded to,

a practical implication of these findings is to train the QE in a manner that would facilitate

information processing and sensorimotor efficiency. Although it is reasonable to expect that the

QE and BP differences found here would correspond with the very best players (i.e.,

professional), without explicitly testing them, the magnitude and direction of the QE period and

relative cortical changes associated with them remain uncertain. Until such observations can be

made, the development of a QE training protocol should be cautiously undertaken.

Summary

The purpose of this investigation was to clarify the role of the QE period in the

preparatory process of a self-paced motor skill. The concurrent exploration of the BP and QE

period under varying levels of anxiety was designed to assess the underlying mechanisms that

link QE duration and performance. Twenty golfers were classified by their USGA handicap









THE ROLE OF THE QUIET-EYE PERIOD AND THE BEREITSCHAFTSPOTENTIAL IN
AROUSAL REGULATION AND MOTOR PREPARATION FOR PERFORMANCE OF A
SELF-PACED MOTOR SKILL


















By

DEREK T.Y. MANN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2007









Furthermore, the results of the early psychophysiological work provide empirical support

for the relationship between pre-performance cortical specificity and the quality of performance

of highly skilled participants, with greater asymmetry corresponding with increased

performance. As such, this work lends credence to Fitts & Posner's 1967 Theory of Skill

Automaticity, such that skilled performance is optimized when conscious processing during skill

execution is minimized.

The systematic observation of the slow wave, Bereitschaftspotential (BP) provides a

window to assess cortical activity central to the planning and initiation of voluntary movement.

During target shooting, research has revealed that marksmen tend to allocate attention to the

features of the target and away from the conscious processing of the motor components of the

task. Accordingly, the cortical signature of successful performance is depicted by an increase in

slow potential negativity. Conversely, sub-optimal and novice performance is characterized by

an attentional focus that is directed toward the requisite mechanics (i.e., gun hold) of the task in

which the cortical signature during the pre-performance period is denoted by slow potential

positivity, or an otherwise decreased readiness to act.

The systematic observation of expert and novice performers across domains has proven

invaluable. Since the seminal work of Chase and Simon (1973), research has supported the

notion that experts posses an extensive knowledge base that facilitates both stimulus recognition

and procedural execution (Richman, Gobet, Stazewski, & Simon, 1996). Furthermore, the advent

of eye-movement registration techniques coupled with various electro-cortical modalities have

advanced the current understanding of the covert psychological behaviors distinguishing the

expert from novice performer. Accordingly, the continued exploration into the role of the quiet









3. Variability in the duration of the QE would account for both inter and intra-group
performance variability. More specifically, the LH group would not only exhibit a longer QE
duration as compared to the HH group, but the QE duration of both LH and HH groups for
successful putts would exceed the QE duration for missed putts. Research with golfers
(Vickers, 1992), marksmen (Janelle et al, 2000), basketball players (Vickers, 1996b),
biathletes (Vickers et al., 1999), and volleyball players (Vickers, 1997, 1998) indicates that
experts not only demonstrate a longer QE when compared to less skilled performers, but
within group performance differences are also notable.

4. Given that the BP is representative of the cortical mechanisms responsible for movement
preparation and that increased negativity in the mean BPpeak and mean BPiate amplitude
characterizes greater movement preparation and cerebral efficiency (Chiarenza, Vasile, &
Villa, 1990; Papakostopoulos, 1978; Taylor, 1978), it was expected that the LH group would
exhibit a greater BPlate amplitude coupled with a greater BPpeak amplitude compared to the
HH group.

5. The mean amplitude of the BPlate and the mean amplitude of the BPpeak were predicted to
discriminate between putts made and putts missed regardless of skill level. It has been
suggested that the BP develops during the premotor period, a temporal period that
corresponds with decreased heart rate and electromyographic activity that is both steady and
tonic (Chiarenza et al., 1990). In accord with Lacey and Lacey's (1978) intake-rejection
hypothesis, research assessing heart rate patterning and performance has reported that heart
rate deceleration during the preparatory period facilitates sensorimotor efficiency, such that
an orienting of attention to task relevant cues is permitted, thereby yielding performance
increases (Boutcher & Zinsser, 1990; Crews, 1989; Hatfield & Hillman, 2001; Molander &
Backman, 1989; Tremayne & Barry, 1990, 2001). Similar to the sensorimotor efficiency
denoted by heart rate deceleration, the amplitude of the BPpeak is more pronounced with
greater task involvement (McCallum, 1976) and is proportional to the preparatory set
(Loveless & Sanford, 1974). Appropriately, an increase in mean BPearly amplitude and mean
BPlate amplitude, coupled with an increase in BPpeak amplitude, would be evident for putts
made as compared to putts missed.

6. According to Eysenck and Calvo (1992), an increase in anxiety is likely to result in a
decrease in processing efficiency, stemming from the additional cognitive/attentional
resources needed to perform a desired task. Furthermore, Eysenck (1982) postulates that,
although anxiety may elicit a negative affective state, such increases in anxiety may also
serve to increase motivation. As such, it is plausible that as environmental task constraints
increase perceived anxiety, the corresponding appraisal of task difficulty may also increase.
Therefore, I hypothesized that increased negativity for each component of the BP (i.e.,
BPearly, BPlate, and BPpeak) amplitude would be greater in the high anxiety condition as
compared to the low anxiety condition due to a relative increase in the complexity of the task
(Lang et al., 1989) and the increased mobilization of resources (i.e., motivation) to
successfully complete the task.

7. The QE period is believed to index the time needed to organize the cognitive and visual
components associated with the orienting of visual attention and execution of a self-paced
task. Cortical and sub-cortical generators associated with the BP are also responsible for the









Although the results of these three investigations support the prediction of expert

efficiency (i.e., fewer fixations of longer durations) they must be taken with caution. First, none

of the investigations maintained an outcome measure, thus failing to identify the efficacy of the

search strategy differences. Second, the temporal brevity of the shooting experiments does not

lend itself to directly testing the efficiency hypothesis (i.e., temporal constraints only permitted

one fixation), although clear differences were noted for fixation location.

Satisfying the limitations of the previous studies, Shank and Haywood (1987) assessed the

abilities of varsity collegiate (n=9) and novice (n=9) baseball batters to predict pitch type (i.e.,

fastball or curveball), while recording the visual search patterns during the preparatory phase

(i.e., wind-up and delivery) of a baseball pitch. Participants viewed a video of a pitcher on a

regulation mound delivering fastballs and curveballs from either the wind-up or the stretch

position from a right-hand batter's perspective. Eye movements were recorded with an Applied

Science Laboratory (ASL) Model 210 Eye-Trac. Results from this investigation support the

general findings from the occlusion research. The expert players were more adept at extracting

pertinent advance cues compared to the novice players, respectively, as signified by the

percentage of correctly identified pitches (i.e., 84.4% vs. 64.3%). Eye movement reaction time

was assessed and defined as the moment from the point of ball release to the next eye movement.

Results revealed a comparable latency period across groups, suggesting that ball tracking is not

occurring and that such "quiet time" may be optimized by experts to process and formulate a

motor plan (see Quiet-Eye), while novice players maybe dedicating this time to searching for a

stimulus response, suggesting that the extraction of advance perceptual cues almost certainly

occurs during the wind-up and the initial stage of the delivery.









to participants while maintaining the same protocol as in Experiment 2. A NAC-V eye

movement registration system was use to record visual search behavior.

Experiment 2 (Helsen & Starkes, 1999). Expert (n=14) and intermediate (n=14) soccer

players viewed 30 slides depicting a typical offensive soccer situation from the perspective of the

ball carrier. Upon slide presentation, the participant was to respond verbally as quickly and

accurately as possible, identifying the most appropriate move (i.e., shoot, dribble, pass). The

results of Experiment 2 indicated that the expert group performed better than the intermediate

group for trials in which dribbling and passing were the appropriate responses for both response

time and accuracy. No skill-based differences were revealed for shooting scenarios. The results

of the visual search data demonstrated that experts maintained fewer fixations than the

intermediate players but did not differ in the duration of fixations. Analysis of the scan-paths

during correct decisions did not reveal any group differences. However, skill-based differences

were illustrated when considering the scan-paths during incorrect decisions. Specifically, the

expert group spent more time viewing the ball carrier, reducing the amount of time spent in

searching other areas, while the intermediate group spent less time fixating on the goal. The

truncated search pattern of the experts during incorrect decisions reflects the behavior more

characteristic of novice players who allocate their attention to a limited focal point, which is

often the ball rather than using a more systematic proportional distribution of fixations across the

visual display.

Experiment 3 (Helsen & Starkes, 1999). The procedure and participants were the same

as those reported in Experiment 2, with two modifications. First, the simulation consisted of 30

different tactical soccer situations that were projected life-size using a dynamic film approach.

Second, the participants performed a tactical decision with the ball (i.e., shoot, dribble, or pass)









BP Activation, Performance and Anxiety

It has been suggested that the amplitude of the BPpeak is more pronounced with greater task

involvement (McCallum, 1976) and is a cortical reflection of the preparatory set of the

participant (Loveless & Sanford, 1974) indicative of sensorimotor efficiency. It was therefore

hypothesized that an increase in mean BPearly amplitude, mean BPiate amplitude, coupled with an

increase in BPpeak amplitude would be evident for putts made as compared to putts missed.

Furthermore, because an increase in perceived anxiety can increase the relative complexity of a

given task (Eysenck & Calvo, 1992), it is reasonable to expect that additional cognitive resources

may be recruited to accommodate the increased effort necessary to sustain performance

(Eysenck, 1982). In accord with Eysenck and Calvo (1992), an increase in anxiety can result in a

decrease in processing efficiency due to the additional cognitive/attentional demands imposed of

the performer necessary to complete the desired task. As a result, it was hypothesized that an

increase in cortical negativity would be evident across BP components (i.e., BPearly, BPiate, and

BPpeak) in the high anxiety condition as compared to the low anxiety condition due to a relative

increase in task complexity and resource mobilization (i.e., motivation) needed to successfully

complete the task under elevated anxiety (Lang, et al., 1989).

Unexpectedly, the BP differences between "hits" and "misses" and between the low and

high anxiety conditions were minimal. Although these differences are not statistically significant,

data were broadly congruent with the original hypotheses. For example, BPpeak amplitude was

invariably more negative across cortical regions (i.e., C3, C4, Cz, P3, and P4) for successful

trials relative to unsuccessful trials, a finding that parallels the earlier work of Crews and Landers

(1993) with golfers, Landers et al., (1991) with archers, and Konttinen and Lyytinen (1992) with

marksmen. However, the inverse was true for the BPearly amplitude; greater negativity was

apparent for unsuccessful trials. With respect to anxiety, both BPearly and BPpeak amplitude were









Lyytinen, 1992) by evaluating the electro-cortical modalities of event-related potentials (ERP)

that occur across a range of frequency bands. According to Fabiani, Gratton, and Coles (2000),

ERPs offer additional insight into the cortical manifestations that precede or follow a discrete

event. The ERP is derived from the average of multiple responses and represents the temporal

relationship of cortical activation to a specific event, thereby providing a time-locked index of

the psychological correlates of performance. The bereitschaftspotential (BP), or readiness

potential (RP) first described by Kornhuber and Deecke (1965), is one class of ERP that lends

itself well to the study of the preparatory period preceding task execution. The BP is a negative

potential that is characterized by a distinct cortical signature that precedes an actual, intended, or

imagined event by 1 to 1.5 seconds and indexes anticipatory attention and movement preparation

(Jahanshahi & Hallett, 2003).

The BP is a visually distinct waveform comprised of three components, each of which are

temporally and cortically diverse (Deecke, Scheid, & Kornhuber, 1969). The early slow rising

negativity (BPearly) reflects the activation of the supplementary motor area (SMA) and begins

approximately 1500 ms prior to movement onset. The early activation of the BPearly has a

widespread scalp distribution with maximal potentials recorded at the vertex (Deecke, 1987).

According to Roland (1984; Roland, Larsen, Lassen, & Skinhoj, 1980), the SMA serves to

retrieve and/or augment the requisite motor commands from memory. Accordingly, the more

elaborate the motor sequence, the more precise the corresponding movement should be, as

indexed by an increase in SMA activation (i.e., increased negativity). The second component,

known as the BPlate, is characterized by a change in the steepness of the waveform's slope, which

occurs approximately 400-500 ms prior to movement onset, and is known to reflect activation of

the primary motor cortex (MI; Deecke, 1987; Shibasaki et al., 1980). Changes in BPlate have been









additional cognitive resources, thereby preventing the decline in available working memory

resources while sustaining performance.

Recent gaze behavior research in sport (e.g., Janelle, Singer, & Williams, 1999; Murray &

Janelle, 2003; Williams & Elliot, 1999; Williams, Vickers, & Rodrigues, 2002) offers empirical

support for the theoretical tenets put forth by Eysenck and Calvo (1992) and Easterbrook (1959),

such that visual search efficiency declines with increases in stress and anxiety, while preserving

performance. Furthermore, skill-based differences in sport have been reliably reported, with

expert performers engaging in fewer fixations of longer duration as compared to less skilled

participants (Mann, Williams, Ward, & Janelle, 2006; Williams, Davids, & Williams, 1999).

Accordingly, such skill-based visual search differences may be the result of greater self-

monitoring of performance processes among novice performers, leaving them with fewer

cognitive resources available for task execution as compared to experts.

The seminal work of Janelle et al. (1999) in sport, examined changes in gaze behavior

under varying levels of anxiety during a simulated auto-racing task. During the course of the

driving task, participants were required to detect and discriminate between relevant and

irrelevant cues randomly presented about the periphery of the visual display while being

subjected to random manipulations in anxiety. Results indicated that as anxiety increased,

processing efficiency and task performance decreased. Additionally, such changes were coupled

with a marked increase in gaze behavior variability (i.e., more fixations of shorter duration). As

such, Janelle et al. (1999) concluded that heightened anxiety can lead to increases in attentional

narrowing, resulting in the ineffective search and use of perceptual cues.

In a related study, Murray and Janelle (2003) employed a dual-task auto-racing simulation

to assess the relationship between visual search behaviors and modulations in anxiety. Murray







BPpeak


BPlate

BPearly


-1.5 -1.0 0.5 0
t
EMG onset


Figure 2-6 Temporal schematic of the Bereitschaftspotential (BP) prior to movement onset.
Adapted from Jahanshahi, M., & Hallett, M. (2003). The Bereitschaftspotential: What
does it measure and where does it come from. In M. Jahanshahi and M. Hallett (Eds.),
The Bereitschaftspotential: Movement Related Cortical Potentials (1-17). New York,
NY: Kluwer Academic/Plenum Publishers.


I


"\,









suggesting less cognitive involvement and effort among expert performers as compared to less

skilled performers (Fitts & Posner, 1967). However, knowledge of the extent to which the expert

exerts less effort or is otherwise more efficient has primarily been the product of deductive

reasoning stemming from decision-making, visual-search, and response time and accuracy

paradigms. Accordingly, the implementation of electrocortical modalities such as

electroencephalography (EEG) have served to identify the covert cognitive processing activity

and the corresponding momentary changes in cortical activity patterns across tasks and skill

levels with excellent temporal resolution supporting the cerebral efficiency hypothesis.

EEG and spectral analysis techniques investigating cortical activation and hemispheric

specialization during the preparatory period of self-paced closed motor skills such as golf putting

(Crews & Landers, 1993), archery (Salazar, Landers, Petruzzello, & Han, 1990), and shooting

(Deeny, Hillman, Janelle, & Hatfield, 2003; Hatfield, Landers, & Ray, 1984, 1987; Hillman, et

al., 2000; Janelle, Hillman, & Hatfield, 2000) have reported a progressive increase in alpha

power in the left hemisphere and a relative stability in alpha power in the right hemisphere of

elite performers as compared to less skilled performers. Given that increased alpha power is

inversely related to cortical activation, the information processing style of the expert is deemed

more efficient than that of the less skilled performer.

In addition to the continuous EEG and spectral techniques, the use of event-related cortical

potentials (ERP) has been a useful tool for determining the time-locked cortical processes

associated with a specific event. For example, the Bereitschaftspotential (BP; Kornhuber &

Deecke, 1964) which occurs in the preparatory period (i.e., 1-1.5s) immediately preceding a

voluntary motor action is believed to represent the requisite preparation for the execution of a

motor act (i.e., the motor program) (Brunia & van Boxtel, 2000). Research with marksmen









cues as compared to the novice players. Unique to this qualitative analysis however, was the

experts' frequent use of the hips as a primary source of information during the ground stroke. It

can be speculated that such a difference emerges during ground strokes as result of the close

proximity of the arm-racquet complex to the hip region. Furthermore, during the ground stroke

the hips may be a better indication of stroke direction; however, further research is warranted to

explore this possibility.

Williams and Davids (1997) conducted two investigations to examine the relationship

between visual search behavior and concurrent verbal reports in a test of advance cue use and

decision-making in soccer. Experiment 1 was a replication of the 11-on-11 anticipation test used

in the previous work of Williams et al. (1994). However, in this investigation, participants

verbalized the location of their attention throughout each viewing sequence. An ASL 4000 SU

eye movement registration system was used to record visual search patterns of experienced

(n=10) and less experienced (n=10) soccer players across a total of 26 soccer action sequences

presented on a 3.5m x 3m projection screen. A reference grid was imposed on the field to

facilitate scoring and to provide a point of distinction for the verbal responses indicating the

location of the final pass destination. Results indicated that no expert-novice differences were

evident for reaction time or response error. Consistent with the previous findings (e.g.,

Abernethy & Russell 1987b; Goulet et al., 1989), no skill-based differences for search rate were

revealed, yet search order differences were noted. Specifically, the expert players displayed a

systematic search strategy, maintaining a visual pivot (i.e., from the box to other areas and back

to the box); however no differences were noted for the verbal reports of attention direction.

Although the expert athletes fixated more frequently to the box, the novice players sustained









occlusion paradigm in Experiment 2 that was discussed in a previous section. Participants

viewed 27 randomly presented tennis serves and identified, as quickly as possible, the type of

serve (i.e., flat, slice, or top-spin) delivered. The number of correct responses, number of

fixations, scan-paths, and favored exchanges (i.e., direction of one fixation to the next) were

assessed and scored according to 11 distinct regions isolated and coded for analysis. The serve

itself was further divided into three distinct segments, the ritual (i.e., precedes serve and includes

ball bounce and footwork), preparatory (i.e., initiation of the ball toss concluding with its apex),

and execution (i.e., begins with knee extension and concluding with ball-racquet contact) phases

in attempt to isolate the nature of the advance cues used. For response accuracy, experts (M=

69.9) correctly identified more serves than did the novice group (M= 52.2). For search rate,

experts displayed more fixations than did novices during the ritual phase and specifically

attended to the head and shoulder/trunk complex. Search patterns were similar across groups

during the preparatory phase, with both groups favoring the server's head and the anticipated ball

location. However, differences emerged again during the execution phase, with the experts

terminating their fixations on the racquet much quicker than the novices, while the novice

players proceeded to track the ball after ball-racquet contact. Expert players were apparently

more attuned to the kinematic cues presented during the ritual phase, thereby aiding the decision-

making process. Moreover, the perception of advance cues corresponded with the decreased

decision time inferred from the temporal brevity of the terminal fixation on the racquet as

compared to the novice players who sought further information in order to confirm their decision

by tracking the flight-path of the ball.

Empirical and methodological advancements: 1990-1998

The limited and equivocal findings presented in the visual search literature were

questioned by Abernethy (1990) who ascribed the lack of skill-based differences in his own









Janelle, C.M., Hillman, C.H., & Hatfield, B.D. (2000). Concurrent measurement of
electroencephalographic and ocular indices of attention during rifle shooting: An
exploratory case study. International Journal of Sports Vision, 6(1), 21-29.

Janelle, C.M., Singer, R.N., & Williams, A.M. (1999). External distraction and attentional
narrowing: Visual search evidence. Journal of Sport & Exercise Psychology, 21(1), 70-91.

Jasper, H.H. (1958). Report of the committee on methods of clinical examination in
electroencephalography. Journal ofElectroencephalography and Clinical Neurophysiology, 10,
370-375.

Jones, C., & Miles, J. (1978). Use of advance cues in predicting the flight of a lawn tennis ball.
Journal of Human Movement Studies, 4, 23 1-23 5.

Jones, G., & Swain, A.B.J. (1992). Intensity and direction dimensions of competitive state
anxiety and relationships with competitiveness. Perceptual andMotor .ill/\, 74, 467-472.

Kerick, S.E., Douglass, L.W., & Hatfield, B.D. (2004). Cerebral cortical adaptations associated
with visuomotor practice. Medicine & Science in Sports & Exercise, 36, 118-129.

Konttinen, N., Landers, D.M., & Lyytinen, H. (2000). Aiming routines and their electrocortical
concomitants among competitive rifle shooters. Scandinavian Journal of Medicine &
Science in Sports, 10(3), 169-177.

Konttinen, N., & Lyytinen, H. (1992). Physiology of preparation: Brain slow waves, heart rate,
and respiration preceding triggering in rifle shooting. International Journal of Sport
Psychology, 23, 110-127.

Konttinen, N., & Lyytinen, H. (1993). Brain slow waves preceding time-locked visuo-motor
performance. Journal of Sport Sciences, 11, 257-266.

Konttinen, N., Lyytinen, H., & Era, P. (1999). Brain slow potentials and postural sway behavior
during sharpshooting performance. Journal of Motor Behavior, 31, 11-20.

Konttinen, N., & Lyytinen, H., & Konttinen, R. (1995). Brain slow potentials reflecting
successful shooting performance. Research Quarterly for Exercise and Sport, 66, 64-72.

Kornhuber, H.H., & Deecke, L. (1965). Himpotentialanderungen bei Willkirbewegungen und
passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale.
PfJ/7gei Archiv, 284, 1-17.

Krane, V. (1994). The mental readiness form as a measure of competitive state anxiety. The
Sport Psychologist, 8, 189-202.

Lacey, B.C., & Lacey, J.L. (1978). Two-way communication between the heart and the brain:
Significance of time within the cardiac cycle. American Psychologist, 33, 99-113.









related potentials) permits a concurrent, real-time look into the underlying cortical structures

contributing to the psychological processes accounting for skill differences. Self-paced tasks

demanding visuomotor coordination such as golf putting, archery, and marksmanship have been

extensively used in the sport sciences. Each of the aforementioned tasks inherently requires the

participant to remain relatively motionless during the preparatory period, reducing the potential

confound of movement artifact in the EEG waveform. Moreover, each task places great demand

on the psychomotor systems (i.e., attention, emotion, motor control and preparation) conducive

to psychophysiological recording (Hatfield & Hillman, 2001).

To date, the majority of research in sport has relied on spectral analysis techniques to

address issues of hemispheric asymmetry during the preparatory period to infer the covert

psychological processes associated with superior performance. As such, research stemming from

the spectral domain has consistently reported cerebral efficiency in favor of the expert. However,

the use of event-related potentials (ERP) is far less prevalent, despite the advantage of the

functional significance of the information processing transactions that are manifested in the

components of the ERP as a function of context (Fabiani, Gratton, & Coles, 2000). Of interest

here is the Bereitschaftspotential. Therefore, the purpose of this section is to provide a review of

the electro-cortical literature in sport as it relates to both the spectral and ERP components

during the preparatory period of a self-paced task.

Spectral Activity

Sport psychophysiologists (e.g., Crews & Landers, 1993; Hatfield, Landers, & Ray, 1984,

1987; Salazar, Landers, Petruzzello, & Han, 1990) investigating the covert cortical processes of

skilled performers have made extensive use of the electroencephalographic (EEG) spectral

analytic techniques, comparing and contrasting levels of hemispheric activation. Specifically, the

decomposition of the EEG waveform, permits the analysis of a specific frequency band in the









to regions of the display, search order, or search rate during the 3-on-3 conditions. Conversely,

visual search differences in the 1-on-l conditions were evident. Specifically, the experts had a

tendency to pivot (i.e., more fixations of shorter duration) their visual attention between the ball

and their opponent's hip region compared to the less-skilled players. Although no statistical

differences were reported for fixation location, the experts spent more time fixating on the area

between their opponent's knees and chest. Williams and Davids (1998) reported an effect size of

1.08 signifying the practical significance of a mean difference of this magnitude for soccer

players in a 1-on-l confrontation.

Experiment 2. Williams and Davids (1998) used the same procedure and participants as

those reported in Experiment 1 with the exception of 2 additional spatial occlusion conditions for

the 3-on-3 task, and four spatial occlusion conditions for the 1-on-l task. Specifically, for the 3-

on-3 anticipation test the spatial occlusion conditions consisted of (el) the removal of all

irrelevant perimeter information by cropping the visual display to include only the positions and

movements of the players and (e2) a reduction of the visual display to include only the ball or the

ball passer. For the 1-on-l anticipation test, the spatial occlusion conditions consisted of the

removal of (el) the opponent's head and shoulders; (e2) the hips; (e3) the lower leg; and (e4) an

irrelevant area of the display. The results from the 3-on-3 conditions showed significant

performance decrements across occlusion conditions for the skilled group (i.e., from (el) to (e2))

and stable performance across conditions for the less-skilled players. Nevertheless, the highly-

skilled players maintained their advantage over the novices performing quicker and more

accurately under both occlusion conditions. The results of the l-on-1 occlusion condition

revealed a similar performance advantage for experts across occlusion conditions; however, no

statistical significance was reported.









LIST OF FIGURES


Figure page

2-1 An information-processing account of the advantages of advance cue usage.
(Adapted from E. Buckolz, H. Prapavesis, and J. Fairs (1988). Advance cues and
their use in predicting tennis passing shots. Canadian Journal of Sport Science,
13(1), 20-30). ............................................................................135

2-2 Radial error for expert and novice badminton players as a function of the degree of
temporal occlusion. (Adapted from B. Abernethy (1988). The effects of age and
expertise upon perceptual skill development in a racquet sport. Research Quarterly
for Exercise and Sport, 59(3), 21-221) .................. ............... ................ ............. 136

2-3 Lateral and depth error for expert and novice wicketkeepers as a function of the
degree of temporal occlusion. (Adapted from D.R. Houlston & R. Lowes (1993).
Anticipatory cue-utilization processes amongst expert)..................... ................ 136

2-4 Error scores for experts and novices. A depiction of an atypical trend in anticipatory
cue use when comparing expert with novice participants. (Adapted from G. Paul and
D. Glencross (1997). Expert perception and decision making in baseball.
International Journal of Sport Psychology, 28, 35-56)...........................137

2-5 Scan-paths of expert, intermediate, and novice boxers. Arrows describe the direction
of gaze movements between locations and the proportional associations between
locations. The size of each circle is proportional to the percentage of fixation at each
location. (Adapted from H. Ripoll, Y. Kerlirzin, J.F. Stein, and B. Reine (1995).
Analysis of information processing, decision-making, and visual strategies in
complex problem solving sport situations. Human Movement Science, 14, 325-349)....137

2-6 Temporal schematic of the Bereitschaftspotential (BP) prior to movement onset.
Adapted from Jahanshahi, M., & Hallett, M. (2003). The Bereitschaftspotential:
What does it measure and where does it come from. In M. Jahanshahi and M. Hallett
(Eds.), The Bereitschaftspotential: Movement Related Cortical Potentials (1-17).
New York, NY: Kluwer Academic/Plenum Publishers............................138

3-1 Putting green dim tensions ........................................... ....................... ............... 140

4-1 Mean cognitive anxiety (Figure 4-la), somatic anxiety (Figure 4-1b), and heart rate
(Figure 4-1c) across skill and anxiety conditions. ................................... ..................... 152

4-2 Performance variability of the HH and LH groups are indicated as the distance from
the target and the magnitude of performance bias across anxiety conditions (i.e.,
Group Centroid Radial Error [GRE]). ........................................ ......................... 154

4-3 Prolonged QE duration across skill but not anxiety highlights the trend supporting
the expert advantage. ............................................................................ .............. 154














*- 12 Years
-- 15 Years
+--- 18 Years
--- Adults


Novice Players
3.2 r


- 12 Years
- 15 Years
---- 18 Years
- Adults


2.4


o 2.0


1.2 I-


I I I I I
-4 -2 Contact 2 Full
Frames Frames Contact 2 Full
Frames Frames Frames Display
(tl) (t2) (t3) (t4) (t5)
Time of Occlusion


I I I I I


-4 -2 Contact 2
Frames Frames Frames
(tl) (t2) (t3) (t4)
Time of Occlusion


Full
Display
(t)


Figure 2-2 Radial error for expert and novice badminton players as a function of the degree of
temporal occlusion. (Adapted from B. Abernethy (1988). The effects of age and
expertise upon perceptual skill development in a racquet sport. Research Quarterly
for Exercise and Sport, 59(3), 21-221).


20 -

10 -
I I I I
T1 T2 T3 T4
Stage of Temporal Occlusion


Non-Expert Depth Expert Depth Non-Expert Lateral Expert Lateral


Figure 2-3 Lateral and depth error for expert and novice wicketkeepers as a function of the
degree of temporal occlusion. (Adapted from D.R. Houlston & R. Lowes (1993).
Anticipatory cue-utilization processes amongst expert).


Expert Players
3.2 r


2.0 -


1.2 I-









rating as either a high handicap (near-expert) or low handicap (expert), to permit skill-based

inferences. Participants completed 45 putts in both the low and high anxiety conditions during

which cognitive anxiety, somatic anxiety, heart rate, QE duration, BP activity, and putting

performance were recorded.

As expected, the LH group's putting performance was superior to the HH group's, with the

LH group successfully completing more putts while missing closer in proximity to the target

across anxiety conditions. Moreover, the QE period of the LH group was longer than the HH

group's, yet this relationship was not maintained across anxiety conditions. Furthermore, QE

duration did not differ for successful (i.e., hits) and unsuccessful (i.e., misses) putts both across

and within participants. As expected however, the LH group displayed greater BP negativity in

the right-central, right-lateral, and the right-parietal regions relative to the HH group and this

increase was significantly related to an increase in QE duration.

Taken together, these results suggest that expert players operate with a greater level of

automaticity and less conscious processing of the requisite movements than their near-expert

counterparts. Moreover and paramount to this investigation, the data lend empirical support to

the motor programming implications of the QE duration over the arousal regulation function as

indicated by the relationships between QE and radial error and QE and BP negativity around the

primary motor cortex, as well as the lack of a demonstrable relationship between anxiety and

QE.









only kinematic information was available proving to be the contributing factor in the expert-

novice difference, with the experts using advance kinematic cues to predict the ball's landing

position. Although somewhat of an abstract inference, subsequent research has directly validated

this point.

Abernethy and Russell (1987a) implemented a spatial occlusion task, in which specific

and relevant display features were masked from the participants. Consistent with the notion that

kinematic cues may discriminate expert and novice performers, Abernethy & Russell (1987a)

hypothesized that expert-novice differences would arise as the result of the expert performer's

ability to pick up on task relevant cues earlier in the movement sequence, cues unattended to by

less skilled performers. The spatial characteristics of the relevant advance kinematic cues used

by expert (n=20) and novice (n=35) badminton players were assessed while predicting the

landing position of a badminton shuttle. Selected areas of the display were occluded, removing

task relevant kinematic cues including the opponent's racquet and arm, racquet only, head and

face, lower body, and irrelevant background features. Results revealed that skill level influenced

the reliance on advance cues, namely relevant kinematic cues. Specifically, expert performance

significantly improved from racquet and arm occlusion to racquet only occlusion, whereas as the

novice performers showed no additional performance change from one condition to the next. It

can be concluded that the expert performers derived task pertinent information from both the

racquet and the arm as compared to the novice group that appeared to benefit only from

information provided by the racquet. Expert-novice differences in the ability to use spatial cues

supports the notion that experts are more adept and in tune with the movement strategies of their

opponents, ultimately improving decision-making under tight time constraints.









maximal height (1125 ms), (3) Preparatory Phase until ball/racquet contact (1208 ms), (4) Ritual

Phase until ball/racquet contact (4710 ms), (5) entire serve without occlusion(5048 ms)). Bard

and colleagues present results are consistent with previous occlusion research in that expert

tennis players are better able to extrapolate and interpolate perceptual information to facilitate

prediction accuracy. Additionally, experts not only use the information they extract differently

they require less information all together (Goulet, Bard, & Fleury, 1989). Unique to this

investigation however, the authors also assessed decision time and concluded that experts were

not only more accurate but they arrived at their decisions much quicker than novices. Perhaps

this distinction is the result of requiring less information to reach their conclusion while

benefiting from the information that appears earlier in the display as compared to the novice

players who depend on information presented much later in the event sequence.

To identify the specific visual cues used during squash performance as well as any

systematic differences that may arise in cue use between expert and novice performers,

Abernethy (1990a) applied the same paradigm to the study of expert and novice squash players

(for the occlusion conditions applied, please refer to Abernethy and Russell, 1987a discussed

earlier). Participants (expert, n=16; novice, n=20) viewed a sequence of film depicting variations

of squash stroke and were required to verbally respond to the force and direction of their

"opponent's" shot (i.e., down or cross for direction and long or short for force) with the intention

of predicting the terminal point of the stroke. Unique to this investigation, participants were

interviewed following the experiment and asked to reflect upon their experience and the

"naturalness"(p.23) of the task as compared to that of an actual squash experience. Additionally,

participants were probed as to the importance of each of the seven perceptual cues made

available during the testing session (i.e., opposition's racquet, head, lower body, torso, dominant









handicap) and the HH (10-12 handicap) groups may not be as covertly evident. Previous research

employing the expert-novice paradigm has typically assessed the overt and covert behaviors of

vastly disparate groups (i.e., national, international level performers versus absolute beginners).

As a result, the existing literature base is comprised of research that has capitalized on the pure

magnitude of expected differences, and as a result, practical inference is often futile. This study

however, has revealed both overt and covert differences between groups that differ in skill, but

are similar in ability. As such, both the noted differences and lack thereof, may more accurately

represent the true mechanisms of the expert advantage.

Alternatively, the extent to which the anxiety manipulation was effective may be subject to

debate. That is, although both the LH and HH groups reported anxiety differences, and these

differences were supported by objective physiological changes, the additional cognitive demands

imposed may not have been sufficient enough to warrant changes in the recruitment and

allocation of cognitive resources that would be reflected in distinct cortical patterns.

Practical Implications and Future Directions

To obtain expert status, athletes must excel in no less than four domains: physiological,

technical, cognitive (tactical/strategic; perceptual/decision-making), and emotional

(regulation/coping; psychological) (Janelle & Hillman, 2003). This investigation corroborates the

notion that expert and near-expert athletes not only differ in their performance proficiency, but

they also differ in their underlying psychological mechanisms responsible for performance (i.e.,

cognitive). The extended QE period of the LH group relative to the HH group speaks to the

cognitive advantage of the expert. The significant relationship between right-central (i.e., C4)

cortical activation and QE duration support the notion of relative sensorimotor efficiency of

expert athletes. It is well understood that expert performers maintain cortical activation levels

that permit a more efficient and effective performance outcome. The visual search literature









from of ones opponent (e.g., stance, racket position, speed, body position etc...) can provide

critical information necessary to foretell future actions. As a result, outcome expectancies are

developed such that the performer engages in either selective preparation (Figure 2-1, J) or

anticipatory mobilization (Figure 2-1, H). Therefore, anticipatory mobilization, a product of

seeking and using information from advance perceptual cues can result in the elimination of

reaction time and the subsequent reduction of movement time by simply permitting the

commencement of a response sequence prior to the completion of an opponent's action

sequence. Alternatively, selective preparation can reduce reaction time by matching a stimulus

cue with a desirable response. For example, preparing to return a flat serve hit down the center

line that plays out as anticipated; resulting in a pairing of the stimulus and the response. The

ability of expert performers to extract perceptual cues can alleviate the temporal constraints

imposed by reaction time. Extensive declarative knowledge can be used to formulate a priori

scan paths to facilitate anticipation permitting extended movement times otherwise restricted.

Extensive empirical work has repeatedly demonstrated quicker response times favoring

this notion of advance cue use. For example, Helsen and Pauwels (1992) presented penalty kicks,

2-on-2, and 3-on-3 video clips to experienced and inexperienced soccer athletes. Participants

were required to physically respond to the scenarios presented on the film by executing a shot on

goal or a pass to a teammate. The findings demonstrated that the experienced players were

quicker and more accurate in their responses.

In an attempt to extend the work of Helsen and Pauwels (1992), Williams, Davids,

Burwitz, and Williams (1994) investigated skill-based differences in anticipation using

experienced and inexperienced soccer players in 11-on-11 soccer situations. Participants in this

investigation responded verbally as quickly as possible to the final pass destination. Consistent









performance cues in order to anticipate and predict future events (Ericcson, Krampe, & Tesch-

Romer, 1993). Experts appear to maintain an uncanny ability to recall past performance

information providing for accurate decisions of "what to do," while demonstrating skill

superiority (i.e., "how to do it"). However, although the expert may possess a rich knowledge

base, processing during the task appears to operate automatically free from cognitive constraints

(Anderson, 1982).

Moreover, other researchers have defined expertise as a superior task specific problem

solving ability resulting from an extensive array of long-term memories that enhance pattern

recognition (Holyoak, 1991). In comparison, Ennis (1994) suggests that experts possess and

maintain a diverse repertoire of skills and strategies that can be employed within specific

situations and contextual requirements. Elite level coaches have confirmed laboratory findings,

stating that the discrimination of expert from novice athletes can be demonstrated along four

dimensions, anticipation (i.e., attention to and interpretation of relative performance cues),

declarative knowledge (i.e., knowledge of tactical and rule based information), self-knowledge

(i.e., sense of own strengths and limitations), and greater cognitive latitude (i.e., quicker and

more diverse problem solving) (Lyoka & Bressan, 1998).

Furthermore, technological and methodological advances permitting the assessment of the

covert processes associated with performance have contributed to a more complete

understanding of the expert advantage. For example, the measurement of electrocortical activity

(i.e., EEG), heart rate (HR), and gaze behavior, have revealed pronounced hemispheric alpha

asymmetry, cardiac deceleration, and longer QE periods respectively, each of which has been

independently associated with superior performance (e.g., Hatfield, et al., 1984; Tremayne &

Barry, 2001; Vickers, 1996a).









Hypothesis 7

The relationship between the amplitude of the BPpeak and QE duration was evaluated using

a Pearson Product Moment correlation. Additionally, Pearson Product Moment correlations were

conducted to assess the relationship between relative anxiety and QE duration and anxiety and

cortical activation within each of the BPcomponents across cortical regions.

Hypothesis 8

As anxiety increased it was expected that both the LH and HH groups would exhibit a

longer QE period. To evaluate the mean QE duration differences in the high anxiety condition as

compared to the low anxiety condition, across skill levels, a repeated measures ANOVA was

employed. Specifically, QE duration was analyzed using a 2 (.kil// LH, HH) x 2 (Anxiety: High,

Low) ANOVA with repeated measures on the last factor.









In a follow-up study, Abernethy and Russell (1987b) replicated the temporal and spatial

occlusion technique as reported earlier (Abernethy & Russell, 1987a). However, in attempt to

extrapolate to the expert advantage of superior anticipatory cue use, eye movement registration

equipment was implemented to assess the search strategies of expert and novice badminton

players (for a discussion of the visual search findings, please see the section entitled Visual

Search). Results from this study echo those of earlier work. Expert players maintained a

significantly lower prediction error rate across all temporal occlusions conditions except (tl),

167 msec prior to racquet-shuttle contact, and across all spatial occlusion conditions except (el),

the occlusion of racquet and arm. In the current context, the authors concluded that only the

experts possessed the necessary knowledge structures to systematically construct perceptual

connections from the information provided in the environment to that which was necessary for

successful performance.

Despite the robust expert-novice differences across a variety of racquet sports, Abernethy

(1988) set out to understand the developmental characteristics of perceptual skill and selective

attention contributing to such differences using both temporal and spatial occlusion techniques.

Matched groups of relative expert and novice badminton players from four distinct age groups

(i.e., 12yrs, 15yrs, 18yrs, adult) were assessed using the same occlusion task previously

employed by Abernethy (1987b). It was hypothesized that the expert group would

systematically use advance cues more proficiently than novices and that this distinction would

become more apparent across groups as the participants became older. Consistent with previous

findings in sport (Abernethy & Russell, 1987a, 1987b; Isaac & Finch, 1983) it was further

predicted that selective attention would vary as a function of skill level. Analysis of the radial

error in the temporal occlusion condition revealed significant performance differences among









contends that the active search of the environment beyond a certain point provides redundant and

often distracting sources of information. However, prolonged fixations, such is the case with the

QE period, permit the detailed processing of information and even the cortical organization

necessary for performance. As such, once a basic understanding of the requisite environmental

cues and movement sequences have been learned, QE training may facilitate the relative cortical

quieting necessary to perform at a higher level. In other words, although deliberate practice is

believed to result in the automaticity of movement, systematically training the QE may augment

practice effectiveness by allowing the mind to process the visual-spatial characteristics of the

task while permitting the organization of the neural networks responsible for movement.

With specific reference to golf, when addressing the putt, the typical golfer will spend a

brief moment estimating the distance, speed, and line of putt to the target. However, at the onset

of the putting stroke, most non-expert players revert to conscious processing of the putting

stroke. For example, Vickers (2004) reported that non-expert golfers often track the putter-head

as it traverses back and through ball contact, a behavior not as evident in highly skilled putters.

Arguably, this ineffective behavior can interfere with the visual-spatial cues previously attended.

Encouraging and/or training the QE in this case can alleviate such inefficient gaze behavior,

thereby rendering the performer more effective.

This investigation was the first to assess the mechanisms responsible for skill-based QE

differences. As previously mentioned, the results from this study support the motor

programming/movement preparation function of the QE duration over the emotion regulation

function. Given that this is a seminal investigation, replication of these findings would lend to

the empirical and theoretical support of the QE period for movement preparation and to

determine whether these findings are generalizable to other self-paced tasks.









eye period may serve to better link the visual, emotional, and motor programming components of

expert performance.









1000 ms of data post movement was retained for the subsequent analysis of the BP and its

constituents. All retained data were baseline corrected (McAdam & Rubin, 1970). Baseline was

operationally defined as the first 1000ms epoch preceding BPearly onset. The data were parsed

according to group (i.e., HH and LH), for which four conditions (i.e., high-anxiety-hit, high-

anxiety-miss, low-anxiety-hit, and low-anxiety-miss) per participant, free of muscular and eye-

movement artifacts were derived, averaged, and measured.

Following the reduction and classification of data, a grand average BP was generated

according to level of expertise, anxiety, and performance to facilitate subsequent analyses. For

each average across each electrode site, a custom LabVIEW analysis program was used to detect

and calculate the BPearly, BPlate, and BPpeak components, measured as follows. It has been

routinely established that BP onset, if present, should appear approximately 1500 ms prior to

EMG onset (Jahanshahi & Hallett, 2003), and was therefore declared as the point of demarcation

for subsequent calculations. Specifically, the duration of the BPearly is the difference in time from

the point of BP onset (i.e., 1500 ms) to the start of the BPlate component. The BPlate component

becomes visually apparent 400-500 ms pre-movement and concludes with the BPpeak (Deecke et

al., 1969; Shibasaki et al., 1980). The BPpeak was calculated as the average amplitude occurring

100 ms prior to movement onset (Slobounov, Tutwiler, Rearick, & Challis, 1999). The BPlate

was calculated as the mean amplitude beginning 500 ms prior to movement onset, concluding

100 ms prior to EMG onset. Lastly, the BPearly was calculated as the average amplitude of the

interval occurring 1500-500 ms prior to EMG onset.

Data Analysis

The following section outlines the data analytic procedures for each of the eight

aforementioned hypotheses. The calculation of effect size estimates (i.e., Cohen's d) served to

quantify skill-based differences across the variety of dependent measures.




Full Text

PAGE 1

1 THE ROLE OF THE QUIET-EYE PERIOD A ND THE BEREITSCHAFTSPOTENTIAL IN AROUSAL REGULATION AND MOTOR PREPARATION FOR PERFORMANCE OF A SELF-PACED MOTOR SKILL By DEREK T.Y. MANN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Derek T.Y. Mann

PAGE 3

3 To the memory of my mom (August 10, 1947 to October 4, 1997). Without her love, support, and sacrifice this journey would not have been possible!

PAGE 4

4 ACKNOWLEDGMENTS The completion of this dissertation is not a reflection of one mans passion for developing human potential, but rather a reflection of one mans potential realized. If not for the love, support, and guidance, this reali zation would not have been possibl e. I would first like to thank my mom and dad, Diane and Dan for their eter nal support and uncondition al love. My sister, Darlene, for always believing in me, and my nephews, Joshua and Michael for their gentle reminders of what life is really about! A special thanks to Dr. Harold Minden, Dr. Jonathan Eto, and Dr. Peter Papadogiannis for th eir friendship and inspiration. I would like to express my grat itude to my mentor, Dr. Christ opher Janelle. Your patience, wisdom, and willingness to challenge me in all facets of my academic and professional development are greatly appreciated. I would like to extend my sincere apprecia tion to the members of my dissertation committee, Dr. James Cauraugh, Dr. Mark Tillm an, and Dr. Tracy Linderholm, for their continued support, insight, and flexibility, permitting a project that I can be proud of. The completion of this project would not ha ve been possible wit hout the technical and methodological support of Dr. Steve Coombes, Melanie Mousseau, and Rob Barnes. I am indebted to you all for the countless hours spent in the developmental stages of this project. Lastly, I would like to thank Melanie Mouss eau. You are a day-to-d ay reminder of what life has to offer. Your infectious smile, thought ful insight, and your willingness to challenge my emotional, intellectual, and spiritual self, has been a tremendous inspiration!

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION..................................................................................................................13 Expertise Related Differences in Cortical Activity................................................................16 Expertise, Visual Search and Emotion Regulation.................................................................21 Limitations.................................................................................................................... ..........24 Statement of the Problem....................................................................................................... .26 The Current Study.............................................................................................................. .....26 Hypotheses..................................................................................................................... .........27 Definition of Terms............................................................................................................ ....30 Assumptions.................................................................................................................... .......31 Significance of the Study...................................................................................................... ..32 2 REVIEW OF LITERATURE.................................................................................................34 Capturing Expertise............................................................................................................ ....34 Definition of Expertise........................................................................................................ ...34 Expert Performance............................................................................................................. ...36 Experts Are Faster and More Accu rate at Recognizing Patterns....................................36 Experts Have Superior Procedural, D eclarative, and Strategic Knowledge...................38 Experts Are Superior at An ticipating Opponents Actions.............................................40 Experts Maintain Superior Perception of Relevant Kinematic Information...................43 Experts Maintain More Efficient and Effective Visual Search Patterns.........................46 Experts Demonstrate Physiological Patterns Indicative of Sensorimotor Efficiency.....46 Visual Perception in Sport..................................................................................................... .48 Occlusion Paradigms.......................................................................................................49 Occlusion Research.........................................................................................................50 Summary........................................................................................................................ ..69 Visual Search.................................................................................................................. ........71 Eye Movement Registration............................................................................................72 Eye Movement Research.................................................................................................73 The seminal years: 1976-1989.................................................................................73 Empirical and methodological advancements: 1990-1998......................................78 Contemporary investigations: 1999-2002................................................................92 Visual search summary and review........................................................................100

PAGE 6

6 Quiet Eye...................................................................................................................... .102 Quiet Eye Summary and Review..................................................................................106 Expertise, Visual Search and Emotion Regulation...............................................................106 Expertise, Visual Search and Emo tion Regulation: Summary and Review.........................112 Cortical Activity and the Preparatory Period........................................................................112 Spectral Activity............................................................................................................113 Expertise Differences in Cortical Activity....................................................................119 Coherence......................................................................................................................122 Bereitschaftspotential....................................................................................................123 Cortical Activity and the Prepar atory Period Summary and Review...................................132 3 METHODS........................................................................................................................ ...139 Participants................................................................................................................... ........139 Instrumentation................................................................................................................ .....139 Putting Surface..............................................................................................................139 Putting Performance......................................................................................................140 Gaze Behavior...............................................................................................................140 Cortical Activity (Ber eitshaftspotential).......................................................................141 Electromyogram............................................................................................................141 Anxiety........................................................................................................................ ..142 Heart Rate..................................................................................................................... .142 Procedure...................................................................................................................... ........142 Data Reduction................................................................................................................. ....145 Putting Performance......................................................................................................145 Electromyogram............................................................................................................145 Heart Rate (BPM)..........................................................................................................145 Gaze Behavior...............................................................................................................146 Cortical Activity............................................................................................................146 Data Analysis.................................................................................................................. ......147 Hypothesis 1..................................................................................................................148 Hypothesis 2..................................................................................................................148 Hypothesis 3..................................................................................................................148 Hypothesis 4..................................................................................................................149 Hypotheses 5 and 6........................................................................................................149 Hypothesis 7..................................................................................................................150 Hypothesis 8..................................................................................................................150 4 RESULTS........................................................................................................................ .....151 Participants................................................................................................................... ........151 Pre-Putt Levels of Cognitive Anxiety, Somatic Anxiety, and Heart Rate....................151 Skill Based Putting Performance and Quiet-Eye Differences Across Anxiety Conditions..................................................................................................................153 Interand Intra-Group Performance Va riability in Quiet-Eye Duration.......................155 BP Activity Across Skill Le vel and Cortical Region....................................................156 BP Activation and Putting Outcome.............................................................................158

PAGE 7

7 Anxiety and BP Activity...............................................................................................158 Quiet-Eye Duration and BPcomponent Activation.............................................................159 Quiet-Eye Duration and Anxiety...................................................................................160 5 DISCUSSION, APPLIED IMPLICATI ONS, AND FUTURE DIRECTIONS ..................162 Discussion..................................................................................................................... ........163 Pre-Putt Levels of Cognitive Anxiety, Somatic Anxiety, and Heart Rate....................163 Skill Based Putting Performance and Quiet-Eye Differences Across Anxiety Conditions..................................................................................................................164 Interand Intra-Group Performance Va riability on Quiet-Eye Duration......................166 BP Activity Across Skill Le vel and Cortical Region....................................................167 BP Activation, Performance and Anxiety.....................................................................170 Practical Implications and Future Directions.......................................................................172 Summary........................................................................................................................ .......174 LIST OF REFERENCES.............................................................................................................176 BIOGRAPHICAL SKETCH.......................................................................................................190

PAGE 8

8 LIST OF TABLES Table page 4-1 Pearson Product Moment correlations demonstrating the regional specificity associated with the relationship betw een anxiety and cortical activation........................160

PAGE 9

9 LIST OF FIGURES Figure page 2-1 An information-processing account of the advantages of advance cue usage. (Adapted from E. Buckolz, H. Prapavesis and J. Fairs (1988). Advance cues and their use in predicting tennis passing shots. Canadian Journal of Sport Science, 13(1), 20-30 )....................................................................................................................135 2-2 Radial error for expert and novice badmin ton players as a function of the degree of temporal occlusion. (Adapted from B. Abernethy (1988). The effects of age and expertise upon perceptu al skill development in a racquet sport. Research Quarterly for Exercise and Sport, 59(3), 21-221)............................................................................136 2-3 Lateral and depth error for expert an d novice wicketkeepers as a function of the degree of temporal occlusion. (Adapted from D.R. Houlston & R. Lowes (1993). Anticipatory cue-utilization processes amongst expert )..................................................136 2-4 Error scores for experts and novices. A depiction of an atypical trend in anticipatory cue use when comparing expert with novice pa rticipants. (Adapted from G. Paul and D. Glencross (1997). Expert percepti on and decision making in baseball. International Journal of Sport Psychology, 28, 35-56)...................................................137 2-5 Scan-paths of expert, intermediate, and novice boxers. Arrows de scribe the direction of gaze movements between locations a nd the proportional asso ciations between locations. The size of each circle is proporti onal to the percentage of fixation at each location. (Adapted from H. Ripoll, Y. Ke rlirzin, J.F. Stein, and B. Reine (1995). Analysis of information processing, deci sion-making, and visual strategies in complex problem solving sport situations. Human Movement Science, 14, 325-349)....137 2-6 Temporal schematic of the Bereitschaft spotential (BP) prior to movement onset. Adapted from Jahanshahi, M., & Hallett, M. (2003). The Bereitschaftspotential: What does it measure and where does it come from. In M. Jahanshahi and M. Hallett (Eds.), The Bereitschaftspotential : Movement Related Cortical Potentials (1-17). New York, NY: Kluwer Acad emic/Plenum Publishers...................................................138 3-1 Putting green dimensions.................................................................................................140 4-1 Mean cognitive anxiety (Figure 4-1a), so matic anxiety (Figure 4-1b), and heart rate (Figure 4-1c) across skill and anxiety conditions............................................................152 4-2 Performance variability of the HH and LH groups are indicated as the distance from the target and the magnitude of performance bias across anxiety conditions (i.e., Group Centroid Radial Error [GRE])..............................................................................154 4-3 Prolonged QE duration across skill but not anxiety highlights the trend supporting the expert advantage........................................................................................................154

PAGE 10

10 4-4 When controlling for anxiety, the LH pa rticipants demonstrate longer quiet eye durations for successful putts as compar ed to missed putts. Conversely, minimal variability in quiet eye durat ion is evident for the HH participant as a function of performance, controlling for anxiety...............................................................................155 4-5 Skill based differences (i.e., mean, SE) acro ss cortical regions and BP components. Figure 4-5a displays a marked increase in left-central negativity across BP components for the LH group. Figure 4-5b illust rates a significant increase in rightcentral negativity across BP components for the LH group. Figure 4-5c displays a marked increase in negativity at the vert ex across BP components for the LH group. Figure 4-5d illustrates minimal hemispheric differences in the left-parietal region between skill. Figure 4-5e illustrates an in crease in right-parietal cortical negativity for the BPearly component (* represents p < .05)..............................................................157 4-6 Performance differences across cortical regions and BP components. Figure 4-6a and Figure 4-6b display marked BP negativ ity across components with minimal differences between hits and misses for left-central and right-central regions respectively. Figure 4-6c illustrates pronounced BPpeak negativity at the vertex. Figure 4-6d and Figure 4-6e illustrate greater BPpeak negativity for left and right parietal regions............................................................................................................... ..159 4-7 Depicts nonsignificant trends in cortical negativity within the three components of the BP for each cortical region for the HA condition as compared to the LA condition across skill level...............................................................................................161

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE ROLE OF THE QUIET-EYE PERIOD A ND THE BEREITSCHAFTSPOTENTIAL IN AROUSAL REGULATION AND MOTOR PREPARATION FOR PERFORMANCE OF A SELF-PACED MOTOR SKILL By Derek T.Y. Mann December 2007 Chair: Christopher M. Janelle Major Department: Health and Human Performance Given the robust empirical support for and prac tical implications of the quiet-eye (QE) period, it was my objective to a ssess the role of the QE peri od in emotion regulation and motor preparation. The concurrent explor ation of the BP and QE period under varying levels of anxiety was designed to assess the principal mechanisms responsible for the psychomotor differences between expert and near-expert performers. Tw enty golfers were classified by their USGA handicap rating as either a high handicap (HH; near-expert) or low handicap (LH; expert) to permit skill-based inferences. Participants co mpleted 45 trials in both low and high anxiety conditions during which cognitive anxiety, somatic anxiety, heart rate, QE duration, BP activity, and putting performance were recorded. Results indicated that the LH golfers are more accurate and less variable in their performance than the HH group, as revealed by measur es of radial error, bi variate variable error, and group centroid radial error. Systematic di fferences in QE duration and BP were also observed, with experts exhibiting a prolonged quiet eye period and greater cortical activation in the right-central region compared to non-experts. A significant association between cortical activation and QE duration was also noted. De spite performing under high and low anxiety

PAGE 12

12 conditions, QE duration and cortical activati on did not fluctuate across conditions. Taken together, the results of this investigation lend prim ary support to the motor programming/motor preparation function of the QE pe riod. Practical and theoretical implications are presented and suggestions for empirical work provided.

PAGE 13

13 CHAPTER 1 INTRODUCTION One of the most challenging putts Ive ever faced was the one I had on the final green of the 1999 Bob Hope Chrysler Classic. It wasnt the length or the break that made it hard, of course. The putt was only about seven feet, with a little tail at the end. It was an eagle putt to win the tournament. And it was for a score of 59, which would be the first sub-60 score anyone on the PGA Tour had ever shot in a final round. I knew that I might never have another chance to set that record. Th e circumstances surrounding the putt challenged my mind. And putting, Ive learned, is all about your mind and your attitude. --David Duvall Anyone who has ever stood over a golf putt with the slightest of importance can easily relate to the anxiety, trepidation, and uncertainty that faced David Duval that day. The game of golf is replete with indelible moments in which players have managed to coordinate both the mind and body through volition to achieve. Ever so prevalent, however, are those missed opportunities in which the athlete succumbs to the performance pressures of arguably the simplest stroke in golf. The golf putt, which accounts for approximately 43 percent of the games strokes (Pelz, 2000) is a simple, self-paced, closed task that requires minimal athleticism. It is perhaps these superficial aspects of the golf putt that yield su ch a quandary for both athl ete and sport scientist alike. Although motorically simple, the difficulty of the golf putt lies in the golfers ability to synchronize sensory information with the mechanis ms necessary to prepare, produce, and control motor behavior (Craig, Delay, Greely, & Lee, 2000; Pfurtschelle r & Neuper, 2003). For example, successful performance mandates that the golfer attend to cues related to distance, direction, and speed; elements that are direct ly influenced by a multitude of environmental conditions (e.g., slope, grain direction). Accordingl y, the visual system mu st attend to the most salient perceptual cues necessa ry to ascertain both distance and direction information, while working memory is called upon for matching str oke tempo with the requisite speed of the impending stroke.

PAGE 14

14 An extensive body of evidence sugg ests that the visual system is the dominant perceptual system by which all other perceptual systems are attuned (Abernethy, 1996; Janelle, Hillman, & Hatfield, 2000; Posner, Nissen, & Klein, 1976; Van Wynsberghe, Noback, & Carola, 1995). The inability to coordinate the visual and motor systems while regulating affective states may confound the mechanical elements of the golf putt, rendering it a difficult and often frustrating task. As such, the ability to attain, master and demonstrate perfor mance proficiency of motorically simple tasks under varying contex tual conditions (e.g., high pressure) can prove demanding even for the most skilled athletes (Singer, 2000), sugges ting that optimizing attentional processes during the preparatory period immediately pr eceding task execution for a self-paced task is of paramount importance (Hillman, Appa ries, Janelle, & Hatfield, 2000). Given that the visual system is the dominant perceptual system, researchers have dedicated considerable effort to addressi ng the visual search characteris tics and gaze behaviors accounting for the attentional factors that preclude the e xpert advantage. Mann, W illiams, Ward, and Janelle (2006) conducted a meta-analysis encompassing n early three decades of work, examining the many performance metrics and indices of attent ional allocation differences of experts and nonexperts. The results provide further support for the role of visual attention in the expert advantage, revealing that expe rts consistently exhibit fewer fixations ( rpb = 0.26) of longer duration ( rpb = 0.23) Such visual search characteristics index an individuals point of interest and relative attention allocation. The longer the eye remains fixate d on a given target, the more information is thought to be extracted from the display, permitting detailed information processing. Additionally, the number of visual fixa tions during a given period provides an index of the search characteristics representative of the most pertinent cues extracted from the environment, thereby facilitating the decision making process. Given the typically dynamic

PAGE 15

15 context of sport, researchers have interpreted visual search strate gies involving fe wer fixations of longer duration as more efficient, permitting more time for more detailed information extraction (Williams, Davids, & Williams, 1999). Researchers (i.e., Janelle, Hillman, & Ha tfield, 2000; Vickers, 1992, 1996a, 1996b) have turned their attention to additional gaze beha vior indices that may reveal expert-novice differences. Of these indices, a promising and robust observation is that experts exhibit an extended quiet-eye period relative to non-e xperts. According to Vickers (1996a), the quiet-eye (QE) is a temporal period when task relevant environmental cues are processed and motor plans are coordinated for the successful completion of an upcoming task. Specifi cally, the QE period is defined as the elapsed time between the last visual fixation to a target and the initiation of the motor response (Vickers, 1996a). As such, the QE appears to functiona lly represent the time needed to organize the neural networks and visu al parameters responsible for the orienting and control of visual atten tion (Vickers, 1996a). Collec tive analysis of the exta nt literature reveals that experts exhibit long er quiet eye periods ( rpb = 0.62) when compared to less skilled performers (Mann et al., 2006). Furthermore, in tragroup variability has also been reported, suggesting that longer quiet eye periods correspond with increased accuracy (Harle & Vickers, 2001; Janelle et al., 2000; Vickers, 1996a, 1996b; Vickers & Adolphe, 1997). Despite its promise, the underlying mechanism( s) responsible for the robust QE findings remain in question. From a pure visuo-motor perspective, the QE may serve to maximize cerebral efficiency, as reflected in cortical patterns indicative of elite performance (Janelle, Hillman, & Hatfield, 2000; Janelle, Hillman, Ap paries, Murray, Meilli, Fallon, & Hatfield, 2000). That is, previous research has consistently reported cortic al quieting in visuospatial and motor coordination tasks in the left hemisphere as compared to the right hemisphere at temporal,

PAGE 16

16 mid-frontal, occipital, and pa rietal regions (e.g., Crews & La nders, 1993; Haufler et al., 2000). Although counter intuitive, a prolonged QE period may be related to this cortical quieting and subsequent notion of expert effi ciency. As previously stated, e xperts in sport generally make fewer fixations of longer durati on, suggesting a level of informa tion processing efficiency that permits more time spent on task relevant cues and less time in search of these cues. As such, a prolonged QE may permit a similar advantage; as task-salient cues are efficiently acquired, less effort is spent on the acquisition and processing of such cues, permitting the re-allocation of cortical resources away from the information pr ocessing stages of performance and toward the motor programming and execution stages. Alternatively, researchers (Janelle, Hillman, & Hatfield, 2000; Vickers et al., 1999) have suggested that the QE period may serve an emo tion regulation function to maintain processing efficiency (Eysenck & Calvo, 1992) and the ef fective use of relevant perceptual cues (Easterbrook, 1959), sparing attentional resource s necessary for task execution. The following section addresses the two potential mechanisms that may moderate the relationship between the QE period and performance, namely sensorimot or integration and emotion regulation. Expertise Related Differences in Cortical Activity Vickers (1996a, 1996b) has relied heavily on ba sic cognitive neuropsyc hological evidence to postulate the cognitive archite cture that underlies the QE pe riod. In doing so, she cited the work of Posner and Raichle (1991), who propos ed a three-component network for visual attention including the or ienting, executive, and vigilance networks. The orienting network provides for shifts in atten tion, while the executiv e network serves to recognize the most pertinent cues relative to goal directed behavi or. The vigilance network, however, serves to maintain focused attention by facilitating the or ienting system and suppr essing the processing of irrelevant stimuli. A residual effect of the vigi lance network may also be the reorganization of

PAGE 17

17 the neural networks responsible for increased visual-spatial pro cessing and the recruitment of the requisite motor program. To better understand the covert psychological indices of the expert advantage, researchers have made extensive use of electroencephalogr aphy (EEG) and spectral an alysis techniques to investigate cortical activation and hemispheric specialization dur ing the preparatory period of self-paced closed motor skills such as golf pu tting (Crews & Landers, 1993), archery (Salazar, Landers, Petruzzello, & Han, 1990), and shooting (Deeny, Hillman, Janelle, & Hatfield, 2003; Hatfield, Landers, & Ray, 1984, 1987; Hillman, Appa ries, Janelle, & Hatfield, 2000; Janelle, Hillman, & Hatfield, 2000). Analysis of EEG spectra l power, in particular, has revealed that the effectiveness and efficiency of expert performance has a cortical signature that differs from that of non-experts (Deeny et al., 2003; Hatfield et al., 1984; Haufler et al., 2000; Janelle, Hillman, & Hatfield, 2000; Landers et al., 1994 ). That is, as individuals progr essively become more skilled, the cognitive strategies employe d during the planning and executi on of movement become more routine, demanding fewer cortical resources (F itts & Posner, 1967; Smith, McEvoy, & Gevins, 1999), resulting in a demonstrable increase in left hemisphere alpha power (i.e., decrease in cortical activity) and performance. The comparison of cortical activation across hemispheres at corresponding reference sites permits an index of hemispheric asymmetry. Within the psychomotor literature, researchers have demonstrated rela tively stable cortical activation across hemispheres in th e novice performer, where as the expert reliably demonstrates a pronounced asymmetrical ratio, char acterized by a relative increase in left hemisphere to right hemisphere alpha power (i.e., decreased cortical activity). Simply stat ed, the novice performer requires greater conscious processing (i.e., verb al analytic processing) of the task demands resulting in greater left hemisphere activati on (Hatfield et al., 1984) Conversely, the expert

PAGE 18

18 performer operates with greater automaticity and su stained visual-spatial processing as indicated by a decrease in the ratio of cortical activation between left and right hemispheres as the time to execution nears. In the first attempt to direc tly relate the QE period to cortical modulations, Janelle, Hillman, and Hatfield, (2000) assessed the psycho motor performance of marksmen. Elite level performance was characterized by significantly longer QE periods and pronounced hemispheric asymmetry, providing the first em pirical account of a relationsh ip between the QE period and cerebral efficiency. According to Vickers (1996 a) conceptual account of Posner and Raichles (1991) three-component network, the QE period may facilitate such co rtical differences. A limitation of EEG spectral activity, however is the restricted inference from the spontaneous rhythmic oscillations in voltage (Fabiani, Gratton, & Coles, 2000) to a specific brain function or psychological process. Furthermor e, the spectral technique decomposes the continuous EEG signal into specif ic frequency bands (i.e., Alpha, 8-12 and Beta, 13-36) to assess the cortical activity a ssociated with a behavioral state, thereby ameliorating its temporal characteristics. As a result of the associated filter ing, cortical excitati on and/or inhibition occurring at subsequent frequency ranges may be attenuated. Although Event Related Desynchronization (ERD) procedures permit th e precision and inference of time-locked analyses, signal processing of th is nature is restricted to th e desynchronization of specific frequency bands (e.g., Alpha, 8-13Hz). Moreover, the electro-cortical investigations of movement preparation and voliti on require a broad frequency ra nge (DC 40Hz), exceeding that of ERD studies. A number of psychophysiological investigations have addresse d athletes attentional and preparatory states preceding task executi on (e.g., Crews & Landers, 1993; Konttinen &

PAGE 19

19 Lyytinen, 1992) by evaluating the electro-cortical modalities of event-related potentials (ERP) that occur across a range of frequency bands. A ccording to Fabiani, Gratton, and Coles (2000), ERPs offer additional insight into the cortical ma nifestations that precede or follow a discrete event. The ERP is derived from the average of multiple responses and represents the temporal relationship of cortical activation to a specific event, thereby providing a time-locked index of the psychological correlates of performance. The bereitschaft spotential (BP), or readiness potential (RP) first described by Kornhuber and Deecke (1965), is one class of ERP that lends itself well to the study of the preparatory period preceding task execution. The BP is a negative potential that is characterized by a distinct cortic al signature that precedes an actual, intended, or imagined event by 1 to 1.5 seconds and indexes anticipatory attention a nd movement preparation (Jahanshahi & Ha llett, 2003). The BP is a visually distinct waveform comp rised of three components, each of which are temporally and cortically dive rse (Deecke, Scheid, & Kornhube r, 1969). The early slow rising negativity (BPearly) reflects the activation of the supplementary motor area (SMA) and begins approximately 1500 ms prior to movement onset. The early activation of the BPearly has a widespread scalp distribution with maximal pot entials recorded at the vertex (Deecke, 1987). According to Roland (1984; Roland, Larsen, Lassen, & Skinhoj, 1980), the SMA serves to retrieve and/or augment the requisite motor commands from memory. Accordingly, the more elaborate the motor sequence, the more prec ise the corresponding movement should be, as indexed by an increase in SMA activation (i.e ., increased negativity). The second component, known as the BPlate, is characterized by a change in the steepness of the waveforms slope, which occurs approximately 400-500 ms prior to movement onset, and is known to reflect activation of the primary motor cortex (MI; Deecke, 1987; Shibasaki et al., 1980). Changes in BPlate have been

PAGE 20

20 shown to reflect skill differences, such that a d ecrease in negativity is evident in the hemisphere ipsilateral to the activ e limb (Taylor, 1978); however the am plitude contralateral to the active limb increases during skilled pe rformance (Chiarenza, Vasile, & Villa, 1990; Papakostopoulos, 1978). Finally, BPpeak, which reflects the coordinated activ ation of the SMA and MI, is most pronounced over the hemisphere c ontralateral to the responding hand and occurs approximately 50-60 ms prior to movement onset. As Brunia and van Boxtel (2000) state, the components of the readiness potential collectively index the initiation of voluntary, self-paced, motor acts. Preparatory activity in the general context of sensorimotor transformations implicates an integrated neural path linking perception to action (Toni & Passingham, 2003). As such, the BP reflects activation of subcortical and cortical ge nerators (cortico-basal ganglia-thalamo-cortical circuitry) necessary not only in motor execution but also in its preparation (Rektor, 2003). The BP has therefore been speculated to play a role in the detection and pair ing of task relevant environmental features with th e requisite elements of respons e execution (Brunia & van Boxtel, 2000). Accordingly, throughout the preparation and movement phases of skill execution, the visual attention centers (i.e., occipital and parietal cortex ) disseminate requisite commands to motor regions of the cortex (i.e., motor cortex, premotor cortex, supplementary motor area, basal ganglia, and cerebellum; Vickers, 1996a, 1996b), all of which are reflected in the BP. Previous research has revealed that the components of the BP are susceptible to modulation under a variety of e nvironmental and task constrai nts. For example, the mean amplitude of the BPlate has been shown to increase with enhanced motivation (Andreassi, 1980) and reported to nearly double in amplitude with the addition of a monetary incentive (McAdam & Seales, 1969). Perhaps most releva nt here, however, is a pronounced BPlate with increased response accuracy (Becker, Iwase, Jurgens, & Korhuber, 1976; McAdam & Rubin, 1971) in

PAGE 21

21 visuo-motor tasks. More recently, sport rese archers have applied the slow negative ERP paradigm to research with golfers (Crews & Landers, 1993), archers (Lan ders et al., 1994) and marksmen (Konttinen & Lyytinen, 1992; Konttinen, Lyytinen, & Era, 1999) revealing that elite performance is characterized by an increase in cortical negativity in the period immediately preceding task performance (BPpeak), a pattern indicative of th e requisite motor program among experts. Conceptually, the QE period is thought to re present the time needed to organize both the neural networks and visual parameters respons ible for the orienting and control of visual attention (Vickers, 1996a, 1996b). Sim ilarly, cortical activation levels are believed to reflect the cerebral efficiency by which the visuo-spatial pa rameters needed for effective performance are organized. According to Nunez (1995), the cortical efficiency noticed among experts may be the result of decreased cortico-co rtical communication, s uggesting the deactiva tion of irrelevant neural pathways and increased attention. Stabili zation of gaze behaviors (i.e., increased QE) in experts coupled with an increase in cortical quieting may behaviorally represent a pruning (Hatfield & Hillman, 2001, p.378) of irrelevant resources, and the re-allocation of cognitive resources to task relevant components. Therefore, the cortical generators responsible for the BP, which have been shown to correspond with the pr eparation and execution of a motor task, may in turn benefit from the re-allocation of resources, allowing for the development of a more refined motor program. Expertise, Visual Search and Emotion Regulation In addition to its motor planning function, rese archers have suggested that the QE period may also reflect a temporal window for the regul ation of emotion (Janelle, Hillman, Apparies, Murray, Meili, Fallon, & Hatfield, 2000; Janelle Hillman, & Hatfield, 2000; Vickers et al., 1999). Given the level of performance uncertain ty that accompanies highly competitive and

PAGE 22

22 challenging tasks, corresponding increases in stre ss, arousal, and anxiety are expected (Jones & Swain, 1992). According to the Processing Effi ciency Theory (PET; Eysenck & Calvo, 1992), the interaction of individual st ate and trait levels of anxiet y coupled with environmental constraints (e.g., performance pre ssure) directly impact the func tional capacity of attention, rendering performance less efficient and potentially less effective. Cognitive anxiety, which is characterized by worry and an inability to c oncentrate, diverts thought s and cognitions away from task relevant cues and preoccupies cognitions with outcome expectations and evaluation (Liebert & Morris, 1967). According to Baddele y (1986), elevated levels of cognitive anxiety reduce the cognitive resources (i.e., working memo ry) available and necessary to sustain task relevant processing. Empirically, Murray and Janelle (2003) empl oyed a dual-task auto racing simulation to demonstrate that the additional cognitive/attent ional demands imparted by relative increases in anxiety result in increased search rate, rende ring the performer less efficient. The notable reduction in processing efficiency may be in pa rt due to a decreased ability to utilize or discriminate between relevant a nd irrelevant cues. That is, cue utilization becomes diminished under conditions of heightened anxi ety and/or arousal, in which stimulus detection becomes less discriminable and information processing beco mes less effective and efficient (Easterbrook, 1959). Though not a direct confirmation of this no tion, Murray and Janelle (2006) reported ERP (P3) findings consistent w ith such an explanation. Recent gaze behavior research in sport (Mu rray & Janelle, 2003; Williams & Elliot, 1999; Williams, Vickers, & Rodrigues, 2002) offers em pirical support for the theoretical tenets put forth by Eysenck and Calvo (1992) and Easterbr ook (1959). For example, Williams and Elliot (1999) examined the effects of cognitive anxiet y and level of experien ce on anticipation and

PAGE 23

23 visual search behavior in karate kumite. Th e viewing patterns of both groups (Expert and Novice) were altered under the high anxiety condition as compared to the low anxiety condition, with a marked increase in atten tion to peripheral cues. Furthermor e, search rate increased along with greater anxiety among novices More specifically, experts in the high anxiety condition demonstrated an increase in the mean duration of their fixations while the mean fixation duration of the novice group decreased. The corresponding in crease in search activ ity in the novice group can be explained by a comparative decrease in processing efficiency and ineffective cue utilization (Williams & Elliot, 1999). In a related study, Janelle, Singer, and Williams (1999) examined changes in gaze behavior under varying levels of anxiety during a simulated racing task. Resu lts indicated that as anxiety increased, processing efficiency a nd task performance decreased, while the corresponding gaze behaviors became more variable (i.e., more fixations of shorter duration). Janelle et al. (1999) c oncluded that anxiety increases atte ntional narrowing, resulting in the ineffective search and use of cues. Furthermor e, Williams et al. (2002) assessed table-tennis performance under combinations of high and lo w working memory with corresponding changes in anxiety (high and low). As expected, results indicated increased effort, delayed reaction times, and increased search rates while performing unde r the high working memory and high anxiety condition as compared to the low working me mory and low anxiety condition, a pattern indicative of less efficiency. As mentioned, Murra y and Janelle (2003) used a dual-task paradigm to assess the effects of increased anxiety on a si mulated driving task. Consistent with previous research (Janelle et al., 1999; Williams & Elliot, 1999; Williams et al., 2002), driving performance decreased as anxiety increased conc omitantly with search rate, indicative of a decrease in processing efficiency.

PAGE 24

24 Each of the aforementioned inve stigations lends support to the negative effects of anxiety on performance, processing efficiency and cu e utilization. As mentioned, researchers have suggested that the QE period ma y reflect regulation of emotiona l states (Janelle et al., 2000; Vickers et al., 1999) so as to alleviate these detrimental effects. That is, the prolonged QE duration that is characteristic of experts may pe rmit them to preclude processing of irrelevant stimuli, thereby circumventing the deleterious effects of anxiety and/or arousal by permitting the recruitment of task specific res ources. For example, Vickers et al (1999) examined the effects of cognitive stress and physiological arousal on the gaze behavior and shooting accuracy of elite biathletes. Although the QE was influenced by modul ations in cognitive st ress and physiological arousal, QE durations during elite performance were similar across levels of cognitive stress and physiological arousal. Furthermore, Williams, Si nger, and Frehlich (2002) assessed the gaze behaviors of skilled and less sk illed billiard players. They suggested that increased task complexity necessitates increased resources and preparation, and postulated that if the QE is related to cognitive processing a direct relationship between the two should be evident (Williams et al., 2002). Results demonstr ated that as task complex ity increased, so too did the corresponding QE period. Expert-novice differences were also evident. Specifically, experts continued to elicit longer QE peri ods as compared to their novi ce counterparts, and QE duration was proportionally longer on successful shots than on unsuccessful shots across skill levels, suggesting that the QE period may in fact aid in the circumvention of cognitive constraints (Janelle et al., 2000; Vi ckers et al., 1999). Limitations Although the seminal work of Vickers (1996a) sparked a number of studies corroborating the notion that an extended QE period characterizes experts, none of these investigations have assessed alternative theoretical postulates for the underlying ps ychological processes accounting

PAGE 25

25 for such differences. As such, the primary mo tivation for the current investigation was to determine which of the competing accounts of th e QE is the most viable for understanding the robust QE findings reported to date. Notwithstanding the cortical pr ocessing advances originati ng from the early work of Hatfield et al. (1984) delineat ing levels of expertise, electroencephalographic measures of interest have been primarily spectral. Although the use of ERD and coherence analyses have provided additional insight into the expert adva ntage, these differences remain topographical, failing to provide theoretical accounts of th e psychological processe s under investigation (Lawton, Hung, Saarela, & Hatf ield, 1998). Although Crews and Landers (1993) and Konttinen, Lyytinen, and Konttinen (1995) have examined th e slow potential differences across levels of performance, the psychological variables contri buting to cortical differences were primarily ignored in favor of motoric differences. Howe ver, according to Jahashahi and Hallett (2003) there are many psychological variables that can modulate the latency and amplitude of the BP. Therefore, one must consider the preparatory stat e of the individual (i.e., anxiety level) and the impending cortical adaptations (i.e,. changes in latency and amplitude of the BP) necessary to perform. Finally with the exception of the work of Janelle, Hillman, and Hatfield (2000) and Janelle et al. (2000), the concurrent assessment of ocular and cortical indices has not occurred. Simultaneously recording the QE and the corres ponding electro-cortical activity may shed light onto the visuo-motor processes differentiating sk ill levels. Although, cerebral efficiency has been linked with prolonged QE periods these findings warrant repli cation and extension. Given the preparatory implications of the QE as well as the preparatory impli cations of the BP, a

PAGE 26

26 concurrent investigation of these variables may provide greater insight into the underlying psychological processes of both phenomena. In summary, researchers have provided evidence, although infe rential, in support of the potential mechanisms of superior visuo-motor pe rformance as evidenced by the QE. However, to date, researchers have failed to offer a concrete theoretical or mechanis tic account of the QE. As such, the following investigation attempts to explicitly assess two mechanisms, motor programming and/or emotion regulation that ma y account for the robust relationship between QE duration and performance. Statement of the Problem Sport psychophysiologists interested in expe rtise have provided evidence for a unified theme of psychological efficienc y, offering descriptions from th e cortical and visual search domains (Hatfield, Hillman, Appa ries, Janelle, & Vickers, 1999) The extant body of literature permits the conclusion that expert performance is characterized by regional cortical deactivation (i.e., increased alpha power) coupled with efficien t cue utilization (i.e., fe wer fixations of longer duration and prolonged QE). Although a relations hip between the QE period and cerebral efficiency has been reported (Janelle, Hillma n, Apparies, Murray, Meili, Fallon, & Hatfield, 2000), the extent to which such a relationship is evident is tenuous. Therefore, of primary interest in this investigation is the assessment of the underlying mechanisms of the QE period responsible for the psychomotor s uperiority of the expert (i.e., emotion regulation and/or motor planning). The Current Study To assess the extent to which the proposed mechanisms correspond with modulations of the QE period, the BP, which has been proposed as an electrophysiological index of cerebral efficiency during the premotor period (Papa kostopoulos, 1978), was assessed under high and low

PAGE 27

27 anxiety conditions. The preparator y set of the performer, which has been indexed by changes in QE duration, may be equally evident in cortical changes leading up to the point of movement execution. Pursuant to this goal, expert (low handi cap; LH) and near-exper t (high handicap; HH) golfers were observed for putting accuracy under high and low anxiety conditions to assess the extent to which the QE period modulates the form ulation of a motor-program and/or the attention consuming effects of anxiety. Several performa nce variables (i.e., put ting accuracy, RE, MRE, and SRE) were examined to determine if th e quiet-period changes unde r varying levels of anxiety and if so to what extent Moreover, the extent to which th ese changes are reflected in the cortical measure of the BP were assessed. Hypotheses The following eight hypotheses were tested. The hypotheses address expected BP and QE differences between skill levels as well as the predicted rela tionship between the BP and QE under normal conditions. Furthermore, several hypot heses were put forth to address the impact of the arousal-anxiety manipul ation on performance, BP, and QE of HH and LH golfers. 1. The use of a monetary incentive, video camera, and written release would produce elevated cognitive and somatic anxiety levels in the hi gh anxiety condition as compared to the low anxiety condition as measured by the Mental Readiness Form Likert (MRF-L, Krane, 1994) and heart rate (BPM, BIOPAC Systems, In c, Santa Barbara, CA). The use of such manipulations has been demonstrated to be a valid means for invoking an anxiety response across a variety of tasks (Hardy, Mullen, & Jones, 1996; Janelle, 1997, 2002; Murray & Janelle, 2003). 2. Across both the low and high a nxiety conditions, the LH group would perform better (i.e., greater success/failure ratio, and reduced variab ility on missed putts) and exhibit a longer QE period than the HH group. Previous research ha s demonstrated consistent and robust QE differences between skill levels in both ope n and closed tasks (Har le & Vickers, 2001; Janelle, Hillman, Apparies, Murray, Meili, Fallon, & Hatfield, 2000; Mann et al., 2006; Vickers, 1992, 1996a, 1996b; Adolphe, Vickers, & Laplante, 1998; Vickers & Adolphe, 1997).

PAGE 28

28 3. Variability in the duration of the QE would account for both inter and intra-group performance variability. More sp ecifically, the LH group would not only exhibit a longer QE duration as compared to the HH group, but th e QE duration of both LH and HH groups for successful putts would exceed the QE duration for missed putts. Research with golfers (Vickers, 1992), marksmen (Janelle et al 2000), basketball players (Vickers, 1996b), biathletes (Vickers et al., 1999), and volleyba ll players (Vickers, 1997, 1998) indicates that experts not only demonstrate a longer QE when compared to less skilled performers, but within group performance diffe rences are also notable. 4. Given that the BP is representative of the cortical mechanisms responsible for movement preparation and that increased negativity in the mean BPpeak and mean BPlate amplitude characterizes greater movement preparation a nd cerebral efficiency (Chiarenza, Vasile, & Villa, 1990; Papakostopoulos, 1978; Taylor, 1978) it was expected that the LH group would exhibit a greater BPlate amplitude coupled with a greater BPpeak amplitude compared to the HH group. 5. The mean amplitude of the BPlate and the mean amplitude of the BPpeak were predicted to discriminate between putts made and putts missed regardless of skill level. It has been suggested that the BP deve lops during the premotor period, a temporal period that corresponds with decreased heart rate and electromyographic activity that is both steady and tonic (Chiarenza et al., 1990). In accord with Lacey and L aceys (1978) intake-rejection hypothesis, research assessing heart rate pattern ing and performance ha s reported that heart rate deceleration during the preparatory period fa cilitates sensorimotor efficiency, such that an orienting of attention to task relevant cues is permitted, thereby yielding performance increases (Boutcher & Zinsser, 1990; Crews, 1989; Hatfield & Hillman, 2001; Molander & Backman, 1989; Tremayne & Barry, 1990, 2001). Si milar to the sensorimotor efficiency denoted by heart rate decelerat ion, the amplitude of the BPpeak is more pronounced with greater task involvement (McCallum, 1976) a nd is proportional to the preparatory set (Loveless & Sanford, 1974). Appropria tely, an increase in mean BPearly amplitude and mean BPlate amplitude, coupled with an increase in BPpeak amplitude, would be evident for putts made as compared to putts missed. 6. According to Eysenck and Calvo (1992), an in crease in anxiety is likely to result in a decrease in processing efficiency, stemming from the additional cognitive/attentional resources needed to perform a desired task. Furthermore, Eysenck (1982) postulates that, although anxiety may elicit a negative affective state, such increases in anxiety may also serve to increase motivation. As such, it is plau sible that as environm ental task constraints increase perceived anxiety, the corresponding appraisal of task difficulty may also increase. Therefore, I hypothesized that increased negati vity for each component of the BP (i.e., BPearly, BPlate, and BPpeak) amplitude would be greater in the high anxiety condition as compared to the low anxiety condition due to a relative increase in the complexity of the task (Lang et al., 1989) and the increased mobiliz ation of resources (i.e., motivation) to successfully complete the task. 7. The QE period is believed to index the time n eeded to organize the cognitive and visual components associated with the orienting of vi sual attention and execu tion of a self-paced task. Cortical and sub-cortical generators associated with the BP are also responsible for the

PAGE 29

29 preparation and execution of a self-paced tas k. As such, a significant and positive correlation between BP and QE period was predicted. More specifically, I expected that as the duration of the QE period increases, relative increases in the amplitude of the BPpeak would be evident. Furthermore, as anxiety levels increase, a co rresponding increase in both QE duration and BPcomponents were expected across skill levels. 8. Increases in anxiety result in a corresponding increase in cognitive demand, as evidenced by attentional narrowing (Eysenck & Calvo, 1992) a nd the inefficient use of perceptual cues (Easterbrook, 1959). Therefore, I hypothesi zed that both the LH and HH groups would exhibit a longer QE period in the high anxi ety condition as compared to the low anxiety condition. That is, an increase in the QE was hypothesized to circum vent the deleterious effects of anxiety, preventing a decrement in performance (Vickers et al., 1999). As such, a corresponding increase in QE duration would perm it the allocation of atte ntional resources to the information processing of the visual cues previously attended to, while suppressing the processing of competing stimuli, ther eby facilitating performance. Although each of the eight aforementioned hypothe ses were designed to address the covert psychophysiological differences betw een expert and near-expert perf ormers, the ability to isolate the role of the QE duration during a self-paced task is pr incipal to all others. Given the extent to which QE differences have been reported in th e extant literature, expe rtise differences were expected. However, the degree to which the sk ill-based QE differences and BP changes are related offers both pragmatic and theoretical su pport for the role of pre-performance vision previously unaccounted for. Previous research supports the contention th at changes in quietduration are directly related to interand intra-group variab ility, including under high stress conditions (Williams, et al., 2002), lending supp ort to the arousal regulation hypothesis. However, the degree to which the QE period is believed to serve a motor programming function is speculative. Given that the BP is responsible for motor planning and pairing of environmental cues to response specifications (Brunia & van Boxtel, 2000), confirmation of a relationship between the QE period and the various BP co mponents would lend empirical support to the motor programming function of the QE period.

PAGE 30

30 Definition of Terms Anxiety (State). State anxiety is defined as the subj ective feelings of tension, apprehension, nervousness, and worry, coupled with the ac tivation of the autonomic nervous system (Spielberger, 1983). Furthermore, a nxiety is believed to be comprised of cognitive (i.e., thoughts and worries) and somatic (i.e., perception of physio logical arousal such as increased heart rate) components (Liebert & Morris, 1967). Bereitschaftspotential (BP). The BP, first described by Kornhuber and Deecke (1964), is a negative cortical potential that develops approximately 1-1.5 seconds prior to the onset of a selfpaced task, suggesting that the BP has both motor preparation and attention components. Closed Task. This is a skill that is performed in a stable or predictable environment in which the participant is in control of movement onset (Magill, 1998). Cortical Generators. Cortical generators are known as th e distinct regions of the cortex (e.g., Supplemental Motor Cortex, Moto r Cortex) known to have topogr aphical representation of the Bereitschaftspotential (e.g., Fr ontal, Parietal, Temporal). Electroencephalography (EEG). The EEG is a process for recording the electrical potential at the scalp associated with the cortical activity of th e underlying structures. Event Related Potential (ERP). The ERP is a time-locked record ing of cortical activation in accord with the International 10-20 system (Jas per, 1958). More specifically, the ERP is the result of an averaging of samples recorded fr om a continuous EEG that is time-locked to a specific event. As such, the averaging technique draws on the signal-to-noise ratio, such that the components of the waveform not deemed related to a specific event are assumed to randomly vary across samples, thereby rendering only a representation of systematic variation or components of the ERP (i.e., slope, amplitude; Fabiani et al., 2000). Eye-Movement Registration. Eye-movement registration refers to the process of recording the visual search characteristics du ring a perceptual-cognitive, or pe rceptual-motor task. The most common procedure is the corneal-reflection me thod, which records the movement of the eye under the assumption that the central portion of the cornea corresponds with a point of visual interest (Duchowski, 2002; Williams, Davids, & Williams, 1999). Fixations. Fixations are characterized by the tiny eye m ovements associated with tremors, drifts, and microsaccades (Duchowski, 2002) that are n ecessary to stabilize the fovea on a specific target, enabling comprehensive stimulus extractio n and information-processing (Williams et al., 1999). Gaze Behaviors. A gaze is the absolute position of th e eyes in space and depends on both the eye position in orbit and th e head position in space (Sch mid & Zambarbieri, 1991, p.229). Therefore, a gaze behavior refers to a specific coordinated action of the eyes and head, which include a saccade, fixation, smooth-pursuit, vesti bular ocular reflex, and the QE period.

PAGE 31

31 Perceptual-Cognitive Skill. The ability to identify and acqui re environmental information for integration with existin g knowledge such that appropriate responses may be selected and executed (Marteniuk, 1976). Quiet-Eye Period. The QE period, a component of the gaze behavior, is a measure defined as the elapsed time between the last visual fixation to a target and the initiati on of the motor response (Vickers, 1996a). Self-Paced Task. This is a skill that is performed free from external temporal constraints or prompts, while under the volition of the participant. Spectral Analysis. Spectral analysis is the process of separating an EEG time series into its constituent frequencies (e.g., Alpha 8-12Hz, Beta 13-36Hz) by subjecting the raw EEG data to Fast Fourier Transform (FFT). Spectral Power. The decomposition of the cortical leve ls expressed by means of a spectral analysis is inversely related to activation (Pfurstcheller 1992). That is, spectral power refers to the decrease in activation otherwis e known as the cortical quieting of the structures related to a perceptual-motor task. For example, an increas e in left hemisphere Alpha power (8-13Hz), suggests a decrease in the amount of activation that is experienced. Sub-Cortical Generators. Sub-cortical generators represent th e deep structures of the brain (e.g., Basal Ganglia, Thalamus, Cerebellum) believed to significantly contribute to the development of the motor program and in turn the cortical structures known to have a topographical representation of the Bereitschaftspoten tial (e.g., Frontal, Pa rietal, Temporal). Stimpmeter. The stimpmeter, designed by Edward S. Stimpson and revised and implemented by the USGA in 1978 (USGA, 2006), is a device used to quantify the speed of a putting surface. The stimpmeter is a V-shaped aluminum bar meas uring 36in long and 1/2in wide that is designed to channel the golf ball and reduce unnecessary lateral movement as the ball is released. Positioned 6 inches from the top of the stimpmeter is a precisely milled ball-release notch designed to discharge a ball when the stimpmeter is raised ap proximately 20 degrees from the ground. The bottom of the stimpmeter is leveled to facilitate the desired a ngle of release. Three balls should be released independently from the same starting point and measured for the distance each ball travels from the end of the stimpmeter. The average distance traveled is referenced as the speed of the green. For example, if the average distance traveled of three golf balls released from the same position is 9.5f t then the speed of the green is a 9.5. Visual Search. A two-stage process conducted to identi fy relevant cues in which a memory dependent systematic search (i.e., serial se arch) or a memory-less random search of the environment is coupled with a decision proce ss confirming relevant s timulus identification. Volition. Volition refers to the conscious will to ac t; a voluntary act that is endogenously driven and free of externally imposed restrictions (Libet, 2003). Assumptions The current investigation was conducte d under the following assumptions:

PAGE 32

32 1. The Mental Readiness Form Likert (Krane, 1994) is an appropriate measure for assessing levels of cognitive and somatic anxiety prior to and during the course of the experimental session. 2. The use of a platform covered with a nylon NP 50 artificial putting su rface (Synthetic Turf International, STI, Jupiter, FL) outfitted with a 4.25in regulation size golf hole adequately simulates an actual putting surface, thereby promoting ecological validity. 3. The putting task (i.e., 12 ft flat putt) and the corresponding dependent measure of putting accuracy (i.e., made vs. missed putts, bias, and consistency) are appropriate measures of performance. 4. Vertical and horizontal electro-oculogram meas ures used to derive QE duration are both appropriate and accurate. 5. The extensor carpi ulnaris (ECU) of the right ar m is an appropriate muscle to demarcate the fiducial time point necessary for the o ff-line reduction and analysis of the Bereitschaftspotential and the co rresponding cessation of the QE period. Significance of the Study The importance of understanding the complex in tegration of systems associated with expert performance is essential for the advan cement of both theory and practice. To date, technological and methodological advances have pe rmitted a rapid increase in research related to expert performance, which has lead to a grea ter conceptual understa nding of the expert performer while executing self-p aced, closed tasks. The work of Hatfield et al. (1984) precipitated an increase in theory driven research examining the information processing, cortical adaptations, and central nervous system differen ces of experts performing sensorimotor tasks. However, it was not until rece ntly that researchers have begun to examine the various mechanisms of perceptual-cognitive expertise in an integrative manner (e.g., Hillman et al., 2000; Janelle, Hillman, & Hatfield., 2000). Of primary importance are theoretical and mechanistic developments stemming for the use of electroencephalographic and visual search technology within an expert-nov ice paradigm. For example, the use of continuous EEG and spectral analysis have reliably demonstrated the role of the alpha band, an d to a lesser extent the

PAGE 33

33 role of the beta I band in the preparatory period of marksmen, golfers, and archers, indicating the virtual automaticity or cerebral efficiency of the expert as denoted by a rela tive increase in alpha power in the left temporal region. Furthermore, the recent integration of eye movement registration techniques, including the QE period, has offered yet another piece of evidence in support of the expert advantage. However, the spec ific role that the quie t-period serves is still relatively unknown. That is, although performance has been demonstrated to vary in relation to the length of the QE period, with prolonged periods generally resulting in increased performance, uncertainty and speculation remain as to whether the QE period serves to facilitate the development and implementation of a requisite motor program or if the QE period serves to ameliorate the harmful performance effects of decreased efficiency. As such, the significance of this investiga tion is founded on the theo retical implications and methodological innovations put forth. Specifi cally, my objective was to determine whether differences in cortical activity and QE characte ristics could differentiate the expert and nearexpert golfer, so as to clarify th e role of the QE period in expert performance. I have proposed an innovative approach for the study of expert performance, incorporat ing a variety of variables that have individually and concomitantly differentia ted the expert and novice performer. Moreover, the exploration of the relationship between the BP and the QE would further enable researchers to understand the integrative role of the percep tual-cognitive, motor, and emotional systems in the production of expert performan ce, while advancing the understand ing of the utility of the BP in goal-directed and ecologically valid research. Lastly, the objective of this research was to extend the current state of the psychophysiolo gical domain in sport by linking time-locked events during the preparat ory period to distinct cortical signa tures during a sensorimotor task.

PAGE 34

34 CHAPTER 2 REVIEW OF LITERATURE Capturing Expertise The performance of experts ac ross domains has intrigued scientists for centuries. More recently, scientific enquiry has attempted to pu t forth a theoretical framework from which the study of expertise can be undertaken in an attempt to deduce the mechanisms that set apart the expert from the novice. Since the principle work of Chase and Simon (1973), researchers have demonstrated that expertise is characterized by an extensive knowledge base that facilitates both stimulus recognition and subsequent procedur al execution (Richman, Gobet, Stazewski, & Simon, 1996). To effectively capture the essence of the expert, the many traits comprising such individuals must be understood. As such, the systematic obser vation and manipulation of both the overt and covert psychological processes of expert and novice performers can provide detailed insight into those mechanisms separatin g the expert from the novice. To begin, a clear operational definition of what consti tutes an expert is warranted. Definition of Expertise Sport expertise has been defined as the ability to consistently demonstrate superior athletic performance (Janelle & Hillman, 2003; Starkes, 1993). Although performance competency and mastery are requisite component s of expertise, unquestionably other equally prominent components exist. Siedentop and Eldar (1989) concur that a primary requisite of expertise is the ability to demonstrate technical proficiency that is reliable and consistent. However, they also emphasize the importance of the athletes aptitude to modify performance skills to meet varying contextual conditions. More prec isely, experts possess an extensiv e procedural and declarative knowledge base (McPherson, 1993,1994, 1999, 2000; Fren ch, Spurgeon, & Nevett, 1995; French & Thomas, 1987) that enables them to extra polate important information from relevant

PAGE 35

35 performance cues in order to anticipate and predict future ev ents (Ericcson, Krampe, & TeschRmer, 1993). Experts appear to maintain an uncanny ability to recall past performance information providing for accurate decisions of what to do, while demonstrating skill superiority (i.e., how to do it). However, al though the expert may possess a rich knowledge base, processing during the task ap pears to operate automatically free from cognitive constraints (Anderson, 1982). Moreover, other researchers have defined expe rtise as a superior task specific problem solving ability resulting from an extensive arra y of long-term memories that enhance pattern recognition (Holyoak, 1991). In comparison, Ennis (1994) suggest s that experts possess and maintain a diverse repertoire of skills and strategies that can be employed within specific situations and contextual requi rements. Elite level coaches ha ve confirmed laboratory findings, stating that the discrimination of expert from novice athletes can be de monstrated along four dimensions, anticipation (i.e., a ttention to and interpretation of relative performance cues), declarative knowledge (i.e., knowledge of tactical and rule base d information), self-knowledge (i.e., sense of own strengths and limitations), and greater cognitive latitude (i.e., quicker and more diverse problem solvi ng) (Lyoka & Bressan, 1998). Furthermore, technological and methodologica l advances permitting the assessment of the covert processes associated with performa nce have contributed to a more complete understanding of the expert advantage. For exampl e, the measurement of electrocortic al activity (i.e., EEG), heart rate (HR), and gaze behavi or, have revealed pronounced hemispheric alpha asymmetry, cardiac deceleration, and longer QE pe riods respectively, each of which has been independently associated with superior perf ormance (e.g., Hatfield, et al., 1984; Tremayne & Barry, 2001; Vickers, 1996a).

PAGE 36

36 In sum, research examining the many underl ying attributes of expertise, including perceptual, cognitive, and motor behavioral domai ns, has generally concluded that experts are more efficient, effective, and accurate in recognizi ng task specific patterns, are more proficient at making decisions, maintain superior procedural and declarative information, have a profound reservoir of contextual cues that are system atically retrievable, and possess an unparalleled ability to foreshadow events and outc omes (Holyoak, 1991; Starkes & Allard, 1993). Expert Performance Experts Are Faster and More Accurate at Recognizing Patterns Skilled performers posses a superior ability to recognize and recall structured patterns during performance within a specific domain as compared to their less skilled counterparts as a result of their advanced knowledge structur es that permit the chunki ng of small bits of information into much larger and more meaningful constituents. Simply, the ability to recognize and recall sport-specific cues is fundamental to me mory structures in whic h previously presented information is encoded into short-term memory and stored in a retrieva ble form in long-term memory (Woo Sohn & Doane, 2003). It is known th at experts acquire knowledge and skills to rapidly program information in long-term memory to facilitate the re trieval of performance information with efficiently processed contextual cues via long-term working memory (Ericsson & Delaney, 1999). The seminal work of deGroot (1978) with chess masters demonstrated th e ability of expert performers to rapidly perceive and store domai n specific patterns into memory. Specifically, chess players were exposed to a board configuration for brief inte rvals. Following exposure they were requested to recite the location of each piece from memory. de Groot (1978) found that chess masters were near flawless in their r ecall, however players of lesser caliber were significantly less effective. Chase and Simon (197 3) extended the work of de Groot (1978) by

PAGE 37

37 implementing a new condition. In addition to being tested on the recall of a typical game board, participants were exposed to a game board with pieces unsystematically arranged. Although Chase and Simon (1973) replicated the initial fi ndings of de Groot, they also found that the pattern recognition abilitie s of the expert group we re context specific. To elaborate, when the chess pieces were arranged in a fashion comparab le to an actual chess match, experts recalled the pieces with amazing accuracy. However, when the pieces were randomly placed, performance of the experts declined dramatically, resembling the recall accuracy of the lesser skilled group. Similarly in sport, researchers have made use of r ecall paradigms, in which participants are exposed to domain specific slides depicting the strategic positioning of pl ayers in offensive or defensive formations. For example, Allard, Grah am, and Paarsalu (1980) assessed the speed and accuracy of pattern reco gnition in basketball play ers and non-players. Participants were exposed to a series of slides that included structured (i.e., offensive formation) and unstructured (i.e., rebound) game situations for brief intervals of four seconds. Following e xposure, participants demonstrated recall ability by identifying player locations on a reduced scale magnetic board. Allard et al. (1980) confirmed the findings of de Groot (1978) and Chase and Simon (1973), revealing that experts were more adept at recal ling structured game knowl edge compared to the non-player group, recognizing critical strategic patterns that formulate during competition. Modifying and extending the work of Allard et al. (1980), Starke s (1987) assessed the expert knowledge structures and pattern recogni tion abilities of elite, skilled, and novice field hockey players. Similar to the Allard et al. (1980) study, partic ipants viewed structured and unstructured slides depicting offensive attack s and turnovers. However, to accommodate the complexity of the field hockey pitch (11-on-11), viewing time was increased from four seconds to eight. Once again, response accuracy and reca ll ability was measured by identifying player

PAGE 38

38 location on a reduced scale magnetic board. The results demonstrated the superior recall abilities of experts, extending the claim that experts possess domain specific pa ttern recognition and recall abilities to the sport co ntext. More recently, research with snooker players and soccer players has corroborated these findings (Abe rnethy, Neal & Koning, 1994; Williams, Davids, Burwitz, & Williams, 1993; Williams & Davids 1995). The notion that long-term working memory is a requisite skill to meet the part icular information-processing demands of domain specific expertise (Ericsson & Delaney, 1999) suppor ts these empirical findings while offering a theoretical account for the expert pattern recognition advantage. Experts Have Superior Procedural, Decl arative, and Strategic Knowledge Declarative knowledge is compos ed of factual information re garding concepts and their interrelationships(Ennis, 1994, p. 167) including rules and definitions (Thomas, 1994). In comparison, procedural knowledge is knowledge a bout how to perform or use the information [necessary to perform] (Ennis, pg. 167). Consid er that the link betwee n a situation and the appropriate action is known as a procedure. Therefore, proced ural knowledge can be further clarified by matching the appropriate movement gi ven that a specific situ ation is coupled with the correct motor execution. As su ch, response selection and motor execution are matched to the stimulus (Thomas, 1994). Researchers interested in determining the extent to which declarative knowledge is responsible for distinguishing expe rt athletes from lesser skilled performers began by adopting a cognitive research paradigm (see Chi, Feltovich, & Glaser, 1981) that required participants to categorize and sort pictures of domain specific stimuli into relevant categories. For example, Allard and Burnett (1985) assessed female bask etball players from the Canadian National Team and novice players for their ability to make associations among the contextual cues provided within each picture. Participants were assessed on the extent (i.e., complex vs. simple) to which

PAGE 39

39 basketball relevant categories were clustered (i.e., strategic form ations vs. jump shots) and the time taken to form the clusters. The expert athl etes systematically divided the pictures into divergent clusters for shots, offensive form ations, rebounds, and defensive formations. In contrast, the novice participants were less stru ctured in the organization of the pictures, identifying only two groups: one-on-one and all ot her formations combined. Allard and Bennett (1985) concluded that experts esta blished higher order knowledge stru ctures that translated into distinct groups based on basketba ll principles, contrary to the s uperficial structures portrayed by the novice group. Although it has been argued that novices, in a ddition to experts, maintain a substantial declarative knowledge base, research has demonstr ated unequivocal differe nces in declarative knowledge between experts and novices (F rench & Thomas, 1987; Thomas, Thomas, & Gallagher, 1993). Research with basketball players further corrobor ates expert-novice knowledge differences. French and Thomas ( 1987) studied the influence of knowledge, both procedural and declarative, on decision-making ability of high and low skill level. Results revealed that the high skill group not only performe d the skills at a higher level, but they also possessed advanced basketball knowledge structures. As another example, Williams and Davids ( 1995) tested soccer players of varying ability (i.e., high-skill, low-skill, and physically di sabled) on a soccer r ecall, recognition, and anticipation ability task. The fi ndings revealed that the highly-skilled players demonstrated superior anticipation, recall, and recognition as co mpared to the low-skill and physically disabled groups, supporting the notion that high level performers maintain a larger and more elaborate declarative knowledge base. Although decl arative knowledge alone cannot account for performance differences in a domain that require s the physical execution of a task, it is believed

PAGE 40

40 to be a fundamental component of skill rather than a conseque nce of experience (Williams & Davids, 1995). Conversely, procedural knowledge, which is more pronounced in expert performers (French & Thomas, 1987; McPherson, 1993), may bette r be able to discri minate expert from novice performers. Research with high and low handi cap golfers, revealed that the high handicap golfers demonstrated less knowledge about how to perform their skills as compared to the low handicap golfers (Thomas & Lee, 1991). This find ing suggests that the lo w handicap golfers not only knew what to do, but they knew how to do it. The coupling of declarative and procedural knowledge arguably translates in to decreased reaction times and subsequently increased decision-making speed and accuracy (A bernethy, Thomas, & Thomas, 1993). Experts Are Superior at Antici pating Opponents Actions The use of advance visual cues has been dem onstrated to facilitate sport performance by means of anticipating opponents actions a nd decreasing overall response time. More specifically, elite performers have been shown to consistently use advance cues otherwise overlooked by less skilled performers (Williams & Davids, 1998; Williams et al., 1999). Contemporary research in sport has examined skill-based differences in a diverse array of sporting contexts (e.g., badminton, tennis, soccer, et c) using a variety of empirical paradigms to confirm the use of advance cues by highly skille d athletes translating into decreased response time. From an information-processing perspec tive, Buckolz, Prapaves is, and Fairs (1998) put forth a conceptual framework depicting the advant ages of advance cue use in sport (Figure 2-1). Specifically, Buckolz et al. (1998) contend that the effective us e of advance perceptual cues serves to alleviate performance restrictions impos ed by temporal constraint s. That is, a priori information (Figure 2-1, A) derived from either contextual cues (e.g., opponents strength and weaknesses, environmental conditions, and curr ent match context) and/or body language cues

PAGE 41

41 from of ones opponent (e.g., stance, racket pos ition, speed, body position etc) can provide critical information necessary to foretell future actions. As a result, outcome expectancies are developed such that the performer engages in ei ther selective prepara tion (Figure 2-1, J) or anticipatory mobilization (Figure 2-1, H). Therefore, anticipatory mobilization, a product of seeking and using information from advance per ceptual cues can result in the elimination of reaction time and the subsequent reduction of movement time by simply permitting the commencement of a response sequence prior to the completion of an opponents action sequence. Alternatively, selec tive preparation can re duce reaction time by matching a stimulus cue with a desirable response. For example, prepar ing to return a flat se rve hit down the center line that plays out as anticipate d; resulting in a pairing of the stimulus and the response. The ability of expert performers to extract perceptual cues can a lleviate the temporal constraints imposed by reaction time. Extensive declarative knowledge can be used to formulate a priori scan paths to facilitate anticipation permitting extended movement times otherwise restricted. Extensive empirical work has repeatedly demonstrated quicker response times favoring this notion of advance cue use. For example, He lsen and Pauwels (1992) presented penalty kicks, 2-on-2, and 3-on-3 video clips to experienced and inexperienced soccer athletes. Participants were required to physically res pond to the scenarios presented on the film by executing a shot on goal or a pass to a teammate. The findings de monstrated that the experienced players were quicker and more accurate in their responses. In an attempt to extend the work of Helsen and Pauwels (1992), Williams, Davids, Burwitz, and Williams (1994) investigated skil l-based differences in anticipation using experienced and inexperienced soccer players in 11-on-11 soccer situations. Participants in this investigation responded verbally as quickly as po ssible to the final pass destination. Consistent

PAGE 42

42 with previous research, the e xperienced players demonstrated superior anticipatory skills. Furthermore, Williams and Davids (1998) examined differences in anticipation and visual search strategies in 3-on-3 and 1-on-1 soccer situa tions. Twelve experienced and 12 inexperienced soccer players were presented with 20 offensive soccer sequences and asked to anticipate final pass location. The results demonstrated that th e experienced players were more adept at anticipating final pass destinati on and did so more quickly as compared to the less skilled participants. Additionally, when aspects of the visual environment were occluded the performance of the advanced group was significan tly hindered while the le ss skilled participants performance remained unaffected, further suggesti ng that advanced visual cues are requisite components of experienced performers swift and accurate decision-making. Recent studies in baseball perceptual d ecision-making have supported expert novice differences. Radlo, Janelle, Barba, and Frehlich (2001) compared groups of baseball players (i.e., varsity players and college stude nts) on a baseball pitch discri mination task that required the participants to identify the type of pitch seen (fastball or curveball) as quickly as possible by pressing one of two buttons. The findings demonstr ated that elite performers were quicker and more accurate at identifying the type of pitc h, supporting the notion that advanced cues, in addition to extensive knowledge structures fac ilitate accurate and expeditious decision-making. Similarly, Abernethy and Russell (1987a, 1987b) demonstrated enhanced anticipatory behaviors of expert performers in badminton. Twenty expert and 35 novice badminton players were required to predict the landing position of a badminton shuttle-cock in response to varying levels of temporal occlusion (s ee later section on Occlusion Paradigms). Systematic expertnovice differences were apparent wi th the expert performers demons trating a prolific ability to use cues presented earlier in the action sequence to predic t stroke outcome. The novice

PAGE 43

43 performers were constrained in their ability and required more elaborate schemas to draw definitive and accurate conclusions. In a similar vein, Abernethy (1990b) re plicated the findings of Abernethy and Russell (1987a; 1987b) using s quash players. The consistent findings of researchers across a variety of sporting cont exts employing a diverse array of research paradigms, has repeatedly confirmed the enhanced abilities of experts to pick-up and process advanced cues that facilitate anticipation and response accuracy. Experts Maintain Superior Perception of Relevant Kinematic Information The use of advance cues by experts to facilita te anticipation speed and accuracy has proven robust, however the nature and extent of the cues necessary for enhanced decision-making in sport was relatively obscure until the advent of the occlusion paradigm. The seminal work of Jones and Miles (1978) assessed 32 profe ssional lawn tennis coaches and 60 novice undergraduates on their ability to predict the landing location of a tennis serve. Participants viewed 24 serves that were equally distributed and randomly presented down the centerline, to the middle of the service area, or to the extr eme right side of the service area. Temporal occlusion varied across three conditions of ball/k inematic exposure including eight frames (336 ms) post ball/racquet contact, 3 frames (126 ms) pos t ball/racquet contact, and one frame (42 ms) prior to ball/racquet contact. The use of a dvance cues was evident across conditions and expertise. Response accuracy remained cons istent across the two post-contact occlusion conditions, yet were significantl y impaired during the pre-contact condition. However, expertise differences were evident during the 126 ms pos t and 42 ms pre-contact conditions, with the expert level coaches demonstr ating significantly better respon se accuracy, suggesting that experience and skill level can influence advance cue usage proficiency. Specifically, kinematic information was readily available during both post-contact occlusi on conditions in which prediction accuracy was relatively stable. Howe ver, during the pre-cont act occlusion condition,

PAGE 44

44 only kinematic information was available proving to be the contributing factor in the expertnovice difference, with the experts using advance kinematic cues to predict the balls landing position. Although somewhat of an abstract inference, subsequent research has directly validated this point. Abernethy and Russell (1987a) implemented a spatial occlusion task, in which specific and relevant display features were masked from the participants. Consistent with the notion that kinematic cues may discriminate expert and novice performers, Aber nethy & Russell (1987a) hypothesized that expert-novice differences would ar ise as the result of the expert performers ability to pick up on task relevant cues earlier in the movement sequence, cues unattended to by less skilled performers. The spatial characteristic s of the relevant advance kinematic cues used by expert ( n =20) and novice ( n =35) badminton players were assessed while predicting the landing position of a badminton shuttle. Selected areas of the display were occluded, removing task relevant kinematic cues including the oppon ents racquet and arm, racquet only, head and face, lower body, and irrelevant ba ckground features. Results revealed that skill level influenced the reliance on advance cues, namely relevant ki nematic cues. Specifically, expert performance significantly improved from racquet and arm occlus ion to racquet only occl usion, whereas as the novice performers showed no additional performan ce change from one condition to the next. It can be concluded that the expert performers de rived task pertinent information from both the racquet and the arm as compared to the novi ce group that appeared to benefit only from information provided by the racquet. Expert-novice differences in th e ability to use spatial cues supports the notion that experts are more adept and in tune with th e movement strategies of their opponents, ultimately improving decisionmaking under tight time constraints.

PAGE 45

45 More recently, Shim and Ca rlton (1999) examined the influe nce of visual display on the anticipation of movement outcome on expert ( n =13) and novice ( n =12) tennis players. Participants observed an expert tennis player execute four shot combinations (i.e., groundstroke and lob either down the line or crosscourt) under three display conditions (i.e., live, 2dimensional, or point-light displa y), at which time they were re quired to perform the appropriate stroke in response. Consistent w ith previous findings, the results indicated that the expert players were more accurate and faster compared to the novice players acro ss the three conditions. However, when comparisons were made within groups, the expert players performed significantly better dur ing the live condition as compared to the 2-dimensional and point-light display conditions, whereas no differences we re noted across condi tions for the novice performers. In a similar study, Ward, Williams, and Bennett ( 2002) examined the effects of perceptual display manipulation in tennis. Experienced ( n =8) and inexperienced ( n =8) tennis players physically responded (i.e., moved one step toward the direction of the ball) to a series of filmed tennis strokes under normal and point-light display conditions. As expected, the findings showed that the experienced group performed better th an the inexperienced group under both conditions. However, both groups under the point-light disp lay condition experience d notable performance decrements. Specifically, the experienced gr oups response accuracy decreased nearly 10 percent, while the inexperien ced groups performance remain ed constant, suggesting that although relevant joint movements were available, critical advance cues under normal conditions are essential to the performance of skilled pl ayers and potentially ove rlooked by less skilled performers.

PAGE 46

46 Experts Maintain More Efficient and Effective Visual Search Patterns The visual search literature has systematica lly illustrated expert-novice differences for fixation location and duration characteristics that are postulated to be indicative of the perceptual strategy used to extract task-relevant information from the environment. Skilled performers apply their advanced knowledge structures as a conceptual framework for adopting more efficient and effective search st rategies characterized by fewer fi xations of longer duration, while fixating on the more information dense areas of the display. From an information-processing perspective, the eye-movement behaviors of expert s is theoretically more efficient and effective because information can be effectively chunke d, allowing for advanced associations and inferences, submitting fewer fixations of longer duration. Furthermore, Vickers (1996a) has proposed a unique gaze behavior appropriately labeled the quiet-eye (QE) period. Simply stated, the QE is a measure of the temporal period between the final fixation to a target and the initiation of a motor response; a period believed to facilitate the coordination of the processing of task relevant environmental cues and the formulation of the requisite motor plan for the successful comple tion of an upcoming task (Vickers, 1996a). Expertise research has routinely demonstrated that experts exhibit l onger quiet eye periods ( rpb = 0.62) when compared to less skilled performe rs (Janelle et al., 2000; Mann et al., 2006; Vickers, 1992, 1996a, 1996b). Quiet-eye research has also revealed intra-group differences, suggesting that longer quiet eye periods correspond with increased accuracy (Harle & Vickers, 2001; Janelle, Hillman, & Hatfield, 2000; Vick ers, 1996a, 1996b; Vickers & Adolphe, 1997). Experts Demonstrate Physiological Patterns I ndicative of Sensorimotor Efficiency The information-processing style of the expert-p erformer has been reliably characterized as more effective, efficient, and less effortful than that of the less skilled. As such, it was postulated that as skill level progressively increases, so too does the automaticity of performance,

PAGE 47

47 suggesting less cognitive involvement and effort among expert performers as compared to less skilled performers (Fitts & Posner, 1967). However, knowledge of the extent to which the expert exerts less effort or is otherwise more effici ent has primarily been the product of deductive reasoning stemming from decision-making, visu al-search, and response time and accuracy paradigms. Accordingly, the implementati on of electrocortical modalities such as electroencephalography ( EEG) have served to identify the co vert cognitive processing activity and the corresponding momentary changes in corti cal activity patterns across tasks and skill levels with excellent temporal resolution s upporting the cerebral efficiency hypothesis. EEG and spectral analysis techniques investig ating cortical activation and hemispheric specialization during the preparatory period of self -paced closed motor skills such as golf putting (Crews & Landers, 1993), archery (Salazar, La nders, Petruzzello, & Han, 1990), and shooting (Deeny, Hillman, Janelle, & Hatfield, 2003; Hatfield, Landers, & Ray, 1984, 1987; Hillman, et al., 2000; Janelle, Hillman, & Hatfield, 2000) have reported a progressive increase in alpha power in the left hemisphere and a relative stab ility in alpha power in the right hemisphere of elite performers as compared to less skilled performers. Given that increased alpha power is inversely related to cortical activation, the inform ation processing style of the expert is deemed more efficient than that of the less skilled performer. In addition to the continuous EEG and spectral t echniques, the use of event-related cortical potentials (ERP) has been a useful tool for de termining the time-locked cortical processes associated with a specific event. For example, the Bereitschaftspotential (BP; Kornhuber & Deecke, 1964) which occurs in the preparat ory period (i.e., 1-1.5s) immediately preceding a voluntary motor action is believed to represent the requisite pr eparation for the execution of a motor act (i.e., the motor program) (Brunia & van Boxtel, 2000). Research with marksmen

PAGE 48

48 (Konttinen & Lyytinen, 1993) and golfers (Crews & Landers, 1993) suggests that an increase in BP negativity corresponds with an increased re adiness to perform and performance excellence. As such, psychophysiological resear ch has validated a cerebral effi ciency hypothesis and further indicates that the expert maintains a welldeveloped sensorimotor program necessary for performance. The early work of cognitive psychologists and the relentless inquiry into human information processing have yielded much of th e information from which the above conclusions have been inferred. As such, the following section will proceed with a cursory review of the pivotal developments leading up to and significan tly contributing to the current understanding of information-processing and cortical changes as ap plied to sport. Furthermore, an attempt to isolate the gaps in the current understanding with respec t to the role of the QE period in motor preparation and/or emotion regulation for succes sful sport performance will be included. Visual Perception in Sport Expert performance is mediated by a number of factors including cogni tive and perceptual motor skills, as well as task specific anatom ical and physiological ad aptations (Ericsson & Lehmann, 1996). Moreover, Janelle and Hillman (2003) postulate that in order to obtain expert status, athletes must excel in no less than four domains: physiological, technical, cognitive (tactical/strategic; perceptual/ decision-ma king), and emotional (regulation/coping; psychological) (p.21). It is comm only understood that expert pe rformance is a product of the delicate balance between innate talents and the amount of practi ce/training (Ericsson, Krampe, & Tesch-Rmer, 1993). Of particular importance is th e latter, in which the at hletes declarative and procedural knowledge base develops and become s both extensive and accessible. The extent of this knowledge can be inferred from indirect pe rception/action coupling a nd more specifically by means of the visual search behaviors and a dvance cue utilization ab ilities of exceptional

PAGE 49

49 performers as compared to le ss skilled performers. The follo wing review is an attempt to summarize a dense literature ba se encompassing the expert-nov ice paradigm and the varied research methodologies used to partition the perc eptual differences noted across skilled and less skilled performers. Occlusion Paradigms In an attempt to reveal the most pertinent a dvance cues present in the environment, or at least those most frequently used by expert athletes, laboratory researchers have made extensive use of the occlusion paradigm. The use of the oc clusion approach for the study of sport includes temporal and spatial techniques. In the case of the temporal paradigm, researchers have identified pivotal points during th e flight of a ball, shuttle, or movement of an opponent at which time further visual information becomes inaccessibl e (e.g., a blank screen appears). Typical sport occlusion paradigms often include experimental conditions simila r to the following adapted from Abernethy and Russells (1987a) investigation of racquet sport athletes to assess advance cue usage: t1: Occlusion of the display 4 frames ( 167msec) prior to racquet-shuttle contact; t2: Occlusion of the display 2 frames ( 83 msec) prior to racquet-shuttle contact; t3: Occlusion of the display at the point of racquet-shuttle contact; t4: Occlusion of the display 2 frames ( 83 msec) subsequent to r acquet-shuttle contact: t5: No occlusion of the display until all ou tward flight of the shuttle was completed. Recognizing the inherent limita tions of temporal constraint s (e.g., incomplete viewing of task), researchers have adapted the paradigm to occlude spatial or event cues. The advent of spatial occlusion techniques allotted the research er more experimental control over what features of the display the participant c ould and could not see. Of primar y importance, spatial occlusion permitted a continuous stream of information to be provided to the partic ipant with the exception of the critical cues in questi on, allowing the researcher to is olate and infer the perceptual

PAGE 50

50 strategies of an athlete. Adapted from Ab ernethy and Russell (1987b) the following sequence depicts a typical approach to the implementa tion of a spatial occlus ion paradigm and its experimental conditions. e1: The players racquet and arm holding the racquet were occluded; e2: The players racquet (but not the arm holding it) was occluded; e3: The players face and head were occluded; e4: The players lower body was occluded; e5: Irrelevant background features were occluded. Occlusion Research In a seminal study, Jones and Miles (1978) adop ted the temporal occlusion paradigm in their investigation of advance cue use in lawn te nnis. In support of the superior ability hypothesis of experts to extract relevant information from advance cues, Jones and Miles concluded that perceptual information germane to decision accur acy is readily availabl e to athletes throughout the flight path of a tennis serve, regardless of performance level. However, as pertinent cues were occluded, level of expertise accounted for significant differences in prediction accuracy. Specifically, when the serve was occluded s hortly (126 ms) after impact, performance differences were notable. Additiona lly, when occluded just (42 ms) prior to ball-racquet contact, performance differences were pronounced in favo r of the top-level performers, signifying the ability of skilled tennis players to effectively ma ke use of advance perceptual information in the performance environment typically provided by the opposing player. In a similar study, Isaacs and Finch (1983) assessed the anticipatory timing of beginning ( n = 34) and intermediate ( n = 16) tennis players. Four tem poral occlusion conditions (i.e., 10 msec before contact; 0 msec at contact; 15 msec post-contact; and 30 msec post-contact) were implemented to examine differences between te nnis proficiency level and the participants ability to accurately predict the placement of a tennis serve. Immediately following each viewing condition participants indicated the anticipated landing positi on of the serve on a specially

PAGE 51

51 designed score sheet which replicated the divisi ons placed on the deuce service court during the filming of the serve. Based on the recommendations of Jones and Miles (1978), Isaacs and Finch assessed not only the percentage of correct respon ses for exact ball location, but also the degree of accuracy for latitude (direction) and longitu dinal (depth) predictions. The results of this investigation mirrored thos e of Jones and Miles (1978). That is players of greater ability were more accurate in their predic tions of landing area across occlusion conditions, with more pronounced differences evident in the 10 ms prior to the ball-racquet contact occlusion condition, the condition with the le ast amount of advance perceptual information requiring the greatest amount of inference. Moreover, the intermediate players also demonstrated superior latitudinal prediction precision across temporal conditions, while longitudinal di fferences were less distinct. Additionally, a significant time effect was evident, as was a significant interaction of ability by time. The authors concluded that in the latter co ndition players of intermediate and novice ability did not posses the requisite skills to identify slight racquet angle va riations during the serve, cues that are subtle and sufficient enough to in fluence the perception of service depth. However, the very nature of stimulus presen tation methods used in occlusion paradigms presents an inherent limitation of this type of research. In accord with the commentary provided by Isaac and Finch (1983), the inability for the in termediate performers to accurately predict longitudinal placement may not be at all related to ineffective cue use, but rather the result of using a two dimensional representation of a thre e dimensional space. Simply, the angle at which the film was recorded may have indirectly occl uded the information necessary to acquire depth perception cues. Nevertheless, skill proficiency ha s again accounted for differences in the ability to predict the landing position of a tennis se rve across temporal oc clusions conditions.

PAGE 52

52 In an attempt to extend the previous rese arch, Abernethy and Ru ssell (1987a) conducted two experiments to independently establish and compare the tempor al and spatial characteristics of the advance cues used by expe rt and novice sport performers. In Experiment 1 five temporal occlusion conditions were used (i.e., 167 msec prio r to racquet-shuttle co ntact; 83 msec prior to racquet-shuttle contact; occlusion at the point of racquet-shuttle contact; 83 msec subsequent to racquet-shuttle contact; and no o cclusion). Participants were 20 expert and 35 novice badminton players who were required to predict the pr obable landing position of a badminton shuttle. Immediately following each trial, all responses were recorded on a scaled repres entation of the receivers court. The findings were consistent with previous results, in that differences in prediction accuracy were notable between e xpert and novice performers across occlusion conditions. That is, from 83 msec prior to racq uet-shuttle contact th rough the no-occlusion condition, expert and novice players differed in their performance accuracy. The authors also concluded that the cues essential for successful directional perception are present during the interval between 83 msec prior to racquet-shuttle contact an d 83 msec post racquet-shuttle contact, while depth perception cues are present in the final 83 msec prior to racquet-shuttle contact. The collective findings suggest advance cu es are apparently critical for depth perception whereas a greater window of opportunity presid es for directional detection (Abernethy & Russell, 1987a). In an effort to isolate experts attention to the most salient of perceptual cues, Abernethy and Russell (1987a) adopted a spatial occlusion paradigm in Experiment 2, allowing for the constant temporal display of perceptual informa tion while controlling the occlusion of explicit cues. The same population of badminton players part icipated in this study as in study one. In a similar fashion, 32 different badminton strokes we re included and subject to occlusion. The

PAGE 53

53 following features of the display were systematica lly removed: (e1) the players racquet and arm holding the racquet; (e2) the play ers racquet; (e3) the players face and head; (e4) the players lower body; and (e5) an irrelevant background feature. The results of experiment two were again consistent with previous occlusion studies. Th e expert group demonstrat ed superior prediction accuracy (i.e., lower radial error, lateral error, and depth error) across occlusions conditions, with the exception of occlusion c ondition (e1), in which expert-novice performance was indistinguishable. It should be noted however, that although with in group comparisons re vealed significant performance improvements from condition (e1) to (e2) for the experts, the novice performers accuracy did not change. These results signify th e fundamental differences between experts and novices for predicting the landing position of the badm inton shuttle. First, experts were able to glean useful information from the opposing player s arm compared to the novice players who were only able to extract useful information from the oppositions racquet. Second, the time in which principle information was removed from a distal (e.g., racquet) to a more proximal (e.g., dominant arm) region signifies the advance use of kinematic information that precedes the motion of the more distal racquet to encode subsequent racquet movement and ball flight information. The two stage experimental appr oach used by Abernethy and Russell (1987a) not only provides additional empirical support for the findings of previ ous researchers (i.e., Jones & Miles, 1978; Isaac & Finch, 1983) employing temporal occlusion techniques, but Abernethy and Russell (1987a) were able to further isolate th e importance of specific perceptual cues (i.e., opposing players arm) necessary for the accurate estimation of the badminton shuttles landing position.

PAGE 54

54 In a follow-up study, Abernethy and Russell (1987b) replicated the te mporal and spatial occlusion technique as reported earlier (Abernethy & Russell, 1987a). However, in attempt to extrapolate to the expert advantage of superior anticipatory cue use, eye movement registration equipment was implemented to assess the search strategies of expert and novice badminton players (for a discussion of the visual search findings, please see the section entitled Visual Search) Results from this study echo those of ear lier work. Expert players maintained a significantly lower predic tion error rate across all temporal occlusions conditions except (t1), 167 msec prior to racquet-shuttle contact, and across all spatial o cclusion conditions except (e1), the occlusion of racquet and arm In the current context, the au thors concluded that only the experts possessed the necessary knowledge structures to system atically construct perceptual connections from the information provided in th e environment to that which was necessary for successful performance. Despite the robust expert-novice differences acr oss a variety of racquet sports, Abernethy (1988) set out to understand the developmental char acteristics of perceptu al skill and selective attention contributi ng to such differences using both tem poral and spatial o cclusion techniques. Matched groups of relative expert and novice badm inton players from four distinct age groups (i.e., 12yrs, 15yrs, 18yrs, adult) were assessed using the same occlusion task previously employed by Abernethy (1987b). It was hypot hesized that the e xpert group would systematically use advance cues more proficient ly than novices and that this distinction would become more apparent across group s as the participants became ol der. Consistent with previous findings in sport (Abernethy & Russell, 1987a 1987b; Isaac & Finch, 1983) it was further predicted that selective attention would vary as a function of skill level. Analysis of the radial error in the temporal occlusion condition re vealed significant performance differences among

PAGE 55

55 age, expertise, and occlusion conditions. Thes e results signify the developmental changes associated with skill proficiency. That is, hi gh skill combined with maturation appears to positively influence perceptual skill but maturati on has little to no effect when performance ability is low. Figure 2-2 depicts this rela tionship between age effects and expertise on anticipatory cue use as a func tion of the occlusion condition. The implementation of the temporal occlus ion condition identified critical age and expertise differences during the most significant viewing times for attaining advance perceptual cues. However, the extent to which selective attention influences prediction accuracy as a function of age and skill could not to be determined. Therefore, in an attempt to bridge this gap Abernethy & Russell (1987a) condu cted a spatial occlusion anal ysis and upheld the previous findings reported earlier. Results indicated that ex perts, regardless of their age, are more adept at extracting perceptual information from both the opponents racquet and arm whereas novice badminton players were inept at using the kinema tic cues provided by the opposing players arm, relying on the racquet only for a dvance perceptual information. Abernethy and Russell (1987b) examined the per ceptual differences of two groups distinct in performance ability (i.e., in ternational level and novice undergraduate students), identifying ostensible expert-novice differences. The comparis on of such variable performers lends itself to large effects. However, Abernethy (1989) probed furt her in an effort to de termine the sensitivity of these differences and whether or not the pe rceptual advantage was merely a product of expertise. Using a temporal and spatial oc clusion task (see Abernethy & Russell, 1987a) intermediate ( n =12, skilled but not elite) and novice ( n =15, undergraduate students) badminton players were assessed on their ability to predict, from the film presented, the probable direction of their opponents stroke. The results of the temporal cond ition show intermediate-novice

PAGE 56

56 differences beginning as early as 167 ms prior to racquet-shuttle contact with the largest differences evident 83 ms prior to contact, conf irming Abernethy and Russells (1987a) previous findings with expert and novice players. The results of the spatial occlusion task also illustrate intermediate-novice differences in the prediction of stroke direction. Most notably, the error rate for occlusion conditions e1 (arm and racquet) and e2 (racquet only) differed significantly for both groups as compared to all other occlusions conditions. However, inconsistent with previous findings (Abernethy & Russell, 1987 a), the intermediate group did not statistically perform better than the novice group during c ondition e1, yet it should be not ed that performance trends were similar to the expert-novice differences of Abernethy and Russell (1987a). These results signify that perceptual advantag es are evident with el evated levels of performance, even though performance may not be elite (Isaac & Finch, 1983). In another study of tennis players, Buckolz, Prapavesis, and Fairs (1988) attempted to delineate the specific advance perceptual cues used by advanced ( n =21) and intermediate ( n =23) tennis players during passing shots. A series of tennis strokes (i .e., down-the-line, cross-court, and lob passing shots, for both forehand and backha nd) were filmed at two different speeds (i.e., 24 frames/second and 48 frames/second) allowing for greater experimental control over the duration of cue exposure. A tempor al occlusion paradigm was used to facilitate th e identification of the cues used by and those that discriminated between the advanced and intermediate players. The authors unconventionally orde red the sequence of film clip s, beginning with the most occluded condition (168 ms at 24frames/second and 84 ms at 48frames/second prior to ball contact) for a given stroke through to the le ast occluded condition (168 ms at 24frames/second and 84 ms at 48frames/second post ball contact) fo r that stroke. Once each occlusion condition had been delivered for a specific stroke, the next stroke was introduced. To further clarify, each

PAGE 57

57 participant viewed each stroke under all occlusion condi tions and at both film speeds prior to viewing the next stroke in the same sequence under the same conditions. The results of Buckolz et al. (1988) are consistent with thos e previously reviewed; experts maintained an advantage over novice performers wh en advance perceptual cues were available. With the exception of down-the-l ine passing shots, significant di fferences were evident during the 168 ms prior to ball-racquet contact and at the moment of ball-racquet c ontact. In addition to providing further empirical support in favor of experts ability to use advance cues, a significant contribution was garnered. Specifically, regardle ss of the duration of temporal viewing period and the amount of early ball-fli ght information, it was more difficult for participants to accurately anticipate the backha nd shot as compared to the forehand. Even though experts are adept at attending to specific kinematic cues that often aid in the predic tion of stroke outcome, experts are unable to anticipate much bette r than chance in the event of a backhand. The previous research, although significantly contributing to the em pirical understanding of expert-novice perceptual differences, neglecte d the potential role of peripheral vision in cue acquisition and utilizatio n. In a modified occlusion paradigm, Davids, Rex Pe Palmer, and Savelsbergh (1989) examined the anticipatory ability of elite, club, and recreational tennis players using a forehand volley. Although an occl usion paradigm was used, the nature of the paradigm and the experimental ta sk differed from those previously discussed. Participants wore a helmet that included Perspex sheets (i.e., opaque, clear, and no-screen), and acted as a visual occlusion device, reducing the visual field by 70 and thus occluding perception of the participants own arm and racquet for the fina l 100-150 ms of ball flig ht. Although researchers typically view perceptual skills as those relative to the flight and landing location of a projectile or the movements of an opponent, pe rceptual skills in this case related to hand-eye-coordination

PAGE 58

58 and the interception of a pr ojectile in flight. In essence, this investigation assessed the role of the visual system and information processing durin g perception-action coupli ng. Participants were rated on placement accuracy and stroke quality while volleying a tennis ball at two separate speeds (i.e., 29.06 m/s and 20.12m/s). The tennis court was divided into distinct scoring zones to tally points for accuracy while quality was determin ed by a direct hit versus a miss-hit. Results indicated that performance was superior in th e slower ball speed condition across skill levels, while there were no reported differences across screen conditions or le vels of expertise. Davids et al. (1989) speculated that to execute a tennis volle y, one does not rely on visual feedback of effectors (i.e., ones own arm and racquet) due to the la rge surface area of the racquet, which is responsible fo r the interception of the projectil e. However, the lack of skillbased differences may be a product of the task. Specifically, a ball machine was used to deliver the ball to be volleyed. With the exception of th e variable ball speed, th e machine operated with relative positional consiste ncy and accuracy. Therefore, in a matte r of a few trials it is suspect that even the most novice perfor mer could anticipate with relativ e accuracy the end-point of the incoming ball relative to ones body position. Alt hough an attempt was made to implement an ecologically valid task, a more dynamic approach (e.g., including backhand shots) is warranted. Additionally, the fact that the effectors were occluded for onl y the final 100-150 ms suggests that relative body positioning may be occurring relative ly earlier in the perceptual-motor process rendering the final temporal period obsolete. Returning to the typical tempor al occlusion paradigms of earli er researchers, Goulet, Bard, and Fleury (1989) assessed expert ( n =10) and novice ( n =10) tennis players on their ability to correctly identify the type of serve presented (i.e ., flat, top-spin, slice) under five occlusion conditions (i.e., (1) Preparatory Phase (875 ms ), (2) Preparatory Phase until elbow reached

PAGE 59

59 maximal height (1125 ms), (3) Preparatory Phase until ball/racquet contact (1208 ms), (4) Ritual Phase until ball/racquet contact (4710 ms), (5) entir e serve without occlusion(5048 ms)). Bard and colleagues present results are consistent with previous occlusion rese arch in that expert tennis players are better able to extrapolate and interpolate perceptual information to facilitate prediction accuracy. Additionally, experts not only use the information they extract differently they require less information all together (G oulet, Bard, & Fleury, 1989). Unique to this investigation however, the authors also assessed decision time and conclu ded that experts were not only more accurate but they arrived at their decisions much quicker than novices. Perhaps this distinction is the result of requiring less information to reach their conclusion while benefiting from the information that appears earli er in the display as compared to the novice players who depend on information presente d much later in the event sequence. To identify the specific visual cues used during squash performance as well as any systematic differences that may arise in cu e use between expert and novice performers, Abernethy (1990a) applied the same paradigm to the study of expert and novice squash players (for the occlusion conditions applied, please refer to Abernethy and Russell, 1987a discussed earlier). Participants (expert, n =16; novice, n =20) viewed a sequence of film depicting variations of squash stroke and were required to verbal ly respond to the force and direction of their opponents shot (i.e., down or cro ss for direction and long or short for force) with the intention of predicting the terminal point of the stroke. Unique to this investigation, participants were interviewed following the experiment and aske d to reflect upon thei r experience and the naturalness(p.23) of the task as compared to th at of an actual squash experience. Additionally, participants were probed as to the importance of each of the seven perceptual cues made available during the testing sessi on (i.e., oppositions racquet, h ead, lower body, torso, dominant

PAGE 60

60 arm, non-dominant arm, and court position). Resu lts from the temporal occlusion condition for lateral error again suppo rted previous findings, indicating th at experts extracted information earlier in the visual display (i .e., 160-80 ms prior to ball/racque t contact) and more information (i.e., kinematic cues) than novice players, enhanci ng the prediction accuracy for stroke outcome. Additionally, experts were better able to predict stroke forc e (depth) as compared to novice squash players across occlusions conditions. Surprisingly, both the expert and novice groups continually improved prediction accuracy from (t 1) through (t5). Previous findings have noted that novice performers typically do not improve pe rformance until much later in the display (t3), however this does not appear to be the case here. According to Aber nethy (1990a), squash players may have a tendency to reveal more acces sible information about stroke depth earlier in their preparation than badminton players, implying that the exis ting occlusion condition failed to suppress a sufficient amount of advance information. Results from the spatial occlusion conditi on for lateral error mirrored those of the temporal condition, in that the experts were more accurate in their predictions across conditions. However, contrary to previous investigations (i.e., Aberne thy, 1988; Abernethy & Russell, 1987a) that have reported differe nces under occlusion condition (e1) -(e2) for experts and as early as (e2)-(e3) for novice performers, no within gr oup performance differences were noted across occlusion conditions. Again, expert ise differences were noted for depth error, with the skilled players predicting more accurately across all o cclusion conditions. In accord with previous findings, both experts and novices demonstrated performance decr ements when both the racquet and arm were occluded. Yet, unique to squash, th e opponents head was also a significant source of advance information as illustrated by syst ematic performance decrements in prediction accuracy across groups while the head was occluded.

PAGE 61

61 Previous researchers (e.g., Jones & Miles, 1978 ) have questioned the ecological validity of laboratory-based investigations. Therefore, in response to such skepticism, Abernethy (1990a) employed a subjective self-report in dex assessing the naturalness (i.e., ability to replicate and provide a life-like simulation) of the visual display characteristics and temporal stress associated with squash performance. Results revealed no differences between experts and novices with a mean group rating of 4.53 out of 7 for display ch aracteristics and 3.47 out of 7 for temporal stress. This finding suggests th at participants were relativel y engaged by the film task but thought the trial-to-trial period was too great failing to capture the tempo of competition. Abernethy (1990a) also collected subjective data to isolate the importance of specific perceptual cues that may account for the robust expert-novice differences. Both the expert and novice groups identified the racque t and the arm holding the racquet as the most important cue, a finding consistent with Abernethy and Russell ( 1987a). Empirically, novi ce performers appear unable to utilize advance kinematic informati on in a manner similar to the experts, yet no subjective differences identifying th e most pertinent sources of in formation were noted. As such, experts apparently not only know where to look but they also main tain a wealth of declarative information, information that is seemi ngly unavailable to the novice performer. The laboratory basis of the previous occl usion studies, althou gh sound methodologically, have been questioned for their ecological validity. For example, Jones a nd Miles (1978) discuss the inherent sterility of the la boratory and the inability of a la boratory setting/task to accurately elicit comparable performance states. Such lim itations may confound the empirical estimates of perceived expert-novice differences. Additiona lly, Abernethy and Russell (1987b) questioned the validity of the film presentation used in previous research and suggested that the use of film may neutralize any notable expe rt-novice differences.

PAGE 62

62 With direct consideration gi ven to the potential confounds of the laborato ry, Abernethy (1990b) implemented a two experiment desi gn. Experiment 1 was conducted inside the laboratory and Experiment 2 on a squash c ourt using a temporal occlusion technique supplemented with the collection of eye movement data. The overall visual search findings of Experiment 1 will be discussed in a later section. In the first experiment Abernethy (1990b), s quash players were re quired to predict the stroke outcome from a filmed representation of an opposing player. Experimental conditions included the same temporal conditions as repo rted in Abernethy and Russell (1987b). Analyses of prediction accuracy for both stroke depth a nd direction across occlus ion conditions revealed significant expert-novice differences confirming the utility of adva nce cues for skilled players. In light this robust finding, improved predicti on performance was found for stroke depth as compared to stroke direction across skill levels. This finding was contrary to previous research (i.e., Isaacs & Finch, 1983) that had reported increased difficulty for players to detect slight racquet variations and kinematic cues responsib le for changes in depth. Overall, the findings support the rapid abilities of expert athletes to make use of more cues germane to prediction accuracy (Abernethy & Russell, 1987a, 1987b; J ones & Miles, 1978; Isaac & Finch, 1983). However, despite this finding, onl y slight visual search characteristics were evident, with the expert allocating more fixations to their opponents arm and head c oupled with fewer fixations to the contact zone as compared to the novice performer. Moreover, across the entire visual search sequence no skill-based differences fo r fixation duration were evident. In the second experiment, in which a fi eld-based assessment was conducted, the same basic conclusions were derived. Simply, al though the experts outperformed the novice comparison group, the visual search behaviors of the experts were virtually indistinguishable

PAGE 63

63 from the search behaviors of the novice group. Abernethy further concluded the perceptual advantage of the expert, although evident, is not the result of access to specific visual information, but rather the capability to extract and utilize information fr om key fixation points (Abernethy, 1990b, p. 75). Analyses of individual, interc eptive, racquet-sports have be en the dominant choice for the bulk of research on advance visual cue use in sp ort. Although a wealth of information has been obtained from such work, little has been done to identify the critical cues for successful anticipation in team sports that are arguably inundated with more complex visual environments. Often in team sports, the anti cipation of and reaction to the development of offensive plays results in their successful prev ention. In-line with this logic, Wright, Pleasants, and Gomez-Meza (1990) examined the perceptual strategies of experienced ( n =12) and novice ( n =12) volleyball players ability to identify the s piker (i.e., left, right, or center net position) to whom the setter intended to pass. A temporal occl usion paradigm was used to syst ematically alter the amount of visual information pre and post se tter contact. The five conditi ons were as follows: C1: 167 ms (5 frames) prior to initial setter contact; C2: at initial setter contact; C3: 167 ms (5 frames) after initial setter contact; C4: 333 ms (10 frames) after initial setter contact; C5: 499 ms (15 frames) after initial setter cont act. All decisions were recorded a nd analyzed for response accuracy and subjective reports were collected in an attempt to identify the most salient cues. Results indicated that the experienced group demonstrated superior response accuracy across occlusion conditions (i.e., C1-C3) with the most pronounced differences occurring at initial s etter contact (C2). No differences were noted for conditions C4-C5; both groups performed with 100 percent accuracy. Furthermore, the information garnered from the self-reports for assessing the perceptual strategies of these participan ts was distinct. The experien ced group reported focusing on the

PAGE 64

64 setters body, followed by hands and ball flight Conversely, the novice group identified the overwhelming importance of ball flight, followed by the spikers move, and setters body as the most pertinent cues. These variations in percei ved locus of attention are exclusive to this investigation, as the study by Ab ernethy (1990b) found no differences in self-reported direction of attention between experts and novice squash players. Based on the unique characteristics of each sport one can only speculate as to the root of these differences. Regardless, the work of Wright et al. (1990) extended beyond racquet sport research to vol leyball signifying the importance and effective use of advance visual cues by experienced athletes across multiple sports. Houlston and Lowes (1993) further addressed the depth and direction prediction while assessing the cue-utilizati on processes of expert ( n =6) and non-expert ( n =6) wicketkeepers. A film occlusion technique with four distinct temporal conditions incl uding; (t1), ball release; (t2), 156 msec post release; (t3), 234, sec post release; (t4), 390 ms ec post release) was used. Wicketkeepers were required to view a series of distinct deliveries and anticipate the landing position of the ball. A target mat was positioned in front of the batting wicket that was used to assist in the scoring of pitches. All predictions were recorded on a scaled version of the target mat and were evaluated for radial error as well as depth and lateral error. The results of this investigation were atypical to those previously noted. That is no differences between group prediction accuracy were noted across occlusion conditions. However, significant prediction differences were evident for lateral versus depth error; both the e xpert and novice groups predicted lateral error more accurately across occlusion conditions as compared to depth predictions with a sign ificant improvement in depth percep tion occurring under (t4). The results suggest that lateral estimations can reliably occur early in the pe rceptual process, while depth

PAGE 65

65 estimations require more time and more detail to be extracted in order to approach a level of accuracy resembling lateral estimations (see Figure 2-3). Up to this point in the review, the majority of the investigations of advance visual information in sport have relied on the use of la boratory based film occlusion techniques, with one notable exception (Davids, Re Pe Palmer, & Savelsbergh, 1989).In an attempt to extend the findings from the laboratory to the real world, Starkes, Edwards, Dissanayake, and Dunn (1995) examined the role of experience and skill in the us e of advance visual cues in volleyball. Skilled ( n =8) and novice ( n =8) volleyball players were assessed on their ability to predict the landing position of a volleyball serve while standing in th e center of the servi ce reception side of a regulation volleyball court. Participants were fitted with liquid crystal visual occlusion spectacles, which were used to occlude the volleyba ll serve at three distinct stages (i.e., (e1) precontact (ball reached highest point in toss), (e2) contact (hand struck the ball), and (e3) postcontact (prior to the ball cro ssing the net)). Participants obs erved the serves under the various occlusion conditions and then pl aced numbered markers on the fl oor indicting the anticipated landing position. Results from this investiga tion confirmed differen ces in expert-novice perceptual abilities; with the experts perfor ming significantly better across all occlusion conditions. A significant effect for occlusion conditions was al so noted, signifying that precontact occlusion proved most challenging to bo th skill levels. No differences were noted between the contact and post-cont act occlusion conditions. It can therefore be concluded from this field study of advance perceptual cue use th at experts extract more pertinent cues allowing for more accurate responses. In a re-examination of the differences in anticipatory decision-making in tennis among expert, intermediate, and novice level players, Tenenbaum, Levy-Kolker, Sade, Liebermann, and

PAGE 66

66 Lidor (1996) used a temporal oc clusion paradigm. Eight differe nt tennis strokes were video taped (1. Cross-court slice, 2. Forehand down th e line, 3. Backhand down the line, 4. Backhand (winner) down the line, 5. Fore hand volley (winner) near the ne t, 6. Serve, 7. Backhand drop volley cross-court, 8. Forehand cr oss-court) and were presented in the laboratory using the following six occlusion conditions: 1. 12 frames (-480 msec) prior to ball-racquet contact. 2. 8 frames (-320 msec) prior to ball-racquet contact. 3. 4 frames (-160 msec) prior to ball-racquet contact. 4. At the point of ball-racquet contact. 5. 4 frames (+160 msec) subseque nt to ball-racquet contact. 6. 8 frames (+320 msec) subseque nt to ball-racquet contact. After viewing each occluded stroke participan ts responded as quickly as possible to the landing position of the ball. A response sheet with a scaled replica of the tennis court was used to record the predicted landing locations. Radial error for response accuracy was assessed across occlusion conditions. The results of this investigati on mirrored those of previous research, with the expert and intermediate groups outperforming the novice group in the advance occlusion conditions (i.e., 480, -320, and msec prior to ballracquet contact). No differences were noted between the expert and intermediate groups across occlusi on conditions, suggesting th at the benefits of advance perceptual skill may reach an asymptotic point when a level of performance competency has been attained. In conclusion, early temporal cues (i.e., cues up to th e point of ball-racquet contact) appear to be detected and relied upon by skilled players, wh ile remaining unprocessed by less skilled performers. Interceptive sports such as the racquet spor ts discussed here (i.e ., tennis, badminton, and squash) are predominant across sport science research on cue utiliz ation. One of the few exceptions to this rule is the work of Paul and Glencross (1997) with expert ( n =15) and novice

PAGE 67

67 ( n =15) baseball players, which examined the use of visual information throughout the duration of a baseball pitch using a typical te mporal occlusion paradigm, includ ing: (t1) 80 ms prior to the moment of ball release (MOR); (t 2) at MOR; (t3) 80 ms after MO R; (t4) 160 ms after MOR; and (t5) 240 ms after MOR. Batters were required to pr edict, as quickly as po ssible, the location of the pitch as it passed through a gr id imposed over the strike zone Unlike previous findings (e.g., Abernethy & Russell, 1987a; Goul et et al., 1989; Jones & Miles, 1978) experts and novices did not differ in their ability to pr edict the landing position of the projectile. Even more peculiar was the direction of the trend (Figure 2-4). Systematic differences have been demonstrated indicating the ability of high skilled players to make be tter use of advance cues with differences in performance ability marginalized later in the temporal condit ions. However, the opposite trend was depicted here; as more time was allotte d to view the flight of the ball (i.e., 10 m) the larger was the difference in mean error scores between the expert and novice gr oups. Finally, when viewing pitches it appears that the first 80 ms (t3) after the pi tch has been released is the most crucial to pitch detection, as the most notable performance differences between skill levels occurred at this point in time. In addition to group differences, (t3) is a critical time frame for pitch recognition, that is, disti nguishing the type of pitch (e.g. curveballs vs. fastballs). Experts ability to extract advance cues fo r pitch recognition and prediction accuracy is questionable in the study of expert and novice baseball players. However, consistent with previous research, the temporal relationship be tween perceptual cue use and prediction accuracy still holds in favor of the expert performers, albe it occurring later in th e cue extraction process. Perhaps one explanation is the required time necessary to visually acquire the spin of the ball for the corresponding pitch, a cue essential for rec ognition and response accuracy. In comparison to racket sports, pitchers in baseba ll are trained to mask the delivery of the ball until the absolute

PAGE 68

68 latest point in the delivery, this process may confound early prediction accuracy in this investigation as compared to the robust early expert-novice differences as seen elsewhere (Abernethy & Russell, 1987a, 1987b). Previous research has reported equivocal finding s as to the proficienc y for anticipation of lateral and depth positioning among expert and novice athletes, with depth positioning being more difficult to predict with re lative accuracy. Such difficultie s may prove troubling to sports (i.e., cricket, baseball, and tenni s) that require acute depth per ception while anticipating lateral direction. For example, a bowler in cricket will bounce the ball in front of the batter and depending on the spin, trajectory, and speed, the ball will react differently with the ground, forcing the batter to anticipate and react accordingl y. The batters task in cricket is not unlike that of baseball, requiring the use of advance perceptual cues incl uding kinematic cues provided by the bowler. Ball speed and rotati on must also be processed eff ectively for the batter to make contact. Renshaw and Fairweather (2000) assessed the perceptual discri mination ability of national ( n =6), regional ( n =6), and club ( n =6) cricket batters exposed to 5 different types of bowling deliveries (legspinner, toward batters feet; topspinner, drops short followed by high a bounce; googly, away from batter; flipper, close to feet with low bounce; b ackspinner, similar to flipper with lower bounce) from two temporal occlus ion conditions. The first condition contained the bowlers run-up and ball flight up to and in cluding ground contact, while the second condition included the bowlers run-up and the first 80 ms of ball flight. Immediately following each trial the batter verbally reported the delivery type. Th e results revealed that overall the experts were more accurate than the regional and club level pl ayers for delivery discrimination. Consistent with previous findings, skill level was indicative of advance cue use (condition 1), however,

PAGE 69

69 unlike previous findings, the cric ket batters did not benefit from addition ball flight information, suggesting that kinematic cues provided by the bo wlers run-up is critical to successful pitch detection. It should be noted th at perceptual information afte r ground contact was not provided and may have proved useful for pitch detection and performance. Specifi cally, the nature of certain deliveries (i.e., legspinne r and googly) will result in th e ball changing direction after ground contact. This information may play a signif icant role in batting success over and above that of perceptual discrimination. Summary Researchers for nearly three decades have i nvestigated the relationship between advance cue use and anticipation in spor t with the intention of unveiling the core differences between expert and novice athletes. The ability to foresh adow events was believed to result from experts extensive knowledge base and their ability to ap ply that knowledge in a manner that facilitates advance visual perception. The use of occlusion paradigms, introduced to sport researchers by Jones and Miles (1978), was swiftly espoused as the paradigm of choice to probe the perceptual behaviors of athletes. The use of both temporal and spatial occlusion tech niques across a variety of sports including, tennis, badminton, squash, cric ket, baseball, and volleyball, systematically demonstrated expert-novice dispar ities in the use of information presented early in the visual display. A summary of these experiments suggest s that: (1) experts ar e better able to use kinematic cues (such as the dominant arm of a tenni s player) that maintain subtle clues as to the direction and force of a tennis str oke. (2) experts are more adept at using early flight cues of the badminton shuttle to predict stroke location, cu es not utilized by novice performers until much later in the flight, and (3) during volleyball offensive attack formations, advanced players were able to use early ball flight and kinematic cues to predict striking loca tion opposed to the novice players who relied on ball flight information to base their t actical decisions. The findings

PAGE 70

70 reported here have been relatively consistent signifying the attuneme nt of expert level performers to advance cues, otherwise neglected by novice performers. The utility of the occlusion paradigm ha s been clearly confirmed, but the inherent limitations of this approach should not be left uns tated. First, occlusion paradigms, both temporal and spatial, lack ecological vali dity. That is, the cues and deci sions apparent in the laboratory cannot be inferred with confidence to reflect the cues and decisi ons influenced by the modulation of competition arousal, motivation and attenti on (Miles & Jones, 1978). Second, the use of temporal occlusion techniques proh ibits the sequential connections of perceptual information that promote the use of altered cognitive processes. Rarely in sport is the athlete unable to view the opposition in his/her entirety, yet the occlusion para digms inherently restrict the presentation of information. From an information-processing appr oach this may yoke very different connections between perceptual stimuli and declarative knowledge necessary to reach an accurate problemsolution. Third, the use of film and slide presen tations reduce a three-di mensional world into a two-dimensional space, altering th e perceptual and sensory experi ence. For example, Isaacs and Finch (1983) reported greater difficulty and decreas ed response accuracy for depth as opposed to lateral predictions that could la rgely be a product of the artifici al display. Fourth, static slide presentation, although amenable for eye movement registration, fails to capture the dynamic nature of the visual environment within most sporting domains (Abernethy, Burgess-Limerick, & Parks, 1994). Finally, self-report indi ces of perceptual cues used for anticipation in sport have been equivocal, signifying the inab ility to rely on such information to confirm the nature and importance of the perceptual cues actually used in the decision-making process. Regardless of these limitations many advances in the understanding of advance cue use in sport have emerged from the use of occlusion pa radigms. However, the need to validate the

PAGE 71

71 many assumptions put forth is warranted. In an attempt to resolve the many limitations of earlier research, eye-movement registration techniques were adopted. The following section will review the expert-novice differences in visual search strategies across sports. Visual Search The ability for expert athletes to extract adva nced perceptual information has been linked to highly developed knowledge stru ctures that are responsible fo r the appropriate allocation of visual attention and enhanced performance. Visu al exploration of the t ypically dense array of perceptual cues is known as visual search, a pr ocess by which the eyes move about the visual environment (i.e., through saccades and smooth-pursui t) in an effort to locate and attend (i.e., fixate) to the most information rich areas. From an information-processing perspective, it is argued that experts derive more task relevant information from each fixation, as oppo sed to lesser skilled performers who require more saccadic movements to gather equivalent information. Saccadic alterations of foveal location are deemed latent periods of in formation processing, from which minimal environmental information is extracted. Thus, the more saccades produced the more evidence there is of inefficient and ineffective search strategies (Williams, Davids, Burwitz, & Williams, 1993). As such, experts search with a high degr ee of visual acuity and efficiency. Without sufficient time to process task relevant cues, ove rsights and incorrect deci sions are inevitable. In addition to typical fixations, the QE period is believed to be a period of time when task relevant environmental cues are processed and motor plans are coordinated for the successful completion of an upcoming task (Vickers, 1996a). The following section will review from an expert-novice paradigm the progression of the visu al search literature, in cluding a description of the common methodology. Finally, a review of the QE literature will be provided, further exemplifying expert-novice diffe rences in gaze behavior.

PAGE 72

72 Eye Movement Registration Knowing where and when to look is a crucial aspect of succ essful sport performance and, as noted above, the visual display is often vast and saturated with information, both relevant and irrelevant to the task at hand. It is therefore imperative that at hletes are able to recognize the central and most information rich areas of the display and direct their attention appropriately (Williams et al., 1999). The awareness that skilled pe rformers possess enhanced perceptual skills relative to information extraction and cue utili zation has been reliably demonstrated across sporting and non-sporting domains a nd has led to further inquiry in to the role perceptual skill acquisition has on the devel opment of expertise. One popular means of assessing pe rceptual skill and subseque nt allocation of visual attention is through the use of eye tracking syst ems such as the Applied Science Laboratories 5000 series (ASL) eye movement measurement sy stems (Williams et al, 1999). These devices are specifically designed to measure fixations and other eye movements by gathering data generated from a light reflected off the cornea, as well as from a video image of the eye (Williams et al., 1999). The location of a visual g aze is typically assumed to index the focus of attention (Duchowski, 2002). When a visual fixation occurs, it is beli eved that a specific area of the environment is being attended to and that th e most detailed, task relevant information is being obtained, while retrieving re latively less detailed informa tion from surrounding peripheral areas. In addition to visual fi xation locations, scan-paths lend themselves to the subsequent analysis and inference of the efficiency by which information is extracted for task completion (Williams, 2000). Efficient and successful performa nce is often characterized by visual search patterns that involve fewer fixa tions of longer duration, a patte rn indicative of the expert performer maximizing the utility of the display and the time available to formulate a response (Williams, 2000;Williams, Davids, & Williams, 1999).

PAGE 73

73 Eye Movement Research The seminal years: 1976-1989 One of the first attempts to determine the visu al search behaviors of basketball players and the first in sport to empirically validate the purported expertnovice visual search differences when solving strategic problems was conducted by Bard and Fleury (1976) using a NAC eye movement recorder. Expert ( n =5,) and novice ( n =5) basketball players were presented with 84 slides depicting one of 28 differe nt typical offensive basketball schematics. At the onset of the stimulus, each participant was required to identif y the type of solution required by the presented offensive strategy, selecting one of seven possibl e answer choices (i.e., shoot, dribble, four passing options, and stay). Two dependent variable s were assessed, decisiontime and number of visual fixations. Experts demonstr ated fewer visual fixations per trial (3.3 vs. 4.9) than novice players. No differences between groups were noted for decision tim e, suggesting that the visual search patterns of experts were more efficien t (i.e., fewer fixations of longer duration) than novice players. Moreover, distinct cognitive strategi es were apparent, with experts concentrating fixations around the pairings of offensive-to -defensive players, while the novice players maintained more frequent fixations to team mates, neglecting the opposition. However, the efficacy of this difference is unknown, since no attempt was made to evaluate performance outcome. As a follow-up to the previous basketball study (i.e., Bard & Fleury, 1976), a series of experiments with basketball and ice-hockey play ers (Bard & Fleury, 1987 ) was conducted using a NAC eye movement recorder to assess the num ber of fixations and fixation durations across tasks of varying complexity. As in the earlier stu dy, basketball players were presented a series of schematics; this time contextual complexity wa s varied. The results dem onstrated that experts

PAGE 74

74 across the three complexity conditions reporte d fewer fixations of longer duration, while reaching conclusions more quickly. However, no attempt was made to assess response accuracy. The second experiment (Bard & Fleur y, 1987) was conducted with ice-hockey goalkeepers. In an on-ice investigation, goalkeepers fitted with an eye-tr acking device (i.e., NAC eye movement recorder) were exposed to a wris t-shot or slap-shot from an experienced icehockey player and required to move in the appropri ate direction in an attempt to block the shot. Results showed that the expert goalkeepers ev oked a quicker reaction (i.e., time from shot initiation to first overt movement to block th e shot) but did not differ from novices on the number of fixations. However, further analysis revealed that although both the experts and novices emitted one fixation per trial per shot, fixation locations differed across skill level and shot type. Specifically, during th e wrist-shot, the novice group fixa ted on the stick significantly more than the expert players, who spent more time fixating on the puck. During the slap-shot, a more complex task, the experts spent significantly more time fixating on the stick and less time on the puck as compared to the novices. In the third experiment by Bard and Fleury (1987), ice-hockey goalkeepers viewed wristshots only. However, in an effort to create te mporal uncertainty, the offensive player was only permitted to initiate the shot after skating with the puck for one, two, or four seconds. Goalkeepers were again assessed on reaction time and number of fixations. Consistent with the earlier findings, the expert players reacted si gnificantly quicker across conditions, with notable performance decrements (i.e., slower reaction time) corresponding with increased temporal uncertainty (i.e., one second vs. four seconds). Al l goalkeepers establishe d attentional preference for the stick and puck, with the expert players de monstrating a significant preference for the stick over the puck.

PAGE 75

75 Although the results of these three investigations support the prediction of expert efficiency (i.e., fewer fixations of longer duratio ns) they must be taken with caution. First, none of the investigations maintained an outcome measure, thus failing to identify the efficacy of the search strategy differences. Second, the temporal brevity of the shooting experiments does not lend itself to directly testing the efficiency hypothesis (i.e., tem poral constraints only permitted one fixation), although clear differences were noted for fixation location. Satisfying the limitations of the previous st udies, Shank and Haywood (1987) assessed the abilities of varsity collegiate ( n =9) and novice ( n =9) baseball batters to predict pitch type (i.e., fastball or curveball), while recording the visual search patterns duri ng the preparatory phase (i.e., wind-up and delivery) of a baseball pitch. Participants viewed a video of a pitcher on a regulation mound delivering fastballs and curveba lls from either the wind-up or the stretch position from a right-hand batters perspective. Eye movements were recorded with an Applied Science Laboratory (ASL) Model 210 Eye-Trac. Results from this investigation support the general findings from the occlusion research. The expert players were more adept at extracting pertinent advance cues compared to the novi ce players, respectively, as signified by the percentage of correctly identified pitches (i.e ., 84.4% vs. 64.3%). Eye movement reaction time was assessed and defined as the moment from the point of ball release to the next eye movement. Results revealed a comparable latency period ac ross groups, suggesting that ball tracking is not occurring and that such quiet time may be optimized by expert s to process and formulate a motor plan (see Quiet-Eye), while novice players maybe dedicating this time to searching for a stimulus response, suggesting that the extraction of advance per ceptual cues almost certainly occurs during the wind-up and the initial stage of the delivery.

PAGE 76

76 Moreover, significant differences in visual search patterns were also noted (Shank & Haywood, 1987). Expert players spent more tim e fixating (63%) and bl inking (26%) and less time searching the visual displa y (11%) than novices (i.e., 53%, 12%, and 35%, respectively). Of further interest are the specific areas of the display most freque ntly fixated on. Experts attended to the pitchers release point whereas the novice players alternated fixations between the release point of the ball and the head of the pitcher. In sum, the work of Shank and Haywood (1987) provided definitive outcome evidence to support vi sual search differences between experts and novices in both the number of fixations and fi xation durations. Although both groups experience post pitch release latencies of approximately 150 ms (i.e., no difference in reaction time) experts appear to have a cognitive adva ntage, as demonstrated by the e ffective use of advance kinematic cues for the accurate percep tion of the baseball pitch. Abernethy and Russell (1987a, 1987b) made extens ive use of the occlusion paradigms as noted above, but questioned the validity of the conclusions drawn from them, citing a lack of objective certainty as to what a nd where an athlete was attendi ng. Using a temporal and spatial occlusion paradigm, Abernethy and Russell (1987b) collected visual search information with the Polymetric Mobile V0165 eye movement recorder while assessing the stroke prediction of expert and novice badminton players. Analysis of the visual search patterns used by the two groups included visual correction time, dwell time, m ean fixation duration, sear ch rate, and location (note; a fixation was defined as any state in which the eye re mained stationary for a period equal to, or in excess of 120 ms p. 289). To is olate cue preferences, fixation locations were divided into five distinct regi ons (i.e., the opponents arm and racquet, the shuttle during out flight, the opponents trunk and body center, the opponents head and face, the opponents legs and feet).

PAGE 77

77 Abernethy and Russell (1987b) demonstrated that both the expert and novice players exhibited early fixations to similar areas of in terest including the opponent s racquet, head, trunk, and to a lesser extent the lower body. However, the frequency of fixations to a given location was proportionately different acro ss regions, with the experts spe nding more time fixating on the racquet/arm complex (46.27%). In contrast, the novice group pref erred the head (28.98%), the trunk (28.74%) and the racquet/arm complex (24.47 %). Subsequent analyses revealed that regardless of the point of previous interest in the visual search sequen ce, experts and novices did not differ on subsequent fixation locations, suggesting that follow ing early fixation preferences, expert and novice badminton player s are not dissimilar in their visu al search patterns. To clarify, the visual search analysis could not attribute perceptual skill di fferences to more effective or efficient search patterns. Simply, the two skill groups allocated their visual fixations to specific display regions similarly, were comparable in th e order in which the cues were fixated on, and were indistinguishable in th eir corresponding search rate. The lack of significant findings obtained by Ab ernethy and Russell (1987b) may not be the result of the small mean visual search differences between skill levels, but rather the large within group variability. For example, the expert performers may no t have been a truly homogenous sample, suggesting that the members of the expert group may not have all been performing at the same level. However, assuming this is not the case, the findings of Abernethy and Russell (1987b) suggest that perceptual e xpertise is not the result of wh at one sees but rather how one uses the information seen. In a multi-study experiment, Goulet, Bard, and Fleury (1989) assessed the search patterns of expert and novice tennis players preparing to return a tennis serv e using eye movement registration techniques (i.e., NAC eye movement recorder, Model V) in Experiment 1 and an

PAGE 78

78 occlusion paradigm in Experiment 2 that was discussed in a previous section. Participants viewed 27 randomly presented tennis serves and iden tified, as quickly as possible, the type of serve (i.e., flat, slice, or top-spin) delivere d. The number of correct responses, number of fixations, scan-paths, and favored exchanges (i.e., direction of one fixation to the next) were assessed and scored according to 11 distinct regions isolated a nd coded for analysis. The serve itself was further divided into three distinct segmen ts, the ritual (i.e., pre cedes serve and includes ball bounce and footwork), preparat ory (i.e., initiation of the ball toss concluding with its apex), and execution (i.e., begins with knee extension an d concluding with ball-racquet contact) phases in attempt to isolate the nature of the adva nce cues used. For response accuracy, experts ( M = 69.9) correctly identified more se rves than did the novice group ( M = 52.2). For search rate, experts displayed more fixations than did novices during the ri tual phase and specifically attended to the head and shoulder/trunk comple x. Search patterns were similar across groups during the preparatory phase, with both groups favor ing the servers head a nd the anticipated ball location. However, differences emerged again during the execution pha se, with the experts terminating their fixations on the racquet mu ch quicker than the novices, while the novice players proceeded to track the ball after ball-r acquet contact. Expert players were apparently more attuned to the kinematic cues presented dur ing the ritual phase, th ereby aiding the decisionmaking process. Moreover, the perception of advance cues corresponded with the decreased decision time inferred from the temporal brevity of the terminal fixation on the racquet as compared to the novice players who sought further information in order to confirm their decision by tracking the flight-p ath of the ball. Empirical and methodologi cal advancements: 1990-1998 The limited and equivocal findings presented in the visual search literature were questioned by Abernethy (1990) who ascribed the lack of skill-based di fferences in his own

PAGE 79

79 research (e.g., Abernethy & Russell, 1987b) to the insufficiently sens itive eye movement registration equipment used and th e potential confound of laboratory based research. In an effort to overcome these limitations Abernethy ( 1990) employed the more sensitive NAC EMR-V Eyemark recorder in two experiments. Experiment 1, was a laboratory inve stigation of 15 expert and 17 novice squash players who viewed 160 trials of four different squash strokes. Each stroke was temporally occluded at one of five time fram es (see the Occlusion section above for a more detailed description of the pro cedure). Most relevant to this discussion however, are the nine separate areas of interest used when compar ing fixation location, duration, and sequence of fixations while anticipating str oke force and direction. A fixation was operationally defined as any case when the eye-mark remained stationary for at least 120 ms. In comparison, Experiment 2 was designed to assess visual search patter ns using a more ecologically valid task while replicating Experiment 1. In this case while positioned on the midpoi nt of the service line of a squash court four expert and four novice squash players viewed 40 squash strokes of varied locations. The findings from Experiment 1 revealed e xpert-novice performance differences both in terms of stroke force and direction (Abernethy, 1990). Anticipation accura cy was not measured during Experiment 2. The findings from the assess ment of the visual search parameters in Experiment 1 revealed expertnovice differences for cue loca tion. Specifically, experts spent more time fixating on the arm and head of the opponent and less time fixating on ball-racquet contact as compared to the novices. This resu lt indicates the novice pl ayers reliance on ball flight cues and the superior ab ility of experts to use advance cues. These results confirm the spatial and temporal occlusion findings repor ted above (Abernethy & Russell, 1987a, 1987b; Isaacs & Finch, 1983; Jones & Miles, 1978). Moreove r, no differences were noted for fixation

PAGE 80

80 duration or number of fi xations across skill levels. The results of Experiment 2, as in previous work by Abernethy and Russell (198 7b), failed to identify skill-bas ed visual search differences. That is, the search patterns asso ciated with fixation distributi on, order, and duration remained stable across skill levels. Abernethy conclude d that the notable performance differences exemplified in Experiment 1 were not the result of more efficient search strategies but rather the ability to make strategic inference from the info rmation extracted. Again, the experts were better able to make use of the advance cues, while th e novice players relied more on the ball-racquet contact zone and subsequent ball flight information. The majority of research reviewed has relied on interceptive rac quet sports and advance cue use for the prediction of stroke type a nd location. Extending the literature beyond racquet sports Helsen and Pauwels (1990) examined the be haviors of soccer players. Fifteen expert and 15 novice soccer players were presented with 90 slid es of typical offensive soccer situations. The participants were required to view the slide and as quickly as possible verbalize the most correct decision (i.e., shoot, dribble, or pass). Visual search behavior was recorded using a NAC-V Eyemark recorder. The findings from this investig ation indicate that e xperts were faster and more accurate in finding tactical solutions to offe nsive soccer situations. In accord with previous findings (e.g., Abernethy 1990a; Abernethy & Russe ll, 1987a), there was li ttle difference amid the locations of fixations between skill groups. However, in contrast to previous work (e.g., Abernethy 1990a; Abernethy & Russell, 1987a), analys is of the visual search data for number of fixations revealed expert-novice differences. Experts displayed fewer fixations than novices. These results are a sign of the im proved ability of experts to ma ke use of the information they possess, while novices continue to seek out validation. As endorsed by Helsen and Pauwels (1990), fewer fixations allow for faster inform ation-processing time and result in shorter

PAGE 81

81 response time, equating to better performance. Essentially, fewer fixations place less demand on long-term working memory and experts can retrie ve an appropriate response to the problemspace more proficiently with the use of their extensive knowledge base. Returning to racquet sports, Cauraugh, Singer, and Chen (1993) investigated the visual search patterns and anticipation strategies used to predict an opponents tennis stroke. Expert ( n =30) and novice ( n =30) tennis players viewed two different types of tennis serves (i.e., spin or flat) and had to identify its directi on (i.e., left, right, or center cour t) as quickly and accurately as possible. An Eye-Trac, Model 210 was used for recording eye fixa tions. A fixation was operationally defined as a stati onary eye for a minimum of 133 ms. The visual search data revealed significant expert-novice differences. Sp ecifically, of the nine predetermined areas for fixations, the novice players demonstrated a prefer ence for the head and left shoulder, with no other notable differences with experts present. Additionally, experts dedicated more time per location, that is, their fixations were of longer duration than the novice players. When considering the visual search patterns across the thr ee phases of the serve (i.e., preparation, execution, and follow-through), the novice players fi xated more frequently on the head during the preparation phase. Similarly, skill-based diffe rences were evident for ground strokes with the novice players fixating more frequently on thei r opponents hips. No other visual search differences between groups were found. However, consistent with previous findings, the experts were both quicker and more accurate in predic ting the type and location of their opponents shots. Despite the promising results of these studies in terms of id entifying expert-novice visual search differences, with the exception of Helsen and Pauwels (1990), research reviewed to this point has involved inter ceptive racquet sports and the correct identification of stroke force,

PAGE 82

82 location, or type. The nature of the scan-paths and the extent of the expert-novice differences cannot be generalized to closed-tas ks or those that require precis e aiming at a far target. Vickers (1992), interested in the systematic changes in visual search as aiming accuracy improved, assessed low handicap (i.e., 0-8) an d higher handicap (i.e., 10-16) tournament caliber golfers. To determine eye movements and gaze behaviors, part icipants were fitted with a mobile ASL 3001H Eye View recorder, while executing consecutive pu tts from a distance of 3m. Participants were required to continuously putt in sets of twelve unt il ten successful (i.e., hit) and ten unsuccessful (i.e., misses) putts were recorded. For coding an d analysis of the putt, three distinct stroke segments (i.e., preparation, backswing/foreswing, a nd contact phase) and six gaze locations (i.e., feet, ball, club head, cup, and putting surface) were analyzed for fixa tion duration (stabilized gaze for a minimum of 99.99 ms), saccades (eye movement from one location to another with a duration of 133.2 ms), express saccades (rapid shifts lasting 66.6-99.9 ms) and smooth pursuit (tracked object for 99.9 ms or more) movements. Th e results of this inves tigation revealed that the low handicap (LH) golfers used significantly fewer fixations (14.2) per putt than the higher handicap (HH) golfers (19.4), while total putt time was indisti nguishable. In-line with this finding, LH golfers maintained fewer fixations of longer duration duri ng the preparation and backswing/foreswing phases. Furthermore these results revealed that better golfers employ variable gaze behaviors, including more e xpress saccades between fixation locations and fixations of longer duration to the ball and ta rget, while minimizing gaze behaviors to the club head and putting surface. Conversely, the HH golfers dedicated significantly more gaze behaviors (fixations, saccades, et c.) to the putting surface and feet and were more consistent in their gaze behaviors across stroke phase and fixation location.

PAGE 83

83 The work of Vickers (1992) provides direct evidence to support the importance of visual gaze behaviors in an aiming task. Specifically, although advance visual information is imperative for the successful anticipation of an opponent s stroke, informati on gathered during the preparation and backswing/foresw ing phase of a golf putt is equa lly important. Vickers (1992) postulates that fewer fixations of longer duration allows for th e coordination of the eyes and hands, synchronizing the visual informa tion with a distinct motor plan. The paucity of visual search research in open-tasks coupled with many conceptual and methodological limitations (e.g., use of static slides, scaled image presentation, brief film clips, and small sample size) has necessitated the need for the further exploration of skill-based visual search differences in open-skill sports (Willia ms et al., 1994). Williams et al. (1994) conducted an examination of perception and decision-making in sport, assessing skill-based differences in visual search strategies, a nd the corresponding response time and response accuracy while viewing 11-on-11 soccer situations An ASL 4000 SU eye moveme nt registration system was used to record visual search strategies acro ss 13 distinct soccer scenarios from a defensive perspective. Filmed scenarios were displayed on 3m x 3m projection sc reen. A reference grid was imposed on the field to facilitate scoring an d to provide a point of distinction for verbal responses indicating the location of the final pass destination. Results indicated skill-based differences for response time, fixation location, nu mber of fixations, and fixation duration (Note: a fixation was defined as a condition in which the eye remained stationary for a period equal to or in excess of 120 ms). Specifically, although no differences were evident for response accuracy, expert players responded much more quickly, confirming the use of advance perceptual information. Williams et al. (1992) derived a response time latency of 200 ms, indicating that the skilled players were able to anticipate final pass dest ination prior to foot

PAGE 84

84 strike, and consistent with the occlusion re search, the novice player s relied on ball-flight information. Furthermore, contrary to the notion that experts exhibit a more efficient search strategy, it was demonstrated that the experts performed a more el aborate search of the display, frequently shifting gaze from a central location to other information areas and back (e.g., more fixations of shorter duration). A lthough appearing less efficient, Williams et al. (1992) suggested that the broad scope with numerous important elements (i.e., 11-on-11), required such an elaborate search process. To compare, the novice players spent more time watching the ball while the experienced players spent more time gathering information away from the ball, facilitating pattern recognition a nd decision-making proficiency, be haviors indicative of highly skilled players. It can therefore be concluded that the task itself may be equally as important, if not more influential on the search strategies empl oyed, as compared to the level of expertise and skill proficiency alone. The potential for task type and complexity to inversely influence ski ll-based visual search differences was examined by Ripoll, Ker lirzin, Stein, and Reine (1995). Expert ( n =6), intermediate ( n =6), and novice ( n =6) kick-boxers viewed video-re corded images of an opponent portraying different tactical mane uvers (i.e., attacks, openings, or feints) that required a quick and accurate response to block or counter successf ully. The participants used a joystick to react to their opponents maneuvers, while simultaneously recording response time and accuracy. Two experiments were conducted. The fi rst experiment assessed response time and response accuracy only, while the second experiment explored skill-based differences in visual search behaviors using a NAC-V Eyemark recorder. To further clar ify, for Experiment 1, three levels of task complexity (i.e., simple A, simple B, and comp lex) were introduced. During condition simple A, the participants were explicitly in structed to respond to attacks (i.e., punch or kick) and to ignore

PAGE 85

85 other maneuvers. In simple condition B, participan ts were to respond to openings while ignoring the other two possibilities when present. The co mplex condition required the participants to react to both attacks and openings and ignore feints. In Experiment 1, the si mple conditions A and B failed to reveal skill-based diffe rences in response accuracy and response time. It was concluded that expertise does not discriminate between skill levels during a simple task. However, when the participants had to respond to attacks and opening s (complex task), experts were more efficient (i.e., time and accuracy) in their responses as compared to both the intermediate and novice groups. Surprisingly, the novice group outperformed the intermediate group. The findings of Ripoll et al. ( 1995) suggest that the expert boxers were able to recognize the type of movement initiated by the opponent, signifying the pot ential for a more successful counter. Unfortunately, this findi ng was not maintained in a co mparison of the lesser skilled groups. In conjunction with the efficiency diffe rences between the experts and less skilled boxers, skill-based differences for the accuracy of response were revealed. Again there were no differences between the intermediate and novice boxers. It was postulated that the lack of performance differences was due to the homoge neity of the intermed iate and novice groups. Although the novice group did not have any fo rmal combat experience, their practice experiences may have been equivalent to that of the intermediate gr oup (Ripoll et al., 1995). Nevertheless, these findings are suspect. Follow ing Experiment 1 Ripoll et al. (1995) concluded that expertise appears only in the presence of a complex problem space. In an extension of the previous study, Experime nt 2 explicitly assesse d skill-based visual search differences in complex conditions (Rip oll et al., 1995). For the purpose of off-line analysis, the visual display was divided into six distinct regions, includ ing: the (a) head, (b) trunk, (c) arm/fist, (d) pelvis, (e) legs, and (f ) other unidentified fixa tions. The results of

PAGE 86

86 Experiment 2 revealed skill-based visual sear ch differences for number of fixations, fixation location, and fixation duration. Specifically, ex perts (43.3) completed fewer fixations as compared to both the intermediate (105.8) a nd novice (122.67) groups. Moreover, consistent with Williams et al. (1994), task complexity evoke d a more systematic search of the display as evidenced by the experts use of a visual pivot. That is, the experts spent the most time fixating and used the head as a point of reference from which subsequent eye movements originated from and returned to. In contrast, the less skilled boxers favored th e arm/fist complex and the trunk (see Figure 2-5). However, unlike soccer players who showed more fixations of shorter duration as a function of task complexity and expertise (Williams et al., 1994), the work of Ripoll et al. (1995) is consistent with the noti on of visual search efficiency, in that the experts maintained fewer fixations of longer duration, a finding that may be attributab le to the less complex visual display. Furthermore, efficiency may have resulted from the rich knowledge base of the experts that facilitates the matching of a stimulus to a response process. A second, alternative explanation attends to the imminent threat imposed by the sport of French boxing, in which the defender cannot afford to miss a perceptual cue and thus relies more heavily on the peripheral system as opposed to the ra pid shifting of vi sual attention ar ound the display. Returning to racquet sports, Singer, Ca uraugh, Chen, Steinberg, and Frehlich (1996) advanced the work of Goulet et al. (1989) using a simulated tenni s experiment to investigate the visual search, anticipatory behaviors, reacti on time, and decision accuracy of highly-skilled ( n =30) and beginner ( n =30) tennis players. An ASL Mode l 210 Eye-Trac monitoring system was used to record the visual search tendencies of participants while viewi ng a video presentation of serves and ground strokes. Participants were inst ructed to respond as quick ly and accurately as possible to the type of serve (i.e., flat or spin ) and location (i.e., left, right, or center) of ball

PAGE 87

87 placement for each stroke. Anticipation time for each serve was measured using a monitoring system that was activated when the video m odel initiated the ball toss sequence. Similarly, anticipation time was measured for ground stroke s, beginning when the ball bounced on the opponents side of the court in itiating the ground stroke. Deci sion speed for serve type was determined with the participant pressing one of tw o switches with either th eir index finger (flat) or middle finger (spin). Particip ants used a joystick to manipul ate with their dominant hand the anticipated direction of bot h serves and ground strokes. The results of Singer et al. ( 1996) confirmed expert-novice visu al search differences, with the novice players fixating longer on the head duri ng the serve compared to the expert players. No significant skill-based diffe rences were present for ground strokes. Finally, anticipation measures (i.e., speed and accuracy) for the serve and ground strokes revealed that experts were faster and more accurate in their responses than novices. In addition to the quantitative analysis, Singer et al. (1996) conducted a qualitative analysis of the visual search patterns of the two best expert players and two ra ndomly selected novice players. The results of the qualita tive analysis revealed that duri ng the serve, the expert players tracked the ball as it was tossed until the point of contact, at which point they refocused visual attention on the racquet and arm re gion of their opponent, and subse quently tracked the flight of the ball. The novice players were less systematic in their search behaviors until ball-racquet contact was made, at which poi nt the ball was tracked. During ground strokes, the experts focused on the hip region, followed by the racquet, racquet-ball contact, and subsequent tracking of the balls flight path. Again, the novice players were random and unsystematic with a common interest in tracking the ball post-contact. These findings are consistent with the majority of the spatial occlusion research suggesting that experts make better use of advance and distal

PAGE 88

88 cues as compared to the novice players. Unique to this qualitative analysis however, was the experts frequent use of the hips as a primary source of information during the ground stroke. It can be speculated that such a difference emerge s during ground strokes as result of the close proximity of the arm-racquet complex to the hi p region. Furthermore, during the ground stroke the hips may be a better indicati on of stroke direction; however, fu rther research is warranted to explore this possibility. Williams and Davids (1997) conducted two inve stigations to examine the relationship between visual search behavior and concurrent verbal reports in a test of advance cue use and decision-making in soccer. Experiment 1 was a rep lication of the 11-on-11 anticipation test used in the previous work of Williams et al. (1994). However, in this investigation, participants verbalized the location of th eir attention throughout each viewing sequence. An ASL 4000 SU eye movement registration system was used to record visual search patterns of experienced ( n =10) and less experienced ( n =10) soccer players across a tota l of 26 soccer action sequences presented on a 3.5m x 3m projection screen. A reference grid was imposed on the field to facilitate scoring and to provide a point of di stinction for the verbal responses indicating the location of the final pass destin ation. Results indicated that no expert-novice differences were evident for reaction time or re sponse error. Consistent with the previous findings (e.g., Abernethy & Russell 1987b; Goulet et al., 1989), no skill-based diffe rences for search rate were revealed, yet search order differences were not ed. Specifically, the expert players displayed a systematic search strategy, main taining a visual pivot (i.e., from the box to other areas and back to the box); however no differences were noted for the verbal reports of attention direction. Although the expert athletes fixa ted more frequently to the box, the novice players sustained

PAGE 89

89 their attention within the box for longer periods, signifying their inability to differentiate the relevant from the extraneous. Experiment 2 (Williams & Davids, 1997) was a modified version of the 11-on-11 protocol used in Experiment 1, examining the relationship between concurrent verbal reports and visual search during a less complex 3-on-3 soccer tas k. Twelve experienced and 12 less experienced soccer players viewed 20 offensive soccer sequences Participants were requi red to anticipate the pass destination as quickly and accurately as po ssible and were again required to verbalize the location of their attention throughout each viewi ng sequence. The results from Experiment 2 contrasted those from Experiment 1; although both groups were e qually successful in predicting the final outcome, the experienced players res ponded quicker. No differences were noted for search order in either the eye movement condi tion or the verbal report condition. Skill-based differences for fixation location were revealed, mirroring the re sults of the 11-on-11 conditions. Specifically, the novice performers spent more time fixating within the box, while the experienced players used the box as reference point, frequently directing thei r attention to the left and right sides of the display. Williams and Da vids (1997) concluded that the inexperienced players were ball-watchers, again identifying the inability of novice players to differentiate relevant from extraneous cues. The work of Williams and Davids (1997) sugge sts that task complexity may not be a significant mediating variable in the search st rategy of experienced and inexperienced soccer player as indexed by the similar findings in 11-on -11 and 3-on-3 soccer situations. However, the nature of the sport may influence such behavior In three experiments (i.e., Williams & Davids, 1997, Experiment 1 and 2; Williams et al., 1994 ) of perceptual decision-making, experts employed more fixations as compared to the less -skilled players. Additionally, experts adopted a

PAGE 90

90 systematic search with a common area of intere st, a pattern neglected by the more variable approach of the inexperienced players. This patt ern of results contradict s the findings of Goulet et al. (1989), Helsen and Pauw els, (1993), and Ripoll et al. (1995) who found that experts maintained fewer fixations of longer duration. It further appears that novice athletes have the knowledge of what they should be doing, but ar e unable to extract the relevant information necessary to enhance performance as indicated by the verbal protocols used here by Williams and Davids, (1997). Williams and Davids (1998) conducted an exte nsion of their previous work on complex (11-on-11) and less complex (3-on-3) decision-maki ng in soccer with an examination of visual search, anticipation, and expert ise using three complimentary methodological approaches. Experiment 1 examined skill-based visual sear ch and anticipation diffe rences during 3-on-3 and 1-on-1 offensive situations. Experiment 2 used a similar protocol with th e addition of a spatial occlusion condition. Experiment 3, compared the scan-paths and the concurrent collection of verbal reports of participants across 3 levels of task complexity (i.e., 11-on-11, 3-on-3, and 1-on1). An ASL 4000SU eye movement regi stration system was use to reco rd visual search behavior. Experiment 1. Williams and Davids (1998) compared 12 experienced and 12 less experienced soccer players who viewed 20 offensiv e soccer sequences in which they acted as a defender, responding as quickly and as accurately as possible by stepping onto a response pad (i.e., left, right, front, or back ) indicating the anticipated direction and final destination of the ball. The results demonstrated the experts superi or ability to anticipate and respond quicker than the less-skilled players during both the 3-on3 and 1-on-1 conditions. However, the highlyskilled players were only more accurate in their predictions during the 1-on-1 task. Moreover, the results of the visual search data revealed no skill-based differences in the allocation of fixations

PAGE 91

91 to regions of the display, search order, or s earch rate during the 3-on-3 conditions. Conversely, visual search differences in the 1-on-1 conditio ns were evident. Specifically, the experts had a tendency to pivot (i.e., more fixa tions of shorter duration) their visual attention between the ball and their opponents hip region compared to th e less-skilled players. Although no statistical differences were reported for fixation location, th e experts spent more time fixating on the area between their opponents knees and chest. Williams and Davids (1998) reported an effect size of 1.08 signifying the practical significance of a me an difference of this magnitude for soccer players in a 1-on-1 confrontation. Experiment 2. Williams and Davids (1998) used the same procedure and participants as those reported in Experiment 1 with the exception of 2 additional spatial o cclusion conditions for the 3-on-3 task, and four spatial occlusion conditi ons for the 1-on-1 task. Specifically, for the 3on-3 anticipation test th e spatial occlusion conditions consis ted of (e1) the removal of all irrelevant perimeter information by cropping the visu al display to include only the positions and movements of the players and (e2) a reduction of the visual displa y to include only the ball or the ball passer. For the 1-on-1 anti cipation test, the spatial occlus ion conditions consisted of the removal of (e1) the opponents head and shoulders; (e2) the hips; (e 3) the lower leg; and (e4) an irrelevant area of the display. The results from the 3-on-3 conditions showed significant performance decrements across occlusion conditions for the skilled group (i.e., from (e1) to (e2)) and stable performance across conditions for th e less-skilled players. Nevertheless, the highlyskilled players maintained their advantage over the novices performing quicker and more accurately under both occlusion conditions. Th e results of the1-on-1 occlusion condition revealed a similar performance advantage for experts across occlusion conditions; however, no statistical significance was reported.

PAGE 92

92 Experiment 3. Williams and Davids (1998) used the sa me participants and test procedures as in Experiment 1 with the addition of obtaining concurrent verbal reports Consistent with the findings of Experiment 1, the highly-skilled play ers had quicker response times but were equaled in response accuracy by the less-skilled players. Furthermore, both groups reported similar patterns of attention allocation to the left and right sides of the box, yet the less-skilled players spent significantly more time fixating within th e box (i.e., ball and or ball carrier) than the highly-skilled players. No further differences were apparent to search order, yet the experienced players fixations were more e qually distributed across the diffe rent viewing areas (i.e., left, right, and box), with the less skilled pl ayers spending more time within the box. It can be concluded from the work of Williams and Davids (1998) that skilled players are able to make sufficient use of advance performan ce cues, reducing the time necessary to derive a conclusion and respond. More impor tantly, skilled play ers apparently rely less on information provided by the ball and more on the informati on presented throughout the visual display as evident by the frequency of fixations directed at the opposing players hip region during the 1on-1 conditions. Additionally, the di screpancy between verbal report s of frequent alternation of attention allocation and visual fixations of equal distribut ion during the 3-on-3 conditions suggests that skilled players rely more on nonball specific informati on gathered through the peripheral system. Contemporary investigations: 1999-2002 Task type and complexity have both b een points of contention for eye movement researchers (e.g., Ripoll et al., 1995; Williams & Davids, 1997). However, the extent to which performance related anxiety mediates the visual search process has been neglected. Attending to this gap in the literature, Williams and Elliot ( 1999) examined the effects of cognitive anxiety and level of experience on anticipation and visual search behavior in karate kumite. In accord

PAGE 93

93 with the Processing Efficiency Theory (PET; Eysenck & Calvo, 1992), Williams and Elliot (1999) postulated a positive relationship between anxiety, number of visual fixations, and fixation location. Simply, the major tenet of the PET states that as arousal, and more specifically anxiety increases, the breadth of the attention system narrows, rendering information in the peripheral environment neglected or when searched, searched in a more deliberate and inefficient process. Therefore, Williams and Elliot (1999) pred icted that as anxiety increased, so too would the number of fixations to the periphery of the vi sual display as a result of attentional narrowing. Experienced ( n =8) and inexperienced ( n =8) martial artists viewed a series of film sequences depicting both fist and foot st rikes and were required to physi cally respond as quickly and as accurately as possible as if acti ng to avoid the impending strike. Three trained judges were used to assess the accuracy of the pa rticipants responses under low-a nxiety (neutral statements) and high-anxiety conditions (competition incentive). Visual search beha viors were examined with an ASL 4000SU eye movement recorder, with particular attention given to s earch rate and fixation location (i.e., head, chest, shoulders, pelv is, arm/fist, leg/foot and unclassified). For the dependent measures of response accu racy and viewing time, the experts were shown to be more accurate in th eir response to the impending strikes, but did not differ from the novice group in the amount of time they spent viewing the display before responding (Williams & Elliot, 1999). The results of the visual search data failed to demonstrate group differences in fixation location. In spite of the l ack of statistical signi ficance, a trend was pr esent indicating that the experts spent proportionately more time fi xating on the head (45.2% vs. 34.8%) and pelvis (12.9% vs. 4.6%) and less time fixating on the chest (24.3% vs. 34.9%) and arm/fist (7.0% vs. 17.5%) locations as compared to the inexperienced performers. In response to the alterations of cognitive anxiety, both groups deviated from their typical viewi ng patterns under the high

PAGE 94

94 anxiety condition as compared to the low anxiety condition, with an increase in fixations to both the shoulder and arm/fist regions, suggesting th at the experienced performers were no more resilient to the effects of anxiet y than were the inexperienced pe rformers. Search rate was also altered by modulations in anxiet y across skill levels. Specifically, in the high anxiety condition the experienced group showed an increase in the mean duration of their fixations (236.53 ms vs. 219.61 ms) while the inexperienced groups fi xation duration decrea sed (250.58 ms vs. 284.44 ms) in comparison to the low anxiety condi tion. Additionally, both groups exhibited more fixations and more fixation locations under the high anxiety condition as compared to the low anxiety condition, suggesting that as anxiety levels increase th e visual search profiles of experienced and inexperienced perf ormers become more similar. The ecological validity of laboratory studies was questioned earlier in the review with respect to occlusion-based research. A primary c oncern was the inability of static slides to capture the essential display featur es of real world tasks. In re sponse, a methodological shift to the use of film and action sequences occurred; however, the concurrent use of both mediums within the same experiment with the same samp le had not occurred. In an effort to assess the relative importance of the perimetric and optimetri c parameters and visual search in perceptual decision-making in soccer, Helsen and Starkes (1999) conducted a multi-method investigation of skill-based differences. A three-study approach wa s used to examine the mediating variables of skilled perception. Experiment 1 assessed visual ac uity and other hardware issues of perception and will not be discussed further. Experiment 2, using a static slide presentation, assessed the skill-based visual search and an ticipation differences of expert and intermediate soccer players during 11-on-11 open soccer play. In Experiment 3, f ilm rather than static slides were presented

PAGE 95

95 to participants while maintaining the same protocol as in Experiment 2. A NAC-V eye movement registration system was use to record visual search behavior. Experiment 2 (Helsen & Starkes, 1999). Expert ( n =14) and intermediate ( n =14) soccer players viewed 30 slides depicting a typical offensive soccer situa tion from the perspective of the ball carrier. Upon slide presentation, the partic ipant was to respond verbally as quickly and accurately as possible, identifying the most appr opriate move (i.e., shoot, dribble, pass). The results of Experiment 2 indicated that the expe rt group performed better than the intermediate group for trials in which dribbling and passing we re the appropriate resp onses for both response time and accuracy. No skill-based differences were revealed for shooting scenarios. The results of the visual search data demonstrated that experts maintained fewer fixations than the intermediate players but did not differ in the duration of fixations Analysis of the scan-paths during correct decisions did not reveal any group differences. However, skill-based differences were illustrated when consid ering the scan-paths during incorrect decisions. Specifically, the expert group spent more time viewing the ball carrier, reducing the amount of time spent in searching other areas, while th e intermediate group spent less time fixating on the goal. The truncated search pattern of the experts during incorrect decisions reflects the behavior more characteristic of novice players who allocate their attention to a limited focal point, which is often the ball rather than using a more systema tic proportional dist ribution of fixations across the visual display. Experiment 3 (Helsen & Starkes, 1999). The procedure and partic ipants were the same as those reported in Experiment 2, with two m odifications. First, the simulation consisted of 30 different tactical soccer situations that were projected life-size using a dynamic film approach. Second, the participants performed a tactical decision with the ball (i.e., shoot, dribble, or pass)

PAGE 96

96 in response to the film, as quickly and as accura tely as possible. The results of Experiment 2 revealed that experts were fa ster and more accurate in res ponding to the various tactical problems. Specifically, the experts demonstrated faster initiation tim e, ball/foot contact time, and total response time. Of the three areas in whic h the experts demonstrated superiority, initiation time offers an explanation for the ability of the expert group to make use of advance cues, whereas the differences in ball/foot contact and total response time may be accounted for by physical skill differences. The results of the visu al search data mirrored that of Experiment 2. Specifically, experts maintained fewer fixations th an the intermediate players, but they did not differ in the duration of fixations Finally, during the dynamic task the expert players were more inclined to make use of the defenders, free back, and free space as compared to the intermediate performers, who fixated more on the ball and the goal. The work of Helsen and Starkes (1999) suppor ts the notion that experts process advance cues, thereby reducing response time in both film and static soccer presentations. Furthermore, visual search differences were evident by the fe wer fixations performed by the experts, indicative of an efficient search process, and providing su pport for the differential cue accessibility and processing capabilities of expert performers. Savelsbergh, Williams, Van Der Kamp, and Ward (2002) conducted a study examining the skill-based differences in anticipation and visual search during the penalty kick in soccer. An ASL 4000SU eye tracker and an Ascension Technol ogies magnetic head-tracker were used to record the visual search behaviors of expert ( n =7) and novice ( n =7) goalkeepers while viewing a presentation of 120 penalty-kick tr ials. The participants were situ ated behind a joystick that was used to indicate and record both response tim e and accuracy. Accuracy was determined by the placement of the joystick relative to the position of the incoming so ccer ball. To further clarify, if

PAGE 97

97 the position of the joystick intercep ted the coordinates of the flight path of the soccer ball at the time it crossed the goal-line a save was declar ed. Therefore, accuracy was determined by the (a) number of total saves; (b) co rrect side (i.e., correct lateral prediction, but incorrect vertical prediction); (c) correct height (i.e. incorrect late ral prediction, but correct vertical prediction); and (d) proportion of corrections (i.e., percenta ge of corrective movements). The visual search data were assessed for search ra te, and percentage of viewing ti me distributed to the following fixation locations: head, trunk, s houlders, arms, hips, kicking leg, non-kicking leg, ball, and unclassified. The anticipation tests revealed no significant skill-based differences for save percentage (Savelsbergh et al., 2002). However, it is possibl e that the secondary measures of accuracy may be more indicative of skill-based differences. Spec ifically, the expert goal keepers predicted shot height and direction more accurately, committed fewer corrective movements, and began their responses closer to ball-foot contact than th e novice goalkeepers. The nature of the contrived task may therefore, have contributed to the la ck of significant accuracy differences. Simply, the novice performers committed mo re corrective movements throughout the sequence, suggesting that more information was required for these athletes to confirm their decision, a luxury unavailable in a real -world condition. Secondly, experts were better able to anticipat e the height and direction of shots more accurately and initiate their responses just prio r (300 ms) to ball-foot c ontact, suggestive of the effective use of advance cues (Savelsbergh et al., 2002). In contrast, th e novice group initiated their response 500 ms prior to ball-foot cont act. Although such a brief differential existed between the expert and novice pl ayers, the early response time by the novice players may be more suited to a guess, whereas th e expert players are able to ma tch the advance cues with an

PAGE 98

98 appropriate response in the brief ti me period. The results of the visu al search data revealed that the expert goalkeepers performed a more effici ent and less exhaustive search of the visual display. That is, they fixated on fewer areas of the display while using fewer fixations of longer duration. Moreover, the experts garnered more information from the kicking leg, non-kicking leg, ball, and head regions, as compared to th e novice performers who spent more time fixating on the trunk, arm, and hip regions. Overall, thes e results demonstrate the importance of highly reactive participants to acquire perceptual informa tion early in the display in order to process and formulate a correct motor response. In a unique study of the advance cues used in a motor task, Byrnes (2002) examined the visual search behaviors of expert ( n =5), intermediate ( n =5), and novice ( n =5) equestrian riders. An ASL 5000 eye tracking system was used to record visual search behavior s as the participants viewed a computer simulation of the fences located throughout an equestrian course. Participants were not constrained in any manner, therefore they were free to l ook at the course from a variety of angles and perspectives for as long as they desired. The visual search data were assessed for number of fixations, fixation duration, and lo cation of fixations. The results from the investigation revealed that expert riders made more fixations across fences than both the intermediate and novice riders yet no differences were evident between the intermediate and novice groups. Furthermore, no skill-based diffe rences were evident for fixation duration. Fixation locations were analyzed according to predetermined areas of interest. The expert riders directed significantly more fixations within the predetermined areas of interest as compared to both the intermediate and novice riders, yet no di fferences were found betw een the intermediate and novice riders. In conclusion, the work of Byrn es (2002) signifies the meticulous nature of

PAGE 99

99 expert riders, as indicated by th e exhaustive nature in which they search a display in preparing for competition. The relationship between visual perception, an ticipation, and visual search behaviors of experienced ( n =8) and novice ( n =8) tennis players was assessed by Ward, Williams, and Bennett (2002). Participants stood in front of a 3m x 3.5m life-size projection screen which depicted normal and point light displays of tennis ground strokes. From a theo retical perspective, Ward et al. (2002) postulated that experi enced tennis players are more att une to perceptual cues than inexperienced performers. Therefore, when the pe rceptual display was modified to emit minimal detail (i.e., point light display), both response a ccuracy and visual search behavior of the novice performers was believed to suffer, yet the experi enced players were expected to portray similar visual search behaviors while responding comp arably across displays. Response accuracy was assessed by physically responding to the anticipa ted direction of the impending stroke. Two pressure sensitive mats were designated as the starting position, with four additional mats placed around the participant that were us ed to record the response direc tion. Visual search data were recorded using an ASL 5000 eye movement tracking system. Anticipation data revealed skill-based diffe rences for decision time, with the experts responding quicker (Ward et al., 2002). No skill-based differences were evident for response accuracy between groups in either condition. However, a performance decrement of 9.9% was noted under the point light display as compared to the normal display for the experienced group, whereas the inexperienced participants response accuracy score remained stable. This result suggests that the lack of detail is more detrimenta l to experienced athletes. The visual search data revealed that regardless of th e viewing condition, the experienced players spent more time fixating on the head-shoulder and trunk regions as compared to inexperienced players, who spent

PAGE 100

100 more time attending to the racquet area. Search rate data revealed a decreas e in the total number of fixation locations and total fixations whereas fixation duration increased for both groups while viewing the point light display. This result confirms the diffi culty for both experienced and inexperienced individuals to ex tract perceptual cues from poi nt light displays. Although no between group differences were noted, search order revealed that experienced players emitted successive fixations around the torso more than the inexperienced group. Moreover, the inexperienced group altered their fixation patterns when viewing the point light display, fixating more to the racquet, ball, and ball-racquet areas Conversely the experien ced players were more resilient, maintaining their preferred search orde r even when viewing the point light display. The work of Ward et al. (2002) provides s upport for the notion that experienced players possess the ability to extract re levant performance cues even from the most sparse arrays. Although, response accuracy diminished marginally while viewing the point light display, decision time was stable across c onditions (Ward et al., 2002). Visual search summary and review In the wake of previous me thodological and inferential li mitations (e.g., temporal and spatial occlusion) in the study of expert-novice perceptual diffe rences, more recent technological advances, such as the advent of eye movement registration t echniques were thought to provide greater insight into the cue utilization and deci sion making process of skilled and less-skilled performers. Visual search data were further used to draw conclusions ab out attention allocation and more specifically selective attention. It was commonplace that expe rts would require less perceptual information as a result of their ex tensive knowledge base permitting the detection of stimulus redundancy (Abernethy, 1988). Unfortuna tely, the equivocal findi ngs across a variety of sport tasks and furthermore task complexity has generated more que stions than initially answered. Specifically, temporal st ress, whether inherent (i.e., in creased response uncertainty) or

PAGE 101

101 imposed (i.e., experimental control), appears to influence the relati onship between fixation duration and the total number of fixations (Abernethy, 1988, 1990b). However, the practical significance of the syst ematic findings that experts perform much quicker than novices is significan t. Sport by nature requires d ynamic thinking and split second decision-making. Clearly the use of advance per ceptual information aids in this process. Although the visual search behavi ors of expert athletes have been equivocal across tasks of varying type and complexity, the issue may not be at all be re lated to information sought via foveal fixations but rather a result of the in formation collected from the peripheral system (Abernethy, 1988; Williams et al., 1993; Williams et al., 1999). More direct research is warranted assessing the peripheral vision differences in expert and novice performers and their respective abilities to gather, process, and translat e that information into enhanced performance. Moreover, experts may engage in an adva nced level of preattentive processing as suggested by Neisser (1967) therefore constructing a priori strategies to facilitate the decisionmaking process. To further clarify, a common findi ng throughout the visual search literature was comparable response accuracy between skill leve ls, however expertise was indicative of faster decision-times suggesting that certain cues w ould appear to pop-out to the expert while diligent pursuits were required by the less-skilled athletes. Nevertheless, this narrative review has provided inclusive evidence as to the distinct visual search tendencies unique to athlet es of varied skill level. Caut ion should be taken however, when attempting to infer even from the equivocal findin gs reported here. The very nature of the tasks assessed coupled with the incons istent use of experts and novices sufficiently impedes the ability to decisively conclude skill-based visual search differences. In response to this limitation a quantitative analysis of this literature base is warranted.

PAGE 102

102 Finally, it can be concluded that eye moveme nt registration techniques cannot ensure an accurate understanding of the information extracted from the visual display despite the ability to index the location of the eye rela tive to the visual display. As Abernethy (1988) indicates the capacity for athletes to shift th eir attention without additional eye movements fosters the need for a distinction between looking and seeing. To confound this problem fu rther, little is known of the function peripheral vision assumes for information extracti on during the visual search of a dynamic visual display. In sum, eye movement registration techni ques have furthered the understanding of the visual search strategies of athletes. Moreover, as technology continues to push the boundaries of applied research, more definitive ecologically valid conclusions may be drawn as to the performance differences present during actual competition. Quiet Eye Although traditional eye-tracking research has been limited to the measurement of the number, duration, and frequency of visual fixations, the location of fixations, visual pivots, and saccades, more recent trends have focused on a measur e that is believed to be associated with the organization of visual attenti on and information processing, the QE period. Specifically, the quiet-eye period is a measure defined as the elapsed time between the last visual fixation to a target and the initiation of the motor response (V ickers, 1996a). For example, the QE period of a marksman is identified as the elapsed time from the last recorded visual fixation to the commencement of the trigger pull. Literature examining the QE period and whether or not the QE period can mechanistically account for notable expert-novice performa nce differences is the focus of this section of the review. From a theoretical perspective, Vicker s (1996a) proposed the location-suppression hypothesis to account for the requi site behaviors necessary for aiming. For example, consider

PAGE 103

103 tasks such as baseball pitching, archery, dart throwing, pistol or rifle shooting, basketball freethrow shooting, and even billiard s. These tasks require the iden tification of a target in the environment through visual search, followed by the successful merging of environmental coordinates and movement to accomplish the goal successfully. Vickers (1996a, 1996b) postulates that fixations of rela tively longer duration are necessary during the preparatory stages of the task and are directly associated with increased accuracy. Simply, prolonged fixations are not associated with vigilance, but rather a resu lt of the opportunity to process cues that have already been extracted from the environment mo re acutely. Once movement has been initiated new environmental cues are crucial to maintaining focus on the target, however, the execution phase of task completion requires astute focus. According to Vickers (1996a) it is during the execution when suppression occurs; the new visual information presented to the performer is significantly filtered, reducing the probability of distraction from task irrelevant cues. Pre-performance routines in self-paced ta sks are a common thread among applied sport psychologists and athletes looking for a perf ormance advantage. Free throw shooting in basketball is the quintessentia l example of the importance of synchronizing the mind and body prior to motor execution through pre-performan ce routines. However, the link between gaze control and success is limited. Vickers (1996b) studied 16 elite, female basketball players, who competed at the intercollegiate and national levels. However, Vickers dichotomized the group as experts and near-experts based on their free-throw shooti ng percentage during one full season, including playoffs. Those who achieved an accuracy rati ng exceeding 75% were deemed expert, whereas those who shot less than 65% successfully were classified as near-experts. Gaze behavior was recorded during the completion of 10 successful a nd 10 unsuccessful shots. Results indicated that

PAGE 104

104 only the expert group located a targ et early in their search pattern s and fixated on that target until the initiation of the shot. Again, this suggests expert performers exhibit distinct gaze behaviors. Furthermore, the QE period was able to consis tently discriminate be tween skilled and lesser skilled athletes. Complex tracking and aiming tasks such as hitting a baseball, returning a tennis serve, and receiving a volleyball serve are ch aracterized by three distinct segments (Vickers & Adolphe, 1997), beginning with the detection phase (i.e., determining the flight path of the object), followed by the pursuit tracking phase (i.e., followi ng the flight path), an d concluding with the aiming phase (i.e., orienting the body to make contact with the projectile). To address the role of gaze behavior across the three distinct segments associated with the reception of an object, Vickers and Adolphe ( 1997) tracked the gaze behaviors of 12 elite volleyball players as they receiv ed a serve and executed a passi ng shot to a designated target. Experts were classified as those who achieved a reception and pass rating of 64% or better. The near-expert comparison group were those who achie ved a reception and pass rating of less than 50%. Results indicated that the expert group demonstrated unique gaze behaviors as compared to the near-experts. That is, the expert group demo nstrated a pronounced QE period, meanwhile the near-expert group failed to establish a QE period at all, such that the purist tracking of the ball during flight was replaced with a higher inci dence of corrective footwork. Unique to this investigation, the QE period was also ev ident in the reception of an object. To this point QE researchers had neglected to account for the influence that emotions may have on gaze behavior. Traditiona l expertise research has suggested that elite performers have superior mental skills that provide more eff ective emotional regulation and coping strategies (Gould, Weil, & Weinberg, 1981). In an attempt to fill this gap Vick ers, Williams, Hillis,

PAGE 105

105 Rodrigues, & Coyne (1999) examined the effect s of cognitive anxiety and physiological arousal on gaze behavior and shooting accuracy. Eleven el ite biathletes fired 10 shots at a target 5m away while under high anxiety a nd low anxiety conditions. Quie t-eye was influenced by the modulations in cognitive anxiety and physiological arousal, as wa s performance. It is worth noting however, that regardless of the experiment al condition, successful shots (i.e., hits) were associated with prolonged QE periods, while unsuccessful shots were characterized by reduced or non-existent QE periods. It can be argued that as task complexity increases the fundamental rules necessary for successful completion will also change (Wulf & Sh ea, 2002). In the case of a billiards task of varying complexity Williams, Singer, and Frehlich (2002) assessed the gaze behaviors of skilled and less skilled participants. It was proposed th at skilled players woul d generate longer QE periods during the preparation phase. Furtherm ore, as task complexity increased it was hypothesized that respective QE period would also increase. The increased complexity necessitates increased resources and preparation, therefore, if the QE is related to cognitive processing a direct relationship between the two should be evident (Williams et al., 2002). Results confirmed the hypothesis and provided further support for the location-suppression hypothesis and QE as a function of expertise. Moreover, it was demonstr ated that as task complexity increased, so too did the corresponding QE peri od. Expert-novice differences maintained their relationship as experts continue d to elicit longer QE periods, while successful shots and unsuccessful shots across skill levels could be differentiated by the QE gaze behavior. Finally, Janelle et al. (2002) further extended the QE research paradigm to small-bore rifle shooters. Marksmen are renowned for their unm atched ability to re gulate their physiology (Konttinen & Lyytinen, 1992) and focus their attent ion prior to each shot (Konttinen & Lyytinen,

PAGE 106

106 1993; Lawton, Hung, Saarela, & Hatfield, 1998). Consis tent with the notion th at the QE period is associated with the coordinati on and production of the requisite resources to execute a task successfully while filtering out irrelevant environmental cues, it was postulated that expert shooters would engage in prolonge d QE periods when compared to novice shooters. Confirming this hypothesis, Janelle and colle agues (2002) reported performan ce and QE differences between groups, with the expert group outperforming the novices and exhibiting re latively longer QE periods. Quiet Eye Summary and Review Advances in research technology, specifically ad vances in visual search equipment, have provided a unique opportunity for expertise re searchers to probe previously unattainable questions from a mechanistic perspective. The resulting gaze behavior research in sport has provided valuable insight into the mechanisms that may differentiate expert from novice performers. In this case, the QE period, has inde xed a mechanism that appears to be temporally linked with the organization of visual informa tion and necessary for the coordination of the motor pathways required for successful execution of a desired task. To this point, research has reliably demonstrated relatively prolonged QE pe riods as an effective marker for differentiating skilled and lesser skilled athletes. Moreover, these findings have shown consistency across domains as diverse as a rifle shoo ting and billiards and across tasks that require aiming at a target (e.g., billiards and shooting) to those that require the individu al to receive a projectile momentarily while aiming and releasing it to a designated target (e.g., volleyball). Expertise, Visual Search and Emotion Regulation The interaction between the visual and emotional systems has been robustly demonstrated across a variety of tasks (e.g., martial arts, drivin g, billiards, biathlon shooting), suggesting that as anxiety increases, a correspond ing reduction in visual attenti on efficiency and psychomotor

PAGE 107

107 performance is likely (Murray & Janelle, 2003; Williams & Elliott, 1999; Williams, Vickers, & Rodrigues, 2002; Williams, Singer, & Frehlich, 2002) It is a commonly held belief that anxiety leads to attentional narrowing, such that the breadth of attention is reduced and distractibility is increased (Easterbrook, 1959), often resulting in an increase in se arch rate (Janelle, Singer, & Williams, 1999; Murray & Janelle, 2003). Moreover, under periods of heightened anxiety, a mobilization of additional attentional resources is likely to occur in an effort to offset the deleterious effects of anxiety, ultimately rendering a less efficient performance (Eysenck & Calvo, 1992; Janelle, 2002). When the outcome of an ensuing event is unc ertain, participants typically experience elevated levels of stress, ar ousal, and anxiety (Jones & Swai n, 1992). Consequently, the highly complex, competitive, and unpredictable nature of sport often prompts heightened arousal and anxiety. According to the Pro cessing Efficiency Theory (PET; Eysenck & Calvo, 1992), the interaction of individual state a nd trait levels of anxiety couple d with environmental constraints (e.g., performance pressure and/or uncertainty) directly impact the functional capacity of attention. Specificall y, the processing efficiency theory im plicates the capacity and resilience of the short-term working memory system as an inherent limitation in human information processing. Short-term working memory is thus su sceptible to the adverse effects of increased state anxiety, thereby comprising performance. That is, cognitive anxiety, which is characterized by worry and an inability to concentrate, may redirect attention towa rd a preoccupation with thoughts of evaluation and outcome expectations, and away from task relevant cues (Liebert & Morris, 1967). However, a major tenet of the proce ssing efficiency theory states that as anxiety increases, performance decrements may be a voided via the recruitment and allocation of

PAGE 108

108 additional cognitive resources, thereby preventing the decline in available working memory resources while sustaining performance. Recent gaze behavior research in sport (e .g., Janelle, Singer, & Williams, 1999; Murray & Janelle, 2003; Williams & Elliot, 1999; Williams, Vickers, & Rodrigues, 2002) offers empirical support for the theoretical tenets put forth by Eysenck and Calvo (1992) and Easterbrook (1959), such that visual search efficiency declines with increases in stress and anxiety, while preserving performance. Furthermore, skill-based differences in sport have been reliably reported, with expert performers engaging in fewer fixations of longer duration as compared to less skilled participants (Mann, Williams, Ward, & Jane lle, 2006; Williams, Davids, & Williams, 1999). Accordingly, such skill-based visual search differences may be the result of greater selfmonitoring of performance pr ocesses among novice performers leaving them with fewer cognitive resources available for task execution as compared to experts. The seminal work of Janelle et al. (1999) in sport, examined changes in gaze behavior under varying levels of anxiety during a simula ted auto-racing task. During the course of the driving task, participants were required to detect and discriminate between relevant and irrelevant cues randomly presented about the periphery of the visual display while being subjected to random manipulations in anxiety. Results indicated that as anxiety increased, processing efficiency and task performance decr eased. Additionally, such changes were coupled with a marked increase in gaze be havior variability (i.e., more fixations of shorter duration). As such, Janelle et al. (1999) conclude d that heightened anxiety can l ead to increases in attentional narrowing, resulting in the ineffective s earch and use of perceptual cues. In a related study, Murray and Janelle (2003) employed a dual-task auto-racing simulation to assess the relationship between visual sear ch behaviors and modula tions in anxiety. Murray

PAGE 109

109 and Janelle reported that the additional cognitive and attentional demands associated with the relative increases in anxiety re sulted in a significant decrease in processing efficiency. Specifically, as anxiety increased so did search ra te, rendering the performer less efficient. The notable reduction in processing effi ciency may be in part due to a decreased ability to utilize or discriminate between relevant a nd irrelevant cues. According to Easterbrook (1959) narrowing of attention is a natural by-product of the interaction between arousa l and task demands, such that the breadth of attention becomes constrai ned. For example, under normal conditions and moderate arousal, the discrimination of task relevant cues from task irrelevant cues is a relatively effortless process. However, as arousal increases the scope of attention is reduced, restricting the number of cues, both task relevant and irrele vant, that can be simultaneously processed and distinguished, ultimately compromising the use of task relevant cues. Furthermore, Williams et al. (2002) assessed table-tennis performance under combinations of high and low working memory with correspon ding changes in anxiety (high and low). As expected, results indicated incr eased effort, delayed reaction times, and increased search rates while performing under the high working memory a nd high anxiety condition as compared to the low working memory and low anxiety condition, a pattern indicative of less efficiency. In the first expert-novice comparison, Williams and Elliot (1999) examined the effects of cognitive anxiety and level of experience on anticipation and visual search behavior during simulated karate kumite combat situations. Expert and novice martial artists viewed a series of film sequences depicting both fist and foot stri kes. Participants were required to physically counter each simulated strike as quickly and as accurately as possible under low anxiety and high anxiety conditions. Consistent with the Proces sing Efficiency Theory and Cue Utilization Hypothesis, Williams and Elliot (1999) predicted and found evidence to support the idea that as

PAGE 110

110 anxiety increased, so too would th e number of fixations to the pe riphery of the visual display. Specifically, they found that as cognitive anxiet y increased, the viewing patterns of the expert and novice groups changed variably, with a marked increase in attention to peripheral cues. Worth noting however, was a reported decrease in visual search efficiency in the novice group. That is, while the experts in the high anxiety condition demonstrated an increase in their respective mean fixation duration, the mean fi xation duration of the novice group decreased. Consistent with previous findings, the correspo nding increase in search rate among the novice group can be explained by a comparative decrease in processing efficiency and ineffective cue utilization (Williams & Elliot, 1999). Collectively, the aforementione d investigations lend support to the negative effects of anxiety on performance, processing efficiency an d cue utilization. Moreover, these findings can be viewed as illustrating the increase in the cognitive/attentional demands which accompany increases in anxiety and arousal. However, the increased fixation du ration of the expert group in the Williams and Elliott study, may suggest that a decrease in visual search rate permits the allocation of additional cognitive resources to offset the attentional demands imposed by heightened anxiety. As such, the QE period may serve an anxiety regulat ion function during selfpaced tasks. The QE research has demonstrated a unique and robust relationship between performance and gaze stability during the temporal period imme diately preceding task performance. Given the magnitude of this relationship, it is not out of the question to infer that the QE period serves a motor planning function. However, in light of the well-established rela tionship between anxiety, visual search, and performance across tasks of varying complexity, rese archers have suggested that the QE period may also reflect a temporal window for the regulation of emotion (Janelle et

PAGE 111

111 al., 2000; Vickers et al., 1999). That is, the prolonged QE duration that is characteristic of expert and superior performance may fac ilitate the prevention of the pro cessing of irrelevant stimuli, a characteristic commonly associated with increases in anxiety and/or arousa l. Central to Eysenck and Calvos (1992) Processing Effici ency Theory is a self-regulato ry system, a system in place to mediate the effects of anxiety on processi ng and performance (p. 415). The QE period may provide an indirect measure of this regulatory system. In the first exploration of th e relationship between QE durat ion, anxiety, and performance, Vickers et al. (1999) examined the effects of cognitive stress and physiological arousal on the gaze behavior and shooting accuracy of elite biathletes. Although the QE period was influenced by modulations in cognitive stress and physiolo gical arousal, QE durations during best performances were similar across levels of cognitive stress and physiological arousal. Specifically, during the high anxiet y conditions, QE duration for h its continued to exceed the quiet duration for misses. As such, it is plausi ble to infer that a prol onged QE period may serve to alleviate the attention demanding e ffects of anxiety on performance. Furthermore, Williams, Singer, and Frehlich (2002) assessed the gaze behaviors of skilled and less skilled billiard players. Although not a direct assessment of anxiety, Williams et al. (2002) imposed temporal cons traints intended to increase ta sk complexity. They suggested that increased task complexity necessitates incr eased resources and preparation, and postulated that if the QE is related to cognitive processing a di rect relationship between the two should be evident (Williams et al., 2002). Results demonstrated that as task complexity increased, so too did the corresponding QE period. Ex pert-novice differences were also evident. Specifically, experts continued to elic it longer QE periods as compared to their novice counterparts, and QE duration was proportionally longer on successful shots than on unsuccessful shots across skill

PAGE 112

112 levels, suggesting that the QE period may in fact aid in the circumvention of cognitive constraints (Janelle, Hillman, Apparies, Murray, Meili, Fallon, & Hatfie ld, 2000; Vickers et al., 1999). Expertise, Visual Search and Emot ion Regulation: Summary and Review In conclusion, the effects of anxiety on info rmation processing efficiency have been reported across a variety of tasks in which visual search effici ency declines with corresponding increases in anxiety and task difficulty. Most intriguing how ever, is the fact that the QE/performance relationship is not mediated by changes in task complexity (Williams et al., 2002), modulations in cognitive anxiety or phys iological arousal (Vickers et al., 1999). Although, the duration of the QE pe riod has been demonstrated to change as a function of task complexity and other constraints, the performanc e relationship remains robust, with increases in QE consistently being associated with increase d performance and expertise. As such, as task complexity and or arousal and anxiety increase, th e associated increase in QE duration may serve a regulatory function permitting the formulation of a precise motor program necessary for performance. The following section will demonstr ate the cortical adap tations separating the expert from the novice performer. Of pr imary importance across the psychophysiological literature is the reported psychomotor efficiency of the expert, coupled with an increased readiness to perform. As such, the cortical si gnatures associated with such efficiency and readiness to perform may be the result of prolonged QE periods as reported here. Cortical Activity and the Preparatory Period The systematic observation of electro-cortical activity pr ovides a noninvasive, objective index for deducing the concomitant psychologica l processes responsible for the informationprocessing and sensorimotor distinctions of expe rt and near-expert perf ormers. Specifically, the use of continuous electroencephalographic (EEG ) techniques (e.g., spectral analysis and event

PAGE 113

113 related potentials) permits a conc urrent, real-time look into the underlying co rtical structures contributing to the psychological processes acco unting for skill differences. Self-paced tasks demanding visuomotor coordination such as golf putting, archery, and marksmanship have been extensively used in the sport sciences. Each of the aforem entioned tasks inherently requires the participant to remain relatively motionless durin g the preparatory period, reducing the potential confound of movement artif act in the EEG waveform. Moreover, each task places great demand on the psychomotor systems (i.e., attention, emo tion, motor control and preparation) conducive to psychophysiological record ing (Hatfield & Hillman, 2001). To date, the majority of research in spor t has relied on spectral an alysis techniques to address issues of hemispheric asymmetry duri ng the preparatory period to infer the covert psychological processes associated with superior performance. As such, research stemming from the spectral domain has consistent ly reported cerebral efficiency in favor of the expert. However, the use of event-related potentials (ERP) is fa r less prevalent, despite the advantage of the functional significance of the information processi ng transactions that are manifested in the components of the ERP as a function of context (Fabiani, Gratton, & Coles, 2000). Of interest here is the Bereitschaftspotential. Therefore, the purpose of this s ection is to provide a review of the electro-cortical literature in sport as it relates to both the spectral and ERP components during the preparatory period of a self-paced task. Spectral Activity Sport psychophysiologists (e.g., Crews & La nders, 1993; Hatfield, Landers, & Ray, 1984, 1987; Salazar, Landers, Petruzzell o, & Han, 1990) investigating the co vert cortical processes of skilled performers have made extensive use of the electroencephalographic (EEG) spectral analytic techniques, comparing a nd contrasting levels of hemispheric activa tion. Specifically, the decomposition of the EEG waveform, permits the analysis of a specific frequency band in the

PAGE 114

114 spectrum ranging from DC to 44Hz. Sport research however, has pl aced a great de al of emphasis on the alpha (8-12HZ) band and to a much lesser extent, the beta band (13-36Hz) of the power spectrum. Alpha power, which is believed to result from thalamic input to the cortex (Lopes da Silva, 1991), has been used to infer the func tional and mechanistic si gnificance of various cortical structures across hemispheres. Because alpha activity is inversel y related to cortical activation, the comparison of left and right hemi spheres permits the inference of the specific cortical regions necessary for the exec ution of a given visuo-motor task. The early work of Hatfield and collea gues (1982) was the first in a series of investigations (Hatfield, Landers, & Ray, 1984, 1987) examining the covert cognitive processes associated with skilled psychomotor performance of elite marksmen. In this seminal study, cortical activity was recorded during the prepar atory period leading up to the point of trigger pull. As hypothesized, Hatfield and colleague s confirmed alpha power differences across hemispheres, with greater cortic al quieting in the left as comp ared to the right hemisphere. In a subsequent attempt to further isolate th e cortical patterns of expert marksmen during the preparatory period, Hatfie ld, Landers, and Ray (1984) conducted a study in which the 7.5 second pre-shot period was subdivided into three 2.5 second epochs facilitating the analysis of cortical activity as the time to trigger pull near ed. Furthermore, a more elaborate montage was employed including T3, T4, O1, and O2 reference sites. Cortical activa tion across the three 2.5 second epochs revealed a steady global decrease in cortical ex citation. More telling however were the asymmetrical differences noted between the left and right te mporal regions, while relative symmetry remained acro ss the occipital sites. Together, these findings suggest that cortical specificity plays a signifi cant role in the performance of skilled athletes. To elaborate, the decrease in left hemisphere activation may re present a reduction in verbal-analytic processing

PAGE 115

115 as the time to trigger pull near s coupled with a relative increas e in visual-spa tial processing, permitting an ideal preparatory set for ta rget shooting (Hatfield et al.,1984). Given that athletes, coaches, and sports scientists place great importance on preperformance routines for facili tating attentional focus the ea rly psychophysiological work of Hatfield and colleagues sparked a surge of research into the covert attentional focus patterns associated with self-paced closed skill sports. For example, Salazar et al. (1990) examined hemisphere-related temporal changes in EEG activity of 28 elite archers during the 3-second pe riod preceding arrow rel ease. Cortical activity was recorded from left (T3) and right (T4) te mporal areas while performance was measured as the distance from the inside edge of the arrow to the center of the target. Their results revealed a steady cortical state in the right hemisphere coupled with a signif icant decrease in left temporal cortical activity (i.e., increase alpha power). Furthermore, the 1-second period immediately prior to arrow release for the four best and four worst shots for each part icipant was analyzed, indicating that pre-performance cognitive states are reflected in subsequent performance. More specifically, although right hemisphere activity was consistent across both best and worst shots, a significant increase in left hemisphere alpha power was evident for best shots, suggesting that optimal performance may be related to a reduc tion in verbal analytic processing, a finding consistent with the work of Hatf ield and his colleagues (1984, 1987). In another study (Crews & Landers, 1993), the electro-cortical activity and golf putting performance of highly skilled golfers (N = 34) was analyzed with emphasis placed on the primary motor (C3, C4) and temporal (T3, T4) co rtices. Similar to previous research (e.g., Hatfield et al. 1984; Salazar et al. 1990), the pre-performance pe riod was subdivided into three 1second epochs prior to the initia tion of the putting stroke. However, unique to this investigation

PAGE 116

116 was the inclusion of a variety of EEG analysis techniques, including spectr al power and the slow wave analysis of the Bereitschaftspotential. Corroborating previous findings, the spectral analysis revealed a significant increase in th e left hemisphere alpha power while reporting relative stability in th e right hemisphere. Again these resu lts suggest a relationship between performance and a reduction in verbal analytic processing with active vi sual-spatial processing during the preparatory period of skilled performers. To further explore the covert attentional strategies and pr eparation to respond, Crews and Landers (1993) extended their analysis to incl ude the Bereitschaftspotential, an index of the psychomotor readiness to perform. Their results re vealed a negative shift in the cortical potential over the left hemisphere from epoch 2 to epoc h 1, indicating an enha nced preparatory set, coupled with relative stability over the right hemisphere. Consis tent with BP research, the asymmetrical changes are consistent with limb dominance and performance execution (see section Bereitschaftspotential). Lastly, Crews and Landers (1993) hypothesized that each of the EEG measures recorded would correlate with performance, with an increas ed probability for such a relationship occurring at epoch 1 (i.e., the final second before arrow rele ase). Contrary to previo us research, the results of the spectral analysis revealed a significant relationship between the alpha activity of the motor region of the right hemisphere and putting perfor mance error, suggesting that tasks requiring precise movement contributions from both limbs may result in symmetrical cortical contributions. The Bereitschaftspotentia l was not related to performance. In an effort to advance the conceptual understanding of the aforementioned cortical patterns, Hillman, Apparies, Jane lle, and Hatfield (2000) compar ed the EEG spect ral activity of skilled marksmen prior to the execution and withdrawal of shots. To clarify, during competition

PAGE 117

117 marksmen are permitted to reject or withdraw thei r rifle from the target prior to trigger pull, suggesting a state of mental or physical unrest. Accordingly, Hillman and colleagues hypothesized that the cortical signatures of the pre-performance period preceding rejected shots would reflect an inability to adaptively allocate visual-spatial processing, resulting in an increase in right hemispheric EEG alpha power relative to an executed shot. The results of the comparison between executed and rejected shots revealed a relative and progressive increase in alpha power for rejected shots as compared to executed shots, supporting the noti on that the decision to withdraw from a shot is represen tative of incongruent cortical activation for the requisite task. Collectively, the results of the early psychophysiological work provide empirical support for the relationship between pre-performance cor tical activation and the quality of performance of highly skilled participants. The analysis of EEG spectral ac tivity across hemispheres has revealed that the analytic left hemisphere (T3) shows marked increases in alpha power during the time period immediately prior to ta sk execution, while the visual-s patial processing associated with right hemisphere (T4) activ ation remains relatively vigilant during the same period. Overall, as the time to task execution nears, cortical activi ty declines, and the relative shift in hemispheric dominance suggests cortical specific ity among highly skilled performers. Discrepant findings have been reported however with respect to role of the occipital EEG alpha power. For example, the seminal work of Ha tfield et al. (1984) repo rted an increase in alpha power in the left occipita l region (O1), but faile d to replicate that finding in a later study (Hatfield, Landers, & Ray, 1987). Given the apparent discrepancies of occipita l EEG alpha power noted in the previous work of Hatfield et al. (1984), Ha tfield, Landers, and Ray (1987), and Loze, Collins, and Holmes (2001) inferred that the proces sing of visuo-motor stimuli should either remain constant or

PAGE 118

118 decrease in the preparatory period immediately preceding trigger pull. That is, when performing a well-learned skill, the continued acquisition an d analysis of visual information immediately prior to performance may interfere with the fo rmulation and execution of the requisite motor program. As such, cortical idling or quieting in the occipital lobe would suggest a reduction in visual attention. Therefore, Loze et al. (2001) examined the corti cal activity of six expert airpistol shooters across best and worst shots to determine the me diating role of occipital EEG alpha power on shooting performance. Continuous electroencephalograp hic activity was collected fr om Oz, T3, and T4, and was referenced to linked mastoids. Po st acquisition analysis consiste d of the reduction of pre-shot data into three 2 sec EEG epochs for which sp ectral analysis was c onducted, yielding absolute alpha (8-13Hz) power values of best and worst shots. Results indicate d that occipital alpha power was greater in best shots, with the greate st difference in power noted in the third epoch (i.e., the final 2-second period immediately prior to trigger pull). Moreover, not only did alpha power increase during best shots, alpha power wa s found to decrease in the final epoch prior to worst shots. Corroborating previous EEG work (e .g., Hatfield et al., 1984; Hatfield and Landers, 1987), hemispheric asymmetry was al so reported. That is, significa ntly greater alpha power was evident in the left temporal region (T3) as comp ared to the right tempor al region (T4), a finding consistent with the notion of cerebral efficiency. In-line with the dominant theory of motor e xpertise (Fits & Posner 1967), the work of Loze et al. (2001) lends support to the notion that skilled pe rformance is optimized when conscious processing during skill execution is minimized. In this case, visual information processing gives way to the formulation and execution of the nece ssary motor program, increasing the probability of optimal performa nce. Although, compelling, such conclusions are

PAGE 119

119 tenuous, given that the cortical re gions associated with the gene ration and execution of the motor pathways were not assessed. For example, the slow potentials known to have motor preparation and execution implications should be concurrently assessed to determine the actual effect of verbal-analytic and visual-spatia l processing on the motor pathways as opposed to relying on the inferential connections based solely on performance outcome. The early investigations of cortical activity during rifle shooting, archery, and golf putting provided direct accounts of the relationshi p between covert psychological process and performance in highly skilled pe rformers. Unfortunately, the lack of a control or comparison group (i.e., less skilled participants ) in the aforementioned investiga tions renders it difficult to infer the extent to which these co rtical profiles can account for skil l development. That is, are the cortical patterns denoted above requisite for skille d performance? Or are su ch patterns consistent across all skill levels? The direct comparisons of expert and less skilled performers should provide insight into the cogniti ve processing differences as skill becomes more efficient, well learned, or automatic. Expertise Differences in Cortical Activity The development of sport-specific psychomotor abilities and cognitions has been argued to result in less effortful and more automatic pe rformance, which in turn permits highly skilled performers to act with less cogniti ve stress compared to their le ss skilled counterparts. As such, expertise researchers have provided extensive empirical accounts that document the cortical differences of skilled and unskilled performe rs. Extending the aforementioned intra-subject designs, expertise researchers not only hypot hesized a significant relationship between hemispheric activation and perf ormance within groups, but fu rther anticipated cortical differences across skill levels. That is, EEG spectral power was expected to reflect the heightened mental effort associat ed with unskilled performance.

PAGE 120

120 Haufler, Spalding, Santa-Maria, and Hatfield (2000) conducted the first study investigating the covert psychomotor differences of skilled and unskilled performers. In their study EEG spectral activity of the left and right frontal temporal, parietal, and occipital regions of competitive marksmen and novice shooters was reco rded during the 6 second preparatory period of 40 self-paced trials, with the resulting data being partitioned into six 1 second epochs. When compared with the marksmen group, the novice shooters were predicted to exert more cognitive effort as a result of their ac tive engagement in verbal-analytic processing. Confirming their hypothesis, Haufler et al. ( 2000) reported that during the 6 second aiming period, the marksmen exhibited less cortical activity in the left hemisphere as compared to the novice shooters across reference sites with the grea test alpha power evident in the left temporal (T3) region. The notable corti cal differences between the marksmen and novice shooters suggests that during the process of skill acquisiti on, cortical adaptation o ccurs in the order of cortical specialization. As such, it can be inferred that the corti cal areas associated with the greatest expert-novice difference are those most relevant to task performance. In a similar study, Janelle et al. (2000) examined expertise differences in cortical activation during rifle shooting. Unlike other expertise studies, the participants included were similar in years of shooting, differing only in their compe titive experience, suggesting that any notable difference can be inferred to be the direct re sult of skill and not the result of exposure or familiarity with the task. Corr oborating previous research (Crews & Landers, 1993; Hatfield et al., 1984, 1987), Janelle and colleagues reported a di rect intra-subject re lationship across skill levels between performance and hemispheric acti vation. To elaborate, successful performance was characterized by an increase in left hemisphe re alpha power compared to that of the right hemisphere. Moreover, expected expertise differe nces were also reported. Specifically, although

PAGE 121

121 both groups appeared to display si milar cortical pattern s, the experts consistently demonstrated greater hemispheric asymmetry as compared to the novice group, suggesting a pronounced level of cortical specificity. Lending credence to the conclusions of cortical efficiency and specificity of experts is the developmental approach to the study of cerebral cortical activity in novice performers. The early stages of skill acquisition is often uncoor dinated and accompanied by effortful cognitive analysis. However as skill progresses, coordina ted movement requires le ss conscious regulation, and performance becomes autonomous, free from cognitive constraints (Anderson, 1982; Fitts & Posner, 1967). Therefore, in an effort to empirically evaluate the neurocognitive adaptations hypothesized to occur during ski ll acquisition and the developmen t of expertise, Landers, Han, Salazar, Petruzzello, Kubitz, and Gannon (1994) conducted a longitudinal investigation of novice archers. As anticipated, as th e skill of the novice archers deve loped over the course of the 14week training program, so to did their corr esponding cortical patte rns. Specifically, EEG asymmetry was characterized by pronounced alpha power in the left hemisphere as compared to the right, with the most notable differences occurring 0.5 seconds prio r to arrow release. Although lacking an expert comparison group, the co rtical adaptations repor ted by Landers et al. (1994) are consistent with the notion of psycho motor efficiency put forth by Fitts and Posner (1967) and further suggest that the degree of psyc homotor skill acquisition is reflected in EEG cortical asymmetry (H atfield & Hillman, 2001). In a more recent investigation, Kerick, Dougl as, and Hatfield (2004) recorded EEG eventrelated alpha power (ERAP) in an effort to in crease the temporal resolution of alpha power estimates over a 14-week training period in novice pistol shooters. Confirming previous research (i.e., Landers et al., 1994), as performance im proved over the course of the 14-week training

PAGE 122

122 period, ERAP asymmetry evolved with pronounced in creases evident in the left temporal (T3) region as compared to the right. Collectively, the expertise and developmental research has consistently revealed intrasubject and inter-subject hemispheric asymmetry ch aracterized by relative cort ical stability in the right hemisphere and increased al pha power in the left. These results are consistent with the notion of psychomotor efficiency characterized by a decrease in verb al-analytic processing associated with the left hemisphere and a lack of cortical quieting in th e right due to the visuospatial processing associated with precision sports. Coherence The conscious processing of task relevant cues is often considered characteristic of the novice performer as indexed by decreased levels of hemispheric asymmetry and in general elevated levels of cortical activity in the anteri or-temporal regions (i.e., T3 and T4; Hatfield et al., 1984; Hatfield, Landers, & Ray, 1987). Altho ugh the use of asymmetry metrics provide an index of the magnitude and directi on of the differential levels of activation in specific cortical regions across hemispheres, the electroencepha lographic technique of coherence analysis, permits the functional assessment between variou s regions of the cerebr al cortex (Davidson, Jackson, & Larson, 2000). That is, coherence is a frequency band specific analysis that reflects the degree of linear relatedness of two cortical regions during the ti me course of a specific task. The greater the coherence, the greater is the co rrelation between the two points of reference, suggesting active communication between the sites. Conversely, low cohere nce is indicative of relative cortical autonomy. For example, elite level performers are believed to operate free from cognitive constraints, relyi ng less on verbal cues (Anderson, 1982). Accordingly, low coherence is expected between the visuo-spatial, verbal -analytic, and motor programming regions of the cortex as skill proficiency increases (Deeny, Hillman, Janelle, & Hatfield, 2003).

PAGE 123

123 Consequently, Deeny et al. ( 2003) employed coherence procedur es to assess whether alpha (low 8-10Hz and High 10-13hz) and beta (13-22Hz) band coherence of left anterior-temporal region (T3) and the motor planning regions (pre-motor cortex; Fz ) of the cortex are inversely related to skill level. Ten expert and 9 skilled marksmen performed 40 shots each using the Noptel Shooter Training System (ST-2000) during an 80 minute te sting session. Cortico-cortical communication was assessed during four one-sec ond epochs preceding trigger pull. Results indicated that compared to the skilled group, the expert marksmen demonstrated lower coherence between the anterior-temporal and motor regi ons of the cortex for low alpha and beta frequencies. Subsequent analys es further indicated lower coherence between motor regions and all left hemisphere reference sites for high-alpha activity and lower coherence for beta activity between the anterior-temporal s ite and midline locations (i.e., Cz, Pz) in the expert group. The findings of Deeny et al. (2003) s upport the notion of de creased cortical communication between functionall y diverse brain regions in th e elite level performer. As previously mentioned, much of the psychophysiologi cal research in sport has ascribed to the basic tenets of Fitts and Posners (1967) notion of automaticity, such that elite level performance is governed by the automatic pr ocessing of the planning and execution of movement. Moreover, as skill develops, the cortical regions necessary for skill execution become more specialized, therefore relying more on the cortical m echanisms responsible motor programming and execution (i.e, supplementary motor cortex, pr imary motor cortex, and the corresponding subcortical generators) as confir med by the psychological differe nces of expert and skilled marksmen. Bereitschaftspotential The Bereitschaftspotential (BP) (or Readiness Potential) fi rst described by Korhuber and Deecke (1964) is a negative moving cortical slow potential that precedes the onset of self-paced

PAGE 124

124 movements by approximately 1 to 1.5 second. The sl ow negative cortical waves, of which the BP is one, are believed to be associated with arous al, the recruitment and augmenting of responses, and the general facilitation of pr ocesses required for task comple tion (Deecke, 1973). That is, an increase in negativity consistent with the BP is representative of cortical activation or excitation. By deductive reasoning, Brunia (199 3) reports that all functions, including percep tion, attention, and preparation are realized by complex patter ns of facilitation and inhibition in specific neuronal circuits which must be depolarized in order to produce an action. Therefore, it becomes possible to study, through electroe ncephalography, the cortical ac tivity of the pre-motor period given that the depolarization process characterize d by an increase in nega tivity at the surface of the cortex must precipitate an action (Brunia, 199 3). Accordingly, readiness to act is depicted by a cortical signature manifested in motor and at tentional processes duri ng the preparatory period (Brunia, 1993; Deecke, 1973). The BP is a visually distinct waveform (see Figure 2-6), denote d by three critical components that are temporally and cortically distinct (Deecke, Schei d, & Kornhuber, 1969): the BPearly, BPlate, and BPpeak. The slow rising negativity of the BPearly starts approximately 1500 ms prior to movement onset and although demonstrated to have a wide spread scalp distribution with maximal potentials recorded at the vertex, th e early onset and pronounced amplitude implicates the supplementary motor area (SMA). The relative increase in cortical ac tivity associated with the mid-frontal (Cz) region is not altogether surprising given its role in working memory, inhibition, planning, and executive functioning, including the integrati on and regulation of emotion and its corollaries. Specifically, the SMA is reportedly central to the planning and initiation of voluntary movement (Parent, 1996). Fo r example, the BP is absent in movements exogenously driven (i.e., reflexiv e or passive movements) but is further exaggerated in those

PAGE 125

125 movements associated with a consequence for the accuracy of motor execution (Becker, Iwase, Jurgens, & Korhuber, 1976; McAdam & Rubi n, 1971; McAdam & Seales, 1969; Reagan, 1989). As such, any modulation in affect may result in unsystematic variability in the motor pathways, resulting in alterations in the quality of performance (Hatfield & Hillman, 2001). The second component, known as the BPlate, is characterized by a change in the steepness of the waveforms slope, which occurs approxim ately 400-500 ms prior to movement onset. The rapid increase in negativity has been linked to the function of the primary motor cortex (MI). That is, the serial activation of the SMA and MI (preceding movement onset) results in an amplified negativity toward the hemisphere contralateral to movement suggesting the formulation of a more elaborate motor program as indicated by in increased cortical activation (Cui & Deecke, 1999; Deecke & Kornhuber, 1978; Shibasaki et al., 1980; Deecke et al., 1985; Boschert & Deecke, 1986). Lastly, the BPpeak is most pronounced over the hemisphere contralateral to the respondi ng hand and occurs approximately 50-60 ms prior to movement onset. As Brunia and van Boxtel (2000) suggest that the compone nts of the readiness potential comprise a process responsible for the ini tiation of voluntary, self-paced, motor acts. The components of the BP, particularly the BPearly, which has been associated with the premovement activation of the supplemen tary motor areas (SMA), and the BPlate, which is associated with the activation of the primary mo tor cortex (MI), implicate the BP as the only psychophysiological correlate of motivation, prepar ation, intention, and initiation of self-paced goal directed behaviors in man (Deecke & Ko rnhuber, 2003). As such, the BP has been considered a means to address the preparator y processes of voluntary, goal-directed actions while observing the concurrent interaction of attention and motivation (Licht & Homberg, 1990).

PAGE 126

126 The classic BP paradigm first describe d by Kornhuber and Deecke (1964) emphasized single finger movements with physi cal and temporal constraints, in which self-paced movements were accompanied by an inter-trial interval of five seconds or more. Although the initial protocol proved innovative, the nature of such a design may have led to the automatic performance of successive trials, negating the effects associated with voluntary, self-paced actions. Fortunately, subsequent research implemented protocols with fewer temporal and movement constraints, more representative of the fast paced complex m ovements of the real world, lending credence to the investigation of object or goal directed movement s (Jahanshahi & Hallet, 2003). Although the sport science literatu re is replete with electrocortical investigations of expertise, few studies have specifically examin ed the event-related cortical slow potentials preceding a movement, while even fewer have ex amined the specific temporal components of the BP in sport. The early work of Konttinen and Lyytinen (1992) determ ined that the preparatory period of marksmen is functionally occupied by either the motor demands (i.e., stabilizing the gun) or the visuo-spatial components (i.e., sighting the targ et) of the task, each of which can be identified by a distinct cortical signature characterized by the direction of the waveforms deflection. During target shooting, as the marksman allocates attenti on to the features (visuo -spatial processing) of the target, slow potential nega tivity becomes pronounced, a patte rn indicative of increased readiness. Conversely, if attention is directed toward the re quisite mechanics (i.e., gun hold) of the task, slow potential positivity becomes pronoun ced, suggesting a decreased readiness to act. Although general pre-shot cortical trends we re evident among skill levels, Konttinen and Lyytinen (1993) put forth the noti on that inter-individual pre-shot cortical patterns should also emerge during the preparatory period dis tinguishing high and low scoring trials.

PAGE 127

127 The slow wave activity of seven male and five female marksmen was obtained from frontal (Fz), central (Cz), and occipita l (Oz) midline and centro-late ral (C3, C4) sites during the 7.5 second period preceding trigger pull and the 1.5 second period following trigger pull. The marksmen completed 300 shots from a standing position to a target pos itioned 18m away. The average amplitude for each of five 1.5-second ep ochs was calculated and used for subsequent analyses. The results of Konttinen and Lyytinen (1993) revealed that each marksman developed a unique cortical signature, s upporting the concept of intra-indi vidual styles of preparation. Furthermore, and perhaps more importantly, dist inct cortical profiles were evident for high scoring and low scoring shots s uggesting that the cort ical activity during th e preparatory period is representative of the psychological readiness to perform which directly impacts performance outcome. Given the potential of the slow cortical wave for implicating an athletes attentional set, arousal level, and overall read iness to perform, Crews and La nders (1993) assessed the motor and temporal cortical activation levels of hi gh skilled golfers during the three-second period preceding the putting stroke. It was hypothesized that a greater cortical change would be evident in the left hemisphere (T3, C3) as compared to the right (T4, C4), suppor ting previous research indicating greater hemispheric changes in the he misphere contralateral to the dominant limb (see Jahanshahi & Hallett, 2003). Althou gh the results confirme d a greater left hemisphere shift from epoch 2 to epoch 1, performance remained c onsistent, failing to support the findings of Konttinen and Lyytinnen (1993) who reported a si gnificant relationship between cortical changes and performance outcome. Despite the lack of empirical support by Crew s and Landers (1993), the work of Konttinen & Lyytinen (1993a, 1993b) suggest that inter and in tra-individual variability can reflect superior

PAGE 128

128 and failed performance (Konttinen, Lyytinen, & Konttinen, 1995). As such, Konttinen et al. (1995) continued to probe the relationship betw een preparatory set, cortical activity, and performance. Cortical slow potentials were record ed from the frontal mid-line (Fz), centro-lateral (C3, C4), and occipital mid-line (O z) locations during the observat ion of 6 elite (internationally ranked) and 6 sub-elite (nati onally ranked) marksmen while performing 180-200 shots from a distance of 18 meters. Unique to this inve stigation, the 7.5 second preparatory window was divided into five 1.5 second epochs for which pos itive, neutral, negativ e, and irregular wave types could be classified. The objective of the K onttinen et al. (1995) study was to establish a typology of slow potentials reflecting the associa tion between cortical ac tivity and performance. Their results revealed an optimal cortical rela tionship, denoted by front al mid-line deactivation coupled with increased asymmetrical activat ion in the centro-lateral hemisphere. More specifically, Konttinen et al. ( 1995) concluded that the magnitude and topography of optimal performance may be described by an efficient mo tor program, as indicated by decreased frontal activation, paired with intense visuo-spatial pr ocessing as indicated by an increase in right hemisphere negativity. In essence, if a motor program is esta blished and accessible, motor regulation will proceed effectively and efficien tly without much cognitive effort, which is discernible by an increase in SP negativity. Conv ersely, however, if the requisite motor program is either primitive or unavailable, additional effort is necessary to perform the task resulting in an increase in slow positivity (Warren & Karrer, 198 4 as cited in Konttinen et al., 1995). Therefore, the cortical profile of superior performance put forth by Konttinen et al. (1995) lends empirical support to the notion of automatic processing in experts (Fitts & Posner, 1967), while further implicating the need for an external focus of attention for well learned skills (i.e., to the target) in lieu of an internal/cognitive processing appr oach (Anderson, 1982). Overall, the findings of

PAGE 129

129 Konttinen et al. (1995) not only support an optimal cortical profile, but they also lend support to the initial hypotheses put forth by Crews and Landers (1993); that is, slow potential changes can be representative of variability in performance. The initial work of Konttinen and Lyytinen (1992) suggests that increased cortical negativity in the frontal, central and occipital regions reflects a general activation corresponding with an increased preparedness to respond. However, later research has as sociated increases in cortical positivity with unrefined or poorly cons tructed motor programs, suggesting that a slow potential cortical profile may be more representative of the e ffort exuded to maintain motor control (i.e., rifle stability; Konttinen & Lyytinen, 1995), compared with the visuo-motor preparedness associated with increased negativity. To further clarify the prevalence and role of the competing cortical deflections, Konttinen, Lyytinen, and Era (1999) conducted a follow-up study assessing psychomotor effort during shooting performance while concurrently measuring the amplitude, direction, and velocity of postural sway during the pre-shot period (i.e., 7.5 seconds preceding trigger pull). Given that previous research has consiste ntly reported enhanced postural balance among expert marksmen as compared to lesser ski lled performers (Aalto, Pyykko, Il marinen, Kahkonen, & Starck, 1990; Niinimaa & McAvoy, 1983), Konttinen and Lyytinen (1992, 1993), predicted that poor shooting performance would correspond with a heightened allocation of res ources and psychomotor effort and a corresponding decrease in available reso urces and effectiveness for visuo-spatial processing. Accordingly, Konttinen et al. (1999) hypothesized that frontal midline cortical activity reflects the psychomotor effort required prior to trigger pull. Specifically, the additional psychomotor effort expended to stabilize the rifl e during low scoring shots would be reflected in a positive cortical deflection and would override the slow potential negativity associated with

PAGE 130

130 optimal visuo-spatial processing. Conversely, du ring successful or high scoring shots, minimal psychomotor effort was expected, resulting in a pronounced negative deflection indicative of arousal regulation and visuo-spatial processing. In accord with their hypotheses, Konttinen et al. (1999) reported that changes in postural sway were aptly reflected in the frontal regions, with increased postural sway related to increases in frontal positivity, while minimal postural sway was manifested in decreased frontal positivit y. Although, such a relationship between postural sway and cortical activation levels was demons trated, Konttinen et al. (1999) concluded that postural sway alone was not a si gnificant predictor of performance. Rather, the psychomotor demands of the task are reflected in the change s in cortical activati on during the preparatory period, which are better able to discriminate skill levels by observing the cognitive effort expended to regulate performance. In line with Fitts and Posner (1967), Konttinen et al. (1999) suggest that experts spend less physical and psychological energy attending to subs idiary tasks (i.e., re gulate postural sway). That is, the experts had already learned an effective strategy fo r regulating postural oscillations, whereas the less skilled marksmen had to consci ously attend to the regul ation of postural sway, thereby reducing the available cognitive resour ces for performance execution. Modulations in slow potential positivity and negativity have been shown to reflect effort and cognitive processing, both of which have been linked to performance outcome, with greater negativity corresponding with better performance. More recently, Konttinen, Landers & Lyttinen (2000) examined the aiming strategies of competitive marksmen, focusing on the final 1000 ms epoch prior to trigger pull. Specifically, it was hypothesized that the final 1000 ms of the aiming period could discriminate between elite and pre-elite marksmen. That is the elite performers were e xpected to demonstrate a more

PAGE 131

131 optimal aiming period characterized by a prolonged period of rifle stability as compared to the pre-elite performers. Furthermore, the cortical prof ile of the elite performers was expected to be more negative than that of th e pre-elite performers, a profil e indicative of less effortful, automatic processing. Specifically, the BP of six elite and six pre-el ite marksmen was recorded from the frontal (Fz), centro-lateral (C3,C 4), and occipital midline (Oz) cort ical regions. The results indicated that the elite performers exhibited more accurate and less variable performance than the pre-elite performers and were significantly more stable in their respective rifle hold while preparing to pull the trigger. However, contrary to the anticipa ted findings, the BP analyses revealed a greater positive cortical shift in the elite performers as compared to the pre-elite performers. According to Konttinen et al. (2000) the increased positivity is likely due to the sustained gross motor activation, which is not relate d to the central timing mech anism (p.175). As previously mentioned, the work of Karrer et al. (1978) indicate s that a positive shift ma y be the result of an insufficient motor plan necessary to inhibit irreleva nt body movements. As such, the positive deflection reported by Konttinnen and colleagues (2000) may be the result of an increased effort of elite marksmen to prevent extraneous movements and not at all an indication of the preparatory set. Alternatively, th e physical stress imparted by th e rifle hold may have in fact masked the BP negativity among the elit e markmen (Konttinen et al., 2000). Although Konttinen and colleagues (2000) failed to provide empirical evidence to support the BP as a measure for assessing psychomot or differences among expert and near-expert performers, the BP should not deemed inappropr iate for the study of psychomotor performance differences. As previously mentioned the main preparatory function of the marksmen is the inhibition of extraneous motor activity during the stabilization of the rifle hold (Konttinen &

PAGE 132

132 Lyytinen, 1993, 2000). Conversely, the execution of the golf putt requires the facilitation of a pre-planned motor act, a task that is more conducive to the use of the BP as means for investigating the motor program. As such, it is pl ausible that the level of physical involvement (i.e., prolonged and variable muscular activation prior to trigger pull) in the preparatory period leading up to the point of trigger pull in ma rksmen, may overshadow the slow potential nature recordings of the BP. Cortical Activity and the Prepar atory Period Summary and Review The systematic observation of electro-cortical acti vity permits a real-time look into the underlying cortical structures contributing to the psychological processes accounting for psychomotor skill differences. Self-paced tasks demanding visuomotor coordination such as golf putting, archery, and marksmanship place great demand on the psychomotor systems (i.e., attention, emotion, motor contro l and preparation) conducive to psychophysiological recording (Hatfield & Hillman, 2001). The early work of Hatfield and colleagues (1982) sparked a series of investigations (Hatfield, Landers, & Ray, 1984, 1987) which examined the covert cognitive processes associated with skilled psychomotor performance of elite marksmen and other self-paced activities. To date, the majority of research in sport has relied on spectra l analysis techniques to address issues of hemispheric asymmetry consis tently reporting skill automaticity and cerebral efficiency in favor of the expe rt, concluding that the reported decrease in left hemisphere activation with the concomitant increase right hemisphere activat ion may represent a reduction in verbal-analytic processing coupl ed with vigilant visual-spatia l processing as the time to performance execution nears; arguably an optimal preparatory set for self-paced precision sports (Hatfield et al.,1984).

PAGE 133

133 Furthermore, the results of the early psychophysiological work provide empirical support for the relationship between pre-performance cortic al specificity and the quality of performance of highly skilled participan ts, with greater asymmetry corresponding with increased performance. As such, this work lends cred ence to Fitts & Posners 1967 Theory of Skill Automaticity, such that skilled performance is optimized when conscious processing during skill execution is minimized. The systematic observation of the slow wa ve, Bereitschaftspotential (BP) provides a window to assess cortical activity central to th e planning and initiation of voluntary movement. During target shooting, research has revealed that marksmen tend to allocate attention to the features of the target and away from the conscious processing of the motor components of the task. Accordingly, the cortical si gnature of successful performance is depicted by an increase in slow potential negativity. Conversely, sub-optim al and novice performance is characterized by an attentional focus that is directed toward the requisite mechanics (i.e., gun hold) of the task in which the cortical signature dur ing the pre-performance period is denoted by slow potential positivity, or an otherwise decreased readiness to act. The systematic observation of expert and novice performers across domains has proven invaluable. Since the seminal work of Chase and Simon (1973), research has supported the notion that experts posses an exte nsive knowledge base that facili tates both stimulus recognition and procedural execution (Richman, Gobet, Staz ewski, & Simon, 1996). Furthermore, the advent of eye-movement registration techniques coupled with various electro-cortical modalities have advanced the current understand ing of the covert psychologica l behaviors distinguishing the expert from novice performer. Acco rdingly, the continued explorati on into the role of the quiet

PAGE 134

134 eye period may serve to better link the visual, em otional, and motor programming components of expert performance.

PAGE 135

135 Figure 2-1 An information-processing account of th e advantages of advance cue usage. (Adapted from E. Buckolz, H. Prapavesis, and J. Fa irs (1988). Advance cues and their use in predicting tennis passing shots. Canadian Journal of Sport Science, 13(1), 20-30 ). Combats Time Pressure Time Pressure in Effect A Prioi Information Sought Contextual Cues Body Language Cues Opponents/ strengths/ weaknesses/ preferences Climatic Conditions Expectancy Developed Anticipation Selective Preparation IncorrectIncorrect CorrectCorrect Increase Movement Time (MT) Eliminate RT, Reduce MT Increased Reaction Time (RT) Decreased Reaction Time x B C D A E F G H I J Stance Racquet Position Direction of Gaze

PAGE 136

136 Figure 2-2 Radial error for expe rt and novice badminton players as a function of the degree of temporal occlusion. (Adapted from B. Abernethy (1988). The effects of age and expertise upon perceptu al skill development in a racquet sport. Research Quarterly for Exercise and Sport, 59(3), 21-221). Figure 2-3 Lateral and depth error for expert and novice wicketkeepers as a function of the degree of temporal occlusion. (Adapted from D.R. Houlston & R. Lowes (1993). Anticipatory cue-utilization processes amongst expert ). -4 Frames (t1) -2 Frames (t2) Contact (t3) 2 Frames (t4) Full Display (t5) 1.2 1.6 2.0 2.4 2.8 3.2 -4 Frames (t1) -2 Frames (t2) Contact (t3) 2 Frames (t4) Full Display (t5) 1.2 1.6 2.0 2.4 2.8 3.2 Expert PlayersNovice Players 12 Years 15 Years 18 Years Adults 12 Years 15 Years 18 Years Adults Radial Error (meters) Radial Error (meters)Time of OcclusionTime of OcclusionNon-Expert Depth Expert DepthNon-Expert LateralExpert Lateral T1T2 T3 T4 10 40 60 80 100 Mean Error (inches)Stage of Temporal Occlusion 20 30 50 70 90 110

PAGE 137

137 Figure 2-4 Error scores for experts and novices. A depiction of an atypical trend in anticipatory cue use when comparing expert with novice pa rticipants. (Adapted from G. Paul and D. Glencross (1997). Expert percepti on and decision making in baseball. International Journal of Sport Psychology, 28, 35-56). Figure 2-5 Scan-paths of expert, intermediate, and novice boxers. Arrows describe the direction of gaze movements between locations a nd the proportional asso ciations between locations. The size of each circle is proporti onal to the percentage of fixation at each location. (Adapted from H. Ripoll, Y. Ke rlirzin, J.F. Stein, and B. Reine (1995). Analysis of information processing, deci sion-making, and visual strategies in complex problem solving sport situations. Human Movement Science, 14, 325-349). Experts Novices T1 T2 T3 T4 2.0 3.5 4.5 5.5 ErrorStage of Temporal Occlusion 2.5 3.0 4.0 5.0 T5 Experts Novices Intermediates H H H AF AF AF TR TR TR UN PE LE PE UN UN H = Head AF = Arm/Trunk TR = Trunk PE = Pelvis LE = Legs UN = Unidentified

PAGE 138

138 Figure 2-6 Temporal schematic of the Bereitscha ftspotential (BP) prio r to movement onset. Adapted from Jahanshahi, M., & Hallett, M. (2003). The Bereitschaftspotential: What does it measure and where does it come from. In M. Jahanshahi and M. Hallett (Eds.), The Bereitschaftspotential : Movement Related Cortical Potentials (1-17). New York, NY: Kluwer Academic/Plenum Publishers. -1.5 -1.0 0.5 0 0.5s BPearlyBPlateBPpeak EMG onset

PAGE 139

139 CHAPTER 3 METHODS Participants Twenty volunteers were randomly recruited from various golf clubs in the Southeastern United States and ranged in age from 18-35 (experts M = 26.0, SD = 6.85; near-experts M = 26.20, SD = 5.83). Participants were objectively classi fied according to the United States Golf Association (USGA) handicap system, with th e experts (i.e., LH) ranging from a 0-2 ( M = 1.20, SD = 1.23) handicap and the near-expert s (i.e., HH) ranging from a 10-12 ( M = 11.30, SD = 0.82) handicap. The LH group ( n = 10) averaged 14.7 ( SD = 5.95) years of playing experience, and completed an average of 56.50 ( SD = 22.12) rounds of golf over the previous 12 months. In comparison, the HH group ( n = 10) averaged 12.4 ( SD = 4.94) years of playing experience, and completed an average of 24.30 ( SD = 9.69) rounds of golf over the previous 12 months. All participants were right-ha nded and right-eye dominant. Instrumentation The following instruments were used to record the measurement of golf putting performance: QE duration, co rtical activity, heart rate, and anxiety across conditions. Putting Surface Golf putting performance was assessed using a nylon NP 50 artificial putting surface (Synthetic Turf International, STI, Jupiter, FL ) outfitted with a 4.25in. regulation size golf hole. A synthetic putting surface was us ed in lieu of an actual putting green because it permits laboratory control (e.g., temperat ure, lighting, speed of green, slope, etc. ) with analogous ecological validity. The speed of the green was determined using a stimpmeter and in accord with STI rating, the putting su rface measured 10.5 feet, an indication of a fast green. Furthermore, the turf was placed on a platform th at was constructed to ensure a level and flat

PAGE 140

140 putting surface. The testing area was designe d to accommodate a 12ft. putt while leaving 25.75in. of space behind the hole and approximately 22i n. on either side of the hole to allow for the measurement and analysis of accuracy, bias and consistency. Figure 3-1 provides a graphical depiction, detailing the dimensi ons of the putting platform. Figure 3-1 Putting green dimensions. Putting Performance The target (i.e., golf hole) was supplemented with an imposed grid used for assessing accuracy, bias, and consistency (Reeve, Fischman, Christina, & Cauraugh, 1994; Hancock, Butler, & Fischman, 1995). Specifically, a 30in x 40in matrix progressing in 1in. increments on both the vertical and horizontal axes was projected onto the putting surface with a Sharp Notevision LCD Projector (Model XG-NV2U, Tokyo, Japan). The coordinate (0,0) indicates the center of the golf hole. The image was projecte d after each stroke and removed prior to each subsequent stroke to avoid la tent visual assessment of pe rformance or impairment of performance potentially induced by the display of the grid. Gaze Behavior A BIOPAC electro-oculogram amplifier (EOG 100B; BIOPAC Systems, Inc., Santa Barbara, CA), with a bandpass range from DC to 100Hz was used to record eye-movements; specifically, QE duration. Analog data were sa mpled at 1000Hz using an MP 150 analog/digital converter and recorded on-line with AcqKnowledge 7.0 (BIOPAC Sy stems, Inc., Santa Barbara, 192" 48" 30" 144" 18" 4"Start Line

PAGE 141

141 CA) software installed on a Dell XPS comput er (Dell Inc., Austin, TX). The EOG 100B amplifier is a biopotential amplif ier designed to record changes in the corneal-retinal potential as the eye navigates the visual environment relativ e to head position (BIOPAC, 2006; Duchowski, 2002). Simply stated, as the eye moves in the horiz ontal and vertical plan es the corneal-retinal potential adjusts accordingly and is reflected in voltage changes in the range of 15-200V with corresponding eye-movements measuring approximately 20V /degrees of eye-movement (Duchowski, 2002). Cortical Activity (Bereitshaftspotential) Continuous EEG data were collected and amplified 5000 times using the BIOPAC EEG amplifier (EEG100B; BIOPAC Systems, Inc., Sant a Barbara, CA), with a bandpass range from DC to 70 Hz. Analog data were sampled at 1000 Hz using an MP 150 analog/digital converter and recorded on-line with AcqKnowledge 7.0 (B IOPAC Systems, Inc., Santa Barbara, CA) software. A digital marker was generated usi ng LabVIEW 8.0 (National Instruments, Austin, TX) to facilitate the identification of the EMG fiducial time point on the EEG trace to indicate the onset of the putting stroke and correspondi ng BP waveform necessary for post acquisition analysis. Electromyogram To determine movement onset, electromyogram (EMG) activity was collected and amplified 5000 times using a BIOPAC EMG amp lifier (EMG 100B; BIOPAC Systems, Inc., Santa Barbara, CA), with a bandpass range from DC to 70Hz. Analog data were sampled at 1000Hz using an MP 150 analog/digital converter and recorded on-line with AcqKnowledge 7.0 (BIOPAC Systems, Inc., Santa Barbara, CA) softwa re. The EMG data were rectified and used to obtain the fiducial point for averaging the EEG and QE as described below.

PAGE 142

142 Anxiety To assess the modulation of anxiety levels across conditions, the Mental Readiness FormLikert (MRF-L; Krane, 1994) was implemented prior to each putt. Developed with the intention of assessing anxiety levels pr eceding and during competition, th e MRF-L is a convenient and practical alternative to the Competitive State Anxiety Inventory-2 (CSA I-2; Martens et al., 1990). The MRF-L is a three-item assessment, on an 11-point scale, of cognitive anxiety (worried not worried), somatic anxiety (tense not tense) and self-c onfidence (confident not confident). Given that the CSAI-2 is considered the criterion measure of anxiety in sport, reported correlations of 0.76, 0.69, and 0.68 for c ognitive anxiety, somatic anxiety, and selfconfidence between the MRF-L and CSAI-2 dimensions respectively, confirm the MRF-Ls utility in sport (Krane, 1994). For the purpose of this investig ation only the dimensions of cognitive and somatic anxiety were assessed. Heart Rate To measure heart rate, ECG activity was co llected using pre-gelled disposable snap electrodes located on the anterior portion of the left and right forearms. Analog data were sampled at 1000Hz using a BIOPAC EMG amplifie r (ECG 100C; BIOPAC Systems, Inc., Santa Barbara, CA) and was recorded on-line with AcqKnowledge 7.0 (BIOPAC Systems, Inc., Santa Barbara, CA) software. Procedure Upon arriving for testing, partic ipants were informed of the general purpose of the investigation and were provided with a brief to ur of the testing equipment and apparatus. Following the tour, each participant was asked to read and complete a university approved informed consent document and a brief demogra phic questionnaire. Participants were also permitted to ask any questions regarding the experiment.

PAGE 143

143 Upon providing consent, part icipants were prepared for electro-ocular (EOG), electromyographic (EMG), heart rate (ECG) and electroencephalographic (EEG) measurement in accord with the guidelines put forth by the So ciety for Psychophysiological Research (Pivik, Broughton, Coppola, Davidson, Fox, & Nuwer, 1993). Ve rtical (VEOG) and Horizontal (HEOG) bipolar electro-oculograhpic move ments were collected to assess QE duration and to control for ocular artifact in the EEG waveform. Four 4mm Biopac silv er/silver chloride (Ag/AgCl) electrodes (EL204) were positione d above and below the right ey e, and lateral to each eye, adjacent to the left and right orbital fossi. EMG activity was recorded using two 10mm s ilver/silver chloride (Ag/AgCl) electrodes placed 3 cm apart over the muscle belly of the extensor carpi ulnaris (ECU) of the right arm. EMG activity was sampled at 1000Hz and amplified (x 5000) using the Biopac EMG amplifier (EMG 100B). ECG activity was recorded using two pre-gelled snap electrodes placed over the radial artery of the left and right forearm. ECG ac tivity was sampled at 1000Hz and amplified (x 5000) using the Biopac ECG amplifier (ECG 100C). The continuous EEG was recorded with an a rray of 6 silver/silve r chloride (Ag/AgCl) electrodes in accord with the International 10 system (Jasper, 1958) using a lycra electrode cap manufactured by Electrode-Cap International, Inc. (ECI, Ea ton, OH). A central cluster of electrodes was positioned over the le ft, mid-line, and right central (primary motor cortex: C3, Cz, C4) sites to concentrate on those cortical regions known to have implications in motor planning and execution, as well as source generators of the BP (Orgogozo et al. 1979; Roland et al. 1980; Ikeda et al. 1993; Rektor et al. 1994; Huckabee, et al. 2003). Furthe rmore, an additional cluster of electrodes was positioned over the parietal co rtex (P3, P4), a region associated with visuo-

PAGE 144

144 motor control and suspected to have implica tions in QE functioning (Brunia & van Boxtel, 2000). All sites were referenced to linked ears. The mid-frontal (FPz) site served as the ground with electrode impedance being kept below 5 k After being outfitted with the requisite phys iological attire, each participant was individually tested. The testing session consiste d of 10 practice trials followed by 90 additional trials under two counter-balanced conditions (i.e., 45 putts per condition) designed to manipulate state levels of anxiety (i.e., lo w anxiety, high anxiety). The pract ice session served to acclimate the participant to the testing equipment and appara tus. Given that all participants were skilled golfers, a learning effect was not expected, thereby justifying th e minimal number of practice putts. In accord with previous research (e.g. Hardy et al., 1996, Masters, 1992; Murray & Janelle, 2003), the low anxiety c ondition consisted of simple, non-ev aluative directives, in which participants were asked to perfor m to the best of their ability so that the researcher can better understand the characteristics and behaviors associated with the golf putt. Conversely, the high anxiety condition was comprised of the addition of a video camera and the completion of a written release permitting the use of the video footage for broadcast on a nationally televised news program. Last, participants were informed that their performance would be compared to all other participants and that they would be competing for a $100 cas h prize. In accord with BP protocol, participants initiated each movement at their own volition, free from external cues or prompts. The testing session took approxima tely120 minutes, includi ng time for set-up, familiarization, practice, testing, equipment removal, and debriefing.

PAGE 145

145 Data Reduction Putting Performance Putting performance was quantified as the percen tage of total putts made per condition and was used to stratify the QE and BP data into h it and miss categories for subsequent analysis. Furthermore, missed putts were recorded on an (x,y) coordinate system to further specify performance error, including the assessment of bi as and consistency between levels of expertise (Reeve, Fischman, Christina, & Cauraugh, 1994; Hancock, Butler, & Fischman, 1995). The use of the coordinate system permitted the quantification of radial error, (RE), groupcentroid radial error (GRE), and bivariate vari able error (BVE). Simply stated, each metric provided an index of accuracy, bi as, and consistency with respect to the center of a target respectively (Hancock et al., 1995). RE is simply the non-direction single trial distance from the target. GRE indexes the overall magnitude and direction of bias Finally, BVE, defined as the average deviation about the target, allowed for in ference regarding the consistency or lack there of, of a given participants performance. As su ch, the BVE is an index of the intra-subject variability across trials. Electromyogram Given that the onset of EMG activity corresponding with the initiation of the putting stroke is critical to the subsequent da ta reduction and analysis of the dependent measures (i.e., BP and QE), a custom program written in LabVIEW 8.0 (N ational Instruments, Austin, TX) was used to coordinate EMG onset with the off-line reduction of eye-movement and BP data. Heart Rate (BPM) Using AcqKnowledge 7.0 (BIOPAC Systems, In c., Santa Barbara, CA) software, the Rwave data were transformed to BPM in order to obtain the mean hear t rate for the 5-second period prior to movement onset.

PAGE 146

146 Gaze Behavior Gaze behavior, indicated by recorded changes in the corneal-retinal potential as the eye navigates the visual display, was reduced off-line. As the eye moves in orbit, the corneal-retinal potential deviates approximately 20 V /degree of eye-movement (Duchowski, 2002). Given that the eye is known to perturbate approximately 5 degrees to maintain a fixation and that movements greater than 5 degrees suggests a shift in visual gaze (i.e., saccade), a predetermined threshold corresponding with 5 de grees of movement was established. That is, eye-movements greater than 5 degrees or 100 V was used to objectively measure gaze behavior. The measurement and reduction of gaze behavior served two objectives: 1) to ensure an artifact free recording of the BP, and 2) to de note the onset of QE. Ey e movements (e.g., blinks, saccades) exceeding the aforementioned threshold occurring within the 1500 ms preparatory period immediately prior to movement onset resulted in the rejection of that trial from further BP consideration and analysis. The QE was determined as the time between the last deviation in the corneal-retinal potential exceeding the predetermi ned threshold and the onset of the EMG burst recorded from the ECU of the right forearm. Cortical Activity Three common components of the BP (i.e., early, late, and peak BP) were evaluated. First, a digital trigger generated by LabVIEW 8.0 during th e acquisition of cortical data were used to locate the EMG burst respective to movement onset. For each tria l across each electrode site, a custom LabVIEW analysis program calibrated the aforementioned burst of EMG onset to establish a fiducial time point co rresponding with the initiation of the putting stroke. Data were then statistically and visually inspected for ey e-movement and muscular artifact. If rendered clean, the data were reduced to a 3500 ms epoch, in which 2500 ms of data pre-movement and

PAGE 147

147 1000 ms of data post movement wa s retained for the subsequent analysis of the BP and its constituents. All retained data were baseline corrected (McAdam & Rubin, 1970). Baseline was operationally defined as the fi rst 1000ms epoch preceding BPearly onset. The data were parsed according to group (i.e., HH and LH), for which four conditions (i.e., high-anxiety-hit, highanxiety-miss, low-anxiety-hit, a nd low-anxiety-miss) per participan t, free of muscular and eyemovement artifacts were deri ved, averaged, and measured. Following the reduction and classification of data, a grand average BP was generated according to level of expertise, anxiety, and perf ormance to facilitate subsequent analyses. For each average across each electrode site, a custom LabVIEW analysis program was used to detect and calculate the BPearly, BPlate, and BPpeak components, measured as follows. It has been routinely established that BP onset, if present, should appear approximately 1500 ms prior to EMG onset (Jahanshahi & Hallett, 2003), and was therefore declared as the point of demarcation for subsequent calculations. Speci fically, the duration of the BPearly is the difference in time from the point of BP onset (i.e., 1500 ms) to the start of the BPlate component. The BPlate component becomes visually apparent 400-500 ms pr e-movement and concludes with the BPpeak (Deecke et al., 1969; Shibasaki et al., 1980). The BPpeak was calculated as the average amplitude occurring 100 ms prior to movement onset (Slobounov, Tutwiler, Rearick, & Challis, 1999). The BPlate was calculated as the mean amplitude beginni ng 500 ms prior to movement onset, concluding 100 ms prior to EMG onset. Lastly, the BPearly was calculated as the average amplitude of the interval occurring 1500-500 ms prior to EMG onset. Data Analysis The following section outlines the data anal ytic procedures for each of the eight aforementioned hypotheses. The calculation of effect size estimates (i.e., Cohens d ) served to quantify skill-based differences across the variety of dependent measures.

PAGE 148

148 Hypothesis 1 Cognitive and somatic anxiety (MRF-L) and hear t rate (BPM) data were evaluated using separate one-way repeated measure ANOVAs to determine if the presentation of a monetary incentive, video camera, and written release would modulate anxiety levels under the high anxiety condition. Hypothesis 2 Given that the LH group was expected to outpe rform and exhibit a prolonged QE period as compared to the HH group across anxiety conditi ons, both putting performance (RE, GRE, BVE) and QE duration were evaluated using a repeated -measures multivariate analysis of variance (RM MANOVA). Given the reported association of QE duration to performance, MANOVA procedures were preferred over separate univariate ANOVAs. Putting performance and QE duration were analyzed using a 2 ( Skill : LH, HH) x 2 ( Anxiety : High, Low) MANOVA with repeated measures on the last factor. Post hoc procedures included univariate ANOVAs, each at the .05 level (Stevens, 2002). The success to failure ratio between LH and HH groups was assessed using the Chi square statistic. Furthe rmore, to specifically address expertise, performance, and anxiety diffe rences on QE duration a 2 ( Skill : LH, HH) x 2 ( Accuracy : Hit, Miss) x 2 ( Anxiety : High, Low) ANOVA with repeated-meas ures on the last two factors was conducted. Lastly, a Pearson Product Moment Correlation was calculated to explore the relationship between QE duration and RE. Hypothesis 3 Quiet-eye duration has been demonstrated to account for both inter and intra-group performance variability (Janelle et al, 2000; Vickers, 1992,1996a, 1997, 1998; Vickers et al., 1997, 1998). Accordingly, both inter and intra-group di fferences in QE duration were analyzed using a 2 ( Accuracy : Hits, Misses) x 2 ( Skill : LH, HH) ANOVA.

PAGE 149

149 Hypothesis 4 Given that the LH group was exp ected to exhibit a greater BPlate amplitude coupled with a greater BPpeak amplitude compared to the HH group collapsing across anxiety conditions, BP data were evaluated using a repeated-measures multivariate analysis of variance (RM MANOVA). Given the anticipated association of the three BP components (i.e., Early, Late, Peak) across cortical regions (i.e., C3, Cz, C 4, P3, P4), MANOVA proce dures were preferred over separate univariate ANOVAs. Co rtical activation in each of the BP components across skill level and cortical region was analyzed using a 2 ( Skill : LH, HH) x 3 ( BPcomponent: Early, Late, Peak) MANOVA with repeated measures on the la st factor. Post hoc procedures included univariate ANOVAs, each at the .05 level. Hypotheses 5 and 6 The various components of the BP were predicted to account for intra-group (i.e., collapsing across skill) performance variab ility. That is, the amplitude of the BPearly, BPlate and BPpeak, across cortical regions, were predicted to discriminate between putts made and putts missed regardless of skill level. The mean amplitude of the BPearly, BPlate and the BPpeak, were analyzed using 2 ( Accuracy : Hits, Misses) x 3 ( BPcomponent: Early, Late, Peak) MANOVA with repeated measures on the last factor. Furthermor e, as perceived levels of anxiety increase, a relative increase in the mobiliza tion of cognitive resources was expected to occur in order to successfully complete the task at hand. To a ssess relative changes in cortical activity under varying levels of perceived anxiety, the mean amplitude of the BPearly, BPlate and the BPpeak, were analyzed using 2 ( Anxiety : High, Low) x 3 ( BPcomponent: Early, Late, Peak) MANOVA with repeated measures on the last factor. Post hoc procedures included univariate ANOVAs, each at the .05 level.

PAGE 150

150 Hypothesis 7 The relationship between the amplitude of the BPpeak and QE duration was evaluated using a Pearson Product Moment correlation. Additiona lly, Pearson Product Moment correlations were conducted to assess the relationship between rela tive anxiety and QE duration and anxiety and cortical activation within each of the BPcomponents across cortical regions. Hypothesis 8 As anxiety increased it was expected that both the LH and HH groups would exhibit a longer QE period. To evaluate the mean QE dura tion differences in the high anxiety condition as compared to the low anxiety condition, across skill levels, a repeated measures ANOVA was employed. Specifically, QE duratio n was analyzed using a 2 ( Skill : LH, HH) x 2 ( Anxiety : High, Low) ANOVA with repeated measures on the last factor.

PAGE 151

151 CHAPTER 4 RESULTS Participants Ten expert (low handicap) and 10 near-exp ert golfers (high handicap) golfers were recruited from various golf clubs in the Southeas tern United States. To ensure that performance differences were not attributable to random moderating effects pa rticipant charac teristics were held constant with the exception of handicap, practice, and competitive playing experience. Specifically, although the number of years of golf experience ( t(18) = .884, p < .360, d = .20) were similar between the participants, the LH group engaged in significantly more practice ( t(18) = 4.191, p < .001, d = .70) and competitive playing experience ( t(18) = 3.892, p < .001, d = .68), completing an average of 10.80 ( SD = 4.64) competitive events compared to 3.2 events for the HH group ( SD = 4.07). Pre-Putt Levels of Cognitive Anxiet y, Somatic Anxiety, and Heart Rate To assess the effectiveness of the anxiety ma nipulation (i.e., use of monetary incentive, video camera, and a written release) for evoking changes in both cognitive and somatic anxiety levels, and to determine the impact of anxiety an d arousal on quiet eye du ration, cortical activity and subsequent golf putting performance, pre-putt levels of cognitive and somatic anxiety were measured using the Mental Readiness Form Likert (MRF-L; Krane, 1994). To provide an estimate of arousal differences across anxiet y conditions, heart rate (HR) was analyzed. The analysis revealed a significant Condition main effect for the dependent variables: Cognitive Anxiety ( F(1, 895) = 132.18, p < .001, d = .77), Somatic Anxiety ( F(1, 895) = 121.93, p < .001, d = .74) and HR ( F(1, 895) = 172.99, p < .001, d = .88). Suggesting that across skill conditions, both Cognitive Anxiety and HR was greater in the high a nxiety condition as compared to the low anxiety conditi on. Moreover, a significant Skill x Condition interaction for Somatic

PAGE 152

152 Anxiety ( F(1, 895) = 27.41, p < .001, d = .35) was evident indicati ng that the LH group reported greater somatic anxiety in the high anxiety condition as compar ed to the low anxiety condition. In comparison, the HH group reported greater somatic anxiety in the low anxiety condition than their LH counterparts and more a nxiety in the low anxiety conditi on relative to the high anxiety condition. Figure 4-1 Mean cognitive anxiet y (Figure 4-1a), somatic anxiet y (Figure 4-1b), and heart rate (Figure 4-1c) across skil l and anxiety conditions. Significant Skill main effects for Cognitive Anxiety ( F(1, 895) = 131.60, p < .001, d = .77), Somatic Anxiety ( F(1, 895) = 68.37, p < .001, d = .55) and HR ( F(1, 895) = 97.57, p < .001, d = .66) were also found. Follow-up univariat e analyses revealed that th e HH group reported lower levels of Cognitive Anxiety than the LH group in both the low anxiety ( F(1, 898) = 113.45, p < .001, d = .71) and high anxiety ( F(1, 895) = 85.06, p < .001, d = .62) conditions. Conversely, the HH group reported greater levels of Somatic Anxiety than the LH group in the low anxiety condition ( F(1, 898) = 107.34 p < .001, d = .69), yet significantly lower rati ngs in the high anxiety condition ( F(1, 0 1 2 3 4 5 6 7 8 9 10 11Cognitive Anxiety HH LH HH LH High Anxiety Low Anxiety 0 1 2 3 4 5 6 7 8 9 10 11Somatic Anxiety HH LH HH LH High Anxiety Low Anxiety 75 80 85 90 95 100 105 110 115 120 125Heart Rate (BPM) HH LH HH LH High Anxiety Low AnxietyFigure 4-1a Figure 4-1b Figure 4-1c

PAGE 153

153 895) = 14.94, p < .001, d = .26). Lastly, elevated HR was evident among the HH group relative to the LH group in both the low anxiety ( F(1, 898) = 84.64, p < .001, d = .62) and high anxiety ( F(1, 895) = 90.07, p < .001, d = .63) conditions. Figure 4-1 provides a graphic representation of these findings. Skill Based Putting Performance and Quiet-E ye Differences Across Anxiety Conditions It was hypothesized that across both the lo w and high anxiety c onditions, the LH group would perform better than the HH group as measured by RE, BVE, and total number of putts made. Furthermore, the expert participants we re expected to exhibi t a longer QE period than the near-expert group. The LH group was more successful than the HH group ( 2 = 33.59, p < .001, d = .19) but did not differ in the magnitude of their bias (GRE). Pillais Trace was used to interpret the MANOVA results for putting performa nce and QE duration since it is deemed more robust than the alternative test statictics (Liu, 2002). The omni bus test for putting performance and QE duration between the LH and HH particip ants across anxiety conditions was significant, Pillais Trace = .089, ( F(3, 893) = 28.90, p < .001, 2 = .089). Follow-up univariate analyses of variance for QE ( F(1, 895) = 41.509, p < .001, d = .43), RE ( F(1, 895) = 36.305, p < .001, d = .40), and BVE ( F(1, 895) = 34.753, p < .001, d = .39) yielded significant di fferences between skill levels, indicating that the LH group was more accurate a nd consistent, while demonstrating a longer QE duration relative to the HH group (Figures 4-2 and 4-3). However, no significant differences were noted for Anxiety Pillais Trace = .006, ( F(3, 893) =1.735, p < .158, 2 = .006). Figure 4-2 and 4-3 provide a graphic repr esentation of the results. Furthermore, the analysis of expertise, perf ormance, anxiety differe nces, and QE duration revealed a significant main effect for Skill ( F(1, 379) = 35.211, p < .001, d = .61) suggesting that the LH group engaged in a longer QE period relati ve to the HH group across conditions. Despite

PAGE 154

154 the notable Skill differences in QE duration, no differences were noted for Accuracy ( F(1, 379) = 2.628, p = .106, d = .17), suggesting that QE duration was relatively consistent for both putts made and putts missed while controlling for skil l. Although QE duration in creased from the low anxiety to the high anxiety condi tion for both the LH and HH part icipants, this finding was not significant, ( Anxiety F(1, 379) = .101, p = .750, d = .03). No other differences were noted (i.e, Skill x Accuracy x Anxiety F(1, 379) = .045, p = .832, d = .02). Figure 4-2 Performance variability of the HH and LH groups are indicated as the distance from the target and the magnitude of performance bias across anxiety conditions (i.e., Group Centroid Radial Error [GRE]). Figure 4-3 Prolonged QE duration across skill but not anxiety highlights the trend supporting the expert advantage. 1300 1500 1700 1900 2100 2300 2500 2700 2900Quiet Eye Duration (ms) HH L H HH LH High Anxiety Low Anxiet y -80 -60 -40 -20 0 20 40 60 80 -25-20-15-10-50510152025 Lateral Error (cm) Low Anxiety High Anxiety GRE Low Anxiety GRE High Anxiety -80 -60 -40 -20 0 20 40 60 80 -25-20-15-10-50510152025 Lateral Error (cm) Low Anxiety High Anxiety GRE Low Anxiety GRE High AnxietyError Score High Handicap Error Score Low Handicap Error Scores (cm)

PAGE 155

155 1300 1500 1700 1900 2100 2300 2500 2700 2900Quiet Eye Duration (ms) HH LH HH LH Miss HitLastly, Pearson Product Moment correlations were conducted to expl ore the relationship between QE duration and RE, and QE duration a nd anxiety. Results indicat ed a non-significant correlation ( r = .046, p = .389, d = .09) between QE and RE, a nd a non-significant correlation between QE and anxiety (r = -.059, p = .359, d = -.12). Interand Intra-Group Performance Variability in Quiet-Eye Duration Assesssing both interand intra-group QE di fferences while controlling for anxiety, it was hypothesized that the QE duration of both th e LH and HH groups for successful putts would exceed the QE duration for missed putts. Furtherm ore, after collapsing across skill, a prolonged QE duration was expected for succesful putts re lative to missed putts. The results yielded a significant main effect for Skill ( F(1, 1793 ) = 51.989, p < .001, d = .34) but not for Accuracy ( F(1, 1793) = 3.323, p = .068, d = .08), suggesting that expertis e is reflected by a prolonged QE duration No other differences were noted (i.e., Accuracy x Skill interaction, F(1, 1793) = .304, p = .581, d = .03, Figure 4-4). Figure 4-4 When controlling for anxiety, the LH participants demonstrate longer quiet eye durations for successful putts as compar ed to missed putts. Conversely, minimal variability in quiet eye durat ion is evident for the HH participant as a function of performance, controlling for anxiety.

PAGE 156

156 BP Activity Across Skill Le vel and Cortical Region To investigate the hypothesis that both increased negativity in mean BPlate and BPpeak amplitude are characteristics of greater movement preparation and cerebral efficiency, cortical activation levels in each of th e BP components (i.e., early, late, peak) across skill level and cortical region (i.e., C3, Cz, C4, P3, P4) wa s assessed. As anticipated, a non-significant Skill x BPcomponent interaction was found (Pillais Trace = .177, ( F(10, 66) = .640, p = .774, 2 = .177). An overall significant difference in cortical ac tivity was evident for the main effect of Skill Pillais Trace = .690, ( F(5, 14) = 6.245, p = .003, 2 = .690) and BPcomponent, Pillais Trace = .483, ( F(10, 66) = 2.103, p = .036, 2 = .242). Combined, the significant Skill and significant BPcomponent main effects suggest that not only did the LH group demonstrate more BP negativity relative to the HH group, cortical negativity al so increased from the BPearly to BPlate component for both groups, reaching maximal negativity immediatel y prior to movement execution (i.e., BPpeak). Follow-up univariate analyses of variance for the main effect of Skill revealed significant cortical region differences for C4 ( F(1, 18) = 14.171, p = .001, d = 1.77) and P4 ( F(1, 18) = 8.304, p < .010, d = 1.36). Given the relative degree of relatedness am ong BP components, a multivariate analysis of variance was used to examine the temporal loca tion of the cortical re gion differences between groups (Figure 4-5). Results indi cated that for C4, the LH group exhibited greater negativity for each BPcomponent (C4early, F(1, 18) = 6.023, p = .025, d = 1.16; C4late, F(1, 18) = 17.519, p = .001, d = 1.97; and C4peak, F(1, 18) = 27.425, p < .001, d = 2.47) compared to the HH group, while parietal differences were only evident between th e two groups during the early component, P4early ( F(1, 18) = 7.661, p = .013, d = 1.30).

PAGE 157

157 Figure 4-5 Skill based differences (i.e., mean, SE) acros s cortical regions and BP components. Figure 4-5a displays a marked increase in left-central negativity across BP components for the LH group. Figure 4-5b illust rates a significant increase in rightcentral negativity across BP components for the LH group. Figure 4-5c displays a marked increase in negativity at the vert ex across BP components for the LH group. Figure 4-5d illustrates minimal hemispheric differences in the left-parietal region between skill. Figure 4-5e illustrates an in crease in right-parietal cortical negativity for the BPearly component (* represents p < .05). Figure 4-5d Figure 4-5e -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HH LH Early Late Pea k C3 BP Component -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 HH LH Early Late Peak C4 BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HH LH Early Late Peak CZ BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HH LH Early Late Pea k P3 BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HH LH Earl y Late Pea k P4 BP Component V V V V V Figure 4-5a Figure 4-5b Figure 4-5c *

PAGE 158

158 BP Activation and Putting Outcome To investigate the hypothesis that an increase in BPpeak amplitude is characteristic of greater task involvement and sensorimotor efficiency, cortical activation leve ls in each of the BP components (i.e., BPearly, BPlate, and BPpeak) for putts made and missed served as the dependent measures of interest for each cortical region. Although cortical ne gativity increased across each of the BPcomponents, reaching maximal negativity immediat ely prior to movement execution (Pillais Trace = .449, ( F(10, 146) = 4.223, p < .001, 2= .224), the omnibus test for Accuracy failed to reach significance, (Pillais Trace = .031, ( F(5, 34) = .219, p = .952, 2 = .031), suggesting that BP negativity did not vary as a function of putting accuracy (i.e., hit or miss). No other differences were noted. Figure 4-6 provides a graphical depiction for accuracy and BP. Anxiety and BP Activity Given that anxiety may serve to increase mo tivation and thus result in greater task involvement, it was hypothesized that an increase in anxiety would result in greater cortical negativity across each BPcomponents (i.e, BPearly, BPlate, and BPpeak). As such, cortical negativity was assessed under high and low anxiety conditions for each cortical regi on (i.e., C3, Cz, C4, P3, P4). Although cortical negativity increased across each of the BPcomponents, maximal negativity was achieved immediately prior to move ment execution (Pilla is Trace = .433, ( F(10, 146) = 4.223, p < .001, 2 = .216). The omnibus test for Anxiety failed to reach significance, (Pillais Trace = .113, ( F(5, 34) = .864, p = .515, 2 = .113), suggesting that BP negativity was similar across anxiety conditions when controlling for skill or accuracy. However, Pearson a Product Moment correlation was conducted to addr ess the relationships between anxiety and BP negativity. A significant positive relationship between anxiety and BP was evident across several cortical regions and BP components (Tab le 4-1), suggesting that relativ e increases in anxiety may be

PAGE 159

159 reflected cortically. Figure 4-7 provides a graphi cal depiction of the rela tionship between cortical activation and anxiety. Figure 4-6 Performance differences across cortical regions and BP components. Figure 4-6a and Figure 4-6b display marked BP negativ ity across components with minimal differences between hits and misses for left-central and right-central regions respectively. Figure 4-6c illustrates pronounced BPpeak negativity at the vertex. Figure 4-6d and Figure 4-6e illustrate greater BPpeak negativity for left and right parietal regions. Quiet-Eye Duration and BPcomponent Activation It has been documented that both the QE period and the cortical and sub-cortical generators associated with the BP are responsible for th e orientating of visual attention and the V V V V V Figure 4-6a Figure 4-6b Figure 4-6c Figure 4-6d Figure 4-6e -4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 Miss Hit Early Late Peak C3 BP Component -4 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 Miss Hit Early Late Peak C4 BP Component -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 Miss Hit Early Late Peak CZ BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 Miss Hit Early Late Peak P3 BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 Miss Hit Early Late Peak P4 BP Component

PAGE 160

160 execution of a self-paced task. As such, a Pearson Product Moment correlation was conducted to explore the relationship between QE duration and BP. Although emphasis is placed on the BPpeak Quiet-Eye duration relationship, the a sssociation between each component of the BP and QE was assessed. Results indicated a significan t correlation between QE duration and C3peak ( r = .3096, p = .026, d = .65), C4peak ( r = .2874, p = .036, d = .60), and Czpeak ( r = .2901, p = .035, d = .61), suggesting that as QE duration increased so too did BP negativity within the specified regions. No other significant correlations were found (P3peak ( r = .1696, p = .148, d = .34); P4peak ( r = .1574, p = .166, d = .32). Table 4-1 Pearson Product Moment correlations dem onstrating the regional specificity associated with the relationship between a nxiety and cortical activation. Quiet-Eye Duration and Anxiety It was hypothesized that both the LH and HH group would e xhibit a longer QE duration across both the low and high anxiety conditions. That is, an increase in the QE duration is believed to circumvent the deleterious effects of anxiety while maintaining performance. Results indicated that no si gnificant differences were noted for Anxiety ( F(1, 18) = .002, p =.963, d = .02). Figure 4-3 shows the null differences in which QE duration was stable for each skill group across anxiety conditions. C3peakr = 0.28 p = 0.04* d = 0.58 Left-Central C3later = 0.354 p = 0.013* d = 0.76 C3earlyr = 0.339 p = 0.016* d = 0.72 C4peakr = 0.306 p = 0.027* d = 0.64 Right-Central C4later = 0.466 p < 0.001* d = 1.05 C4earlyr = 0.343 p = 0.015* d = 0.73 Czpeakr = 0.211 p = 0.096 d = 0.43 Midline Czlater = 0.288 p = 0.036* d = 0.6 Czearlyr = 0.326 p = 0.02* d = 0.69 P3peakr = 0.392 p = 0.006* d = 0.85 Left-Parietal P3later = 0.11 p = 0.249 d = 0.22 P3earlyr = 0.005 p = 0.487 d = 0.01 P4peakr = 0.482 p < 0.001* d = 1.1 Right Parietal P4later = 0.112 p = 0.246 d = 0.23 P4earlyr = -0.073 p = 0.328 d = -0.15 *Denotes significant correlation p <.05.

PAGE 161

161 Figure 4-7 Depicts nonsignificant tre nds in cortical negativity within the three components of the BP for each cortical region for the HA c ondition as compared to the LA condition across skill level. V V V V V Figure 4-7a Figure 4-7b Figure 4-7c Figure 4-7d Figure 4-7e -5 -4 -3 -2 -1 0 1 2 3 4 5 HA LA Early Late Peak C3 BP Component -4.5 -3.5 -2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 HA LA Earl y Late Pea k C4 BP Component -4.5 -3.5 -2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5 HA LA Earl y Late Pea k CZ BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HA LA Early Late Peak P3 BP Component -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 HA LA Early Late Peak P4 BP Component

PAGE 162

162 CHAPTER 5 DISCUSSION, APPLIED IMPLICATI ONS, AND FUTURE DIRECTIONS The quest to understand the nature of sport e xpertise has lead resear chers to explore the psychophysiological characteristics of superior performance. The resu lting evidence suggests that psychological efficiency underlies expert performance (Hat field et al., 1999), which is characterized by both regional cort ical deactivation (i.e., increased alpha power, Haufler et al., 2000) and effective and efficient cue utilization (i.e., fewer fixa tions of longer duration and prolonged QE, Mann et al., 2006). However, with the exception of the early work of Janelle, Hillman and Hatfield (2000) and Janelle et al., (2000) these two components of expertise have not been simultaneously explored. According to Vickers (1996a), the quiet-eye is a temporal period when task relevant environmental cues are processed and motor plan s are coordinated for the successful completion of an ensuing task. As such, the QE period theore tically represents the time needed to organize the neural networks and visual parameters resp onsible for the orienting and control of visual attention to promote cerebral efficiency (Vic kers, 1996a). In addition to its motor planning function, researchers have suggest ed that the QE period may refl ect an opportunity for emotion regulation, thereby minimizing the deleterious eff ects of anxiety and/or arousal by permitting the recruitment of task specific re sources (Janelle, Hillman, & Hatf ield, 2000; Janelle et al., 2000; Vickers et al., 1999). Performance varies with the length of th e QE period, with prolonged periods generally resulting in increased performance. Th e importance of understanding the complex integration of systems associated with expert performance, how ever, is essential for the advancement of both theory and practice. Therefore, I sought to better understand the mechanisms that underlie the efficacy of the QE period by concurrently a ssessing QE duration, anxiety, and a premotor,

PAGE 163

163 electrophysiological index of cerebral efficien cy (i.e., the BP) among HH and LH golfers who performed a putting task unde r high and low anxiety. Discussion The use of a monetary incentive, video camera, and written release to manipulate levels of cognitive and somatic anxiety and physiological arousal was integral to the overall success of this investigation. The eff ectiveness of this manipulat ion is addressed next. Pre-Putt Levels of Cognitive Anxiety, Somatic Anxiety, and Heart Rate A central tenet of this investigation was to address the potential arousal regulation function of the QE period. Moreover, I explored the impact of heightened anxiety on the neural networks and visual parameters needed for effective perf ormance. The anxiety ma nipulation was predicted to elevate levels of cognitive and somatic anxiety as measured by the Mental Readiness Form Likert (MRF-L, Krane, 1994), and physiological ar ousal as measured by h eart rate (i.e., BPM). The HH and LH groups reported significantly higher cognitive anxiety ratings under the high anxiety condition as compared to the contro l (i.e., low anxiety) c ondition. Somatic anxiety scores revealed a similar pattern for the LH group, but the HH group reported less somatic anxiety in the high anxiety c ondition relative to the control condition. Furthermore, the high anxiety condition promoted heightened levels of arousal across skill le vel, as indicated by substantial increases in HR from the control condition. Although the HH groups scores on somatic anxiety were not in the hypothesized dire ction, the elevated hear t rate exhibited during the high anxiety condition by both skill groups supports the link between increased anxiety and corresponding changes in physiological arousal, and provides further support for the efficacy of the manipulation used in this investigation. It is clear that the anxiety ma nipulation was successful. Given that the each of the primary hypotheses put forth in this inve stigation were dependent on th e anxiety manipulation, it would

PAGE 164

164 have been impossible to draw any conclusions rega rding the role that the QE period may have in emotion regulation without a succ essful anxiety manipulation. Ther efore, the changes in anxiety and arousal permit a direct comparison of the ef fects of anxiety on the cortical and visual mechanisms deemed characteristic of expert performers. Skill Based Putting Performance and Quie t-Eye Differences Across Anxiety Conditions Since the seminal work of Vickers (1996a ), a growing body of evidence has been compiled supporting the notion that fixations of relatively longer duration are ev ident during the preparatory stages of a self-paced task, and can differentiate th e expert from the near-expert performer (Mann et al., 2006). Given that many of the hypotheses outlined in this investigation are dependent on overt and measur able skill based differences, several performance variables were of interest. Putting performance was meas ured as a function of hits and misses, and missed putts were additionally assessed for radial error (RE) and bivariate variable error (BVE). I hypothesized that the LH group would make more putts than the HH group, with reduced variability on missed putts a nd in turn, a longer QE period than the HH group across anxiety conditions. Results confirmed the hypothesized effects of Skill but not Anxiety That is, the LH group successfully holed more putts unde r both the low and high anxiety conditions while exhibiting a longer QE duration than the HH group. As expect ed, the LH group was also more consistent (i.e., less error) with nearer misses than the HH group. Anxiety failed to significantly alter putting accuracy when considered absolutely (i.e., hit, miss) and qualitatively (i.e., RE, BVE). However, for both the HH and LH groups, absolute and qualitative performance was superior in the high anxiety condition, although not statistically si gnificant. In retr ospect, the anxiety manipulation may have actually facilitated motivation and concentration in both groups. In other words, the high anxiety conditi on may not have been severe enough to promote performance

PAGE 165

165 decrement, but influential enough to enhance re source recruitment and subsequent performance (Eysenck, 1982; Eysenck & Calvo, 1992). A sim ilar finding was evident for QE duration, with the LH group exhibiting a slight increase in the QE duration in the high anxiety condition, while a slight decrease in the QE period was noticeable for the HH group. Contrary to expectations, the interaction of skill, performance, and anxiety, on QE duration failed to support the notion that QE durat ion would increase across skill levels in the high anxiety condition. It has been postulated (Janelle, Hillman, & Hatfield, 2000) and tenuously supported (Vickers et al., 1999; Williams, Singer, & Frehlich, 2002) that increases in anxiety and arousal result in a prolonged temporal window necessary for the regulation of emotion. Despite both groups improving their putting performance unde r the heightened anxiety condition, the QE duration increased only marginally for the LH group while it decreased for the HH group. Two explanations for this unexpected finding are plau sible: 1) QE duration does not serve an emotion regulation function but rather serves solely as a motor programming/movement preparation function; or 2) the anxiety condition did not adequately i nduce deleterious effects often associated with elevated cognitive and somatic anxiety (Jones & Hardy, 1990) to effectively determine the extent to which QE may serve an emotion-regulating function. Three distinct pieces of information favor the motor programming/movement preparation function over the latter interpretation. First, a non-significant correlation between anxiety and QE duration was reported, suggesting that QE duration varied irre spective of anxiety. Second, a significant correlation was found between QE du ration and BP. More specifically, as QE duration increased, cortical nega tivity increased around the primary motor cortex (i.e., C3, Cz, C4). Thus, although the assessment of QE dura tion for hits and misses failed to reveal statistically significant diffe rences, the aforementioned relationships provide insight into the

PAGE 166

166 utility of the QEs temporal window for organizi ng relevant cognitive and perceptual cues for improving performance. Third, because both the BP and QE duration are believed to have motor programming implications and we re significantly relate d in the current stu dy, further support for the motor programming hypothesis is evident. Furthe rmore, the lack of relationship between the QE period and anxiety suggests that the QE duration may not serve an emotion regulation function, but that such regulation may take place in the complex pathways linking the cortical and subcortical generators of the motor progr am. Although speculative, th e BP has been shown to reflect changes in arousal and motivation (Andreassi, 1980; McAdam & Seales, 1969), which may circumvent the need for additional regulatory processes. As a result, the QE period may simply regulate the type and amount of visual in formation needed to evoke the requisite motor program, rather than modulate th e effects of anxiety itself. Interand Intra-Group Performance Variability on Quiet-Eye Duration Research with golfers (Vickers, 1992), marksm en (Janelle et al, 2000) basketball players (Vickers, 1996b), biathletes (Vic kers et al., 1999), and volleyb all players (Vickers, 1997, 1998) indicates that experts not only demonstrate a longer QE when compared to near-expert performers, but that the QE is associated with within group performance differences. I therefore hypothesized that variability in the duration of the QE period w ould account for inter and intragroup performance variability. The LH group wa s hypothesized to exhib it a longer QE period relative to HH group, and the QE duration of both the LH and HH groups was predicted to exceed the QE duration for missed putts compared to successful ones. In accord with expectation, the LH group not only performed better on the putting task, but also engaged a longer QE duration relative to the HH group. Contrary to previous research however, QE duration did not account for within group performance differences across skill levels. Caution however, must be taken when in terpreting these statistical findings. Despite the

PAGE 167

167 non-significant differences, the mean QE duration was longer for successful putts compared to missed putts for both the HH and LH groups. BP Activity Across Skill Le vel and Cortical Region The BP reflects the cortical mechanisms invol ved in movement preparation (Brunia & van Boxtel, 2000). In line with previous resear ch that suggests an increase in mean BPpeak and mean BPlate negativity characterize greater movement pr eparation and cerebral efficiency (Chiarenza, Vasile, & Villa, 1990; Papakostopoulos, 1978; Ta ylor, 1978), it was hypothesized that the LH group would show a greater BPlate and BPpeak amplitude compared to the HH group. Congruent with previous research (Deecke, 1987; Konttinen & Lyytinen, 1992; Shibasaki et al., 1980), data from the curr ent study indicated that cortical negativity continued to increase across each BP component, re aching maximal negativity in th e temporal window immediately prior to movement execution. Furthermore, as hypothesized, Skill based differences were discernible across cortical regions and BP compon ents. Specifically, cortical differences were evident over the right-central (C4early, C4late, and C4peak) and right-parietal regions (P4early), indicating the relativ e increase in attention allocation to th e visuo-spatial cues for the LH group to the HH group. Given the cortical specificity of th ese findings, it is reasona ble to conclude that the LH group allocated more attention to the vi suo-motor components of the putting task than their HH counterparts. A distingu ishing feature between experts a nd near-experts is the distinct cortical patterns of the expert performer. The fact that the cortical differences are evident before the golfer actually executes the stro ke suggests the efficient organi zation of task related neural networks (Milton, Solodkin, Hlustik, & Small, 2007). The initial increase in cortical ne gativity associated with the BPearly component as evidenced here (i.e., C4early and P4early) reflects the activat ion of the supplementary motor area, which may serve to retrieve and/or augment the requisite motor commands from memory

PAGE 168

168 (Roland, 1984). This finding supports the contention that the BP may play a role in the detection and pairing of task relevant information with the necessary components of movement (Brunia & van Boxtel, 2000), while reflecting the activation of a neural pa th linking perception to action (Toni & Passingham, 2003). More recently, whol e brain MRI data lend further support to the results presented here, suggesti ng that experts develop a specialized motor program evidenced by right brain activation that integr ates visual information with th e necessary motor commands for performance (Milton et al., 2007). The preparatory set of the perfor mer may also be reflected in BPlate changes. For example, Taylor (1978) repor ted increases in BPlate negativity in the hemisphere contralateral to the active limb. In contrast, the cortical di fferences revealed here (i.e., C4late) are ipsilateral to the dominant limb. The golf putt is a bimanual task by nature, how ever, it is often argued that an expert righthanded golfer will control the putt with his or he r right hand, with the le ft hand simply providing support. Although the ipsilateral cort ical differences reported here apparently contradict those of Taylor (1978), they may be explained by the bima nual coordination required for the golf putt. Data regarding the significant difference in activation of the BPpeak component of the right-central region (i.e., C4peak) for the LH group relative to the HH group is congruent with previous research (Deecke, 1987; Shibasaki et al., 1980). Of the three BP components, the BPpeak component is believed to reflect the coordi nated activation of the SMA and MI. Combined, activation of these structures play a critical ro le in the organization of complex motor sequences that are rehearsed from memory and fit into a pr ecise timing plan (Gerloff et al., 1988). As such, this finding is not altogether su rprising, given that the LH gr oup engages in significantly more practice and competition than the HH group. Indee d, the elite group should have a more refined cortical representation of the ta sk that facilitates the movement and timing patterns of the golf

PAGE 169

169 putt. Practice and experience may contribute to th e elevated right-central cortical activation of the LH group relative to the HH group, such that the preparatory period of th e LH player reflects an attentional process that permits the assessmen t, organization, and recall of the requisite motor program from memory. The HH player likely has not developed such refi ned control, therefore resulting in more deliberate cognitive processes. The majority of sport psychophysiological re search (Haufler et al., 2003) has adopted spectral techniques for inferring the cortical role in psychomotor performance. Although this investigation used a class of ERP (i.e., BP) to examine skilled-based differences, complimentary findings attained from both approaches can be garnered to implicate common cortical mechanisms. As indexed by the collective spectral findings to date, as i ndividuals become more skilled, the cognitive strategies used during th e preparatory process and movement execution stage become more routine, demanding fewer cogn itive resources (Fits & Posner, 1967; Hatfield et al., 1984). Direct comparisons of expert and near-e xpert performers have revealed cortical asymmetry differences, such that the expert perf ormer demonstrates a relative decrease in left compared to right hemisphere cortical activity (i .e., increased alpha power or cortical quieting). This finding suggests that near-e xpert performers require greater conscious processing of the task and its demands. The BP data reflect greater cortical activation in the right hemisphere relative to the left, and moreover, that the LH golfers maintain greater acti vation in the right hemisphere relative to the HH golfers. Corrobor ating the extant spectral wor k, BP evidence gathered in the current study suggests that the LH players al locate more resources to the visual-spatial processing of the task and fewer resources to the conscious processing of the movement, linking the visual-spatial area of the cortex to m ovement preparation and performance.

PAGE 170

170 BP Activation, Performance and Anxiety It has been suggested that the amplitude of the BPpeak is more pronounced with greater task involvement (McCallum, 1976) and is a cortical reflection of the preparatory set of the participant (Loveless & Sanford, 1974) indicative of sensorimotor efficiency. It was therefore hypothesized that an increase in mean BPearly amplitude, mean BPlate amplitude, coupled with an increase in BPpeak amplitude would be evident for putts made as compared to putts missed. Furthermore, because an increase in perceived anxiety can increase the relative complexity of a given task (Eysenck & Calvo, 1992), it is reasonable to expect that additi onal cognitive resources may be recruited to accommodate the increased effort necessary to sustain performance (Eysenck, 1982). In accord with Eysenck and Cal vo (1992), an increase in a nxiety can result in a decrease in processing efficiency due to the a dditional cognitive/attentio nal demands imposed of the performer necessary to complete the desired task. As a result, it was hypothesized that an increase in cortical negativity would be evident across BP components (i.e., BPearly, BPlate, and BPpeak) in the high anxiety condition as compared to the low anxiety condition due to a relative increase in task complexity and resource mobiliz ation (i.e., motivation) needed to successfully complete the task under elevated anxiety (Lang, et al., 1989). Unexpectedly, the BP differences between h its and misses and between the low and high anxiety conditions were minimal. Although these differen ces are not statistic ally significant, data were broadly congruent with the original hypotheses. For example, BPpeak amplitude was invariably more negative across cortical regions (i.e., C3, C4, Cz, P3, and P4) for successful trials relative to unsuccessful tria ls, a finding that parallels the ea rlier work of Crews and Landers (1993) with golfers, Landers et al ., (1991) with archers, and Kon ttinen and Lyytinen (1992) with marksmen. However, the inverse was true for the BPearly amplitude; greater negativity was apparent for unsuccessful trials. With respect to anxiety, both BPearly and BPpeak amplitude were

PAGE 171

171 more negative in the high anxiety condition as compared to the low anxiety condition, again suggesting that additional cognitive effort may have been necessary to offset the increased task demands imposed by the relative increase in anxi ety. Empirical support of this contention is provided by the positive and signifi cant correlations between a nxiety and BP negativity across cortical regions, suggesting that as anxiety increased so too di d BP activity (Table 4-1). The nature of the Anxiety-BP trend lends credence to the notion that changes in anxiety can result in greater task involvement and resource mobilizat ion, which can be reflec ted cortically. This finding corroborates the early work of Andreassi (1980) and McAdam an d Seales (1969) who demonstrated increased BP negativity with e nhanced motivation and in the presence of a monetary incentive. Although the aforementioned trend is in the de sired direction, the magnitude of the BP difference across performance conditions (i.e., hits and misses) was not significant. This lack of statistical difference may be accounted for in the dichotomy (i.e., hit, miss) used to classify putting performance. For example, it is not unr easonable for a well-executed putt to result in a miss and similarly for a poorly executed putt to result in a hit, a pattern that may confound the comparison. Moreover, all misses were classified together, so a ball that missed the target by a mere fraction was scored the same as a ball that missed the target by several feet. This methodological decision may have confounded the data analysis a nd subsequent findings. That being said, the magnitude of the true cortical di fferences across task performance may have been lost in classification. Furthermore, one plausible explanation for the lack of within subjec t associations between QE and putting outcome is the relative homogeneity of the groups. Although the groups in this study clearly differed in putting ability, the relati ve difference in ability between the LH (0-2

PAGE 172

172 handicap) and the HH (10-12 handicap) groups may not be as covertly evident. Previous research employing the expert-novice paradigm has typically assessed the overt and covert behaviors of vastly disparate groups (i.e., natio nal, international leve l performers versus absolute beginners). As a result, the existing literature base is comp rised of research that has capitalized on the pure magnitude of expected differences, and as a resu lt, practical inference is often futile. This study however, has revealed both overt and covert diffe rences between groups that differ in skill, but are similar in ability. As such, both the noted di fferences and lack thereof, may more accurately represent the true mechanisms of the expert advantage. Alternatively, the extent to which the anxiety manipulation was effective may be subject to debate. That is, although both the LH and HH groups reported anxiety differences, and these differences were supported by objective physiological changes, the additional cognitive demands imposed may not have been sufficient enough to warrant changes in the recruitment and allocation of cognitive resources that would be reflected in distinct cortical patterns. Practical Implications and Future Directions To obtain expert status, athletes must excel in no less than four domains: physiological, technical, cognitive (tactical/strategic; perceptual/decision-making), and emotional (regulation/coping; psychological) (Janelle & Hillman, 2003). This investigation corroborates the notion that expert and near-exper t athletes not only differ in th eir performance proficiency, but they also differ in their underlying psychologica l mechanisms responsible for performance (i.e., cognitive). The extended QE period of the LH group relative to the HH group speaks to the cognitive advantage of the expe rt. The significant relationship be tween right-central (i.e., C4) cortical activation and QE dura tion support the notion of relativ e sensorimotor efficiency of expert athletes. It is well understood that expert performers maintain cortical activation levels that permit a more efficient and effective perf ormance outcome. The visual search literature

PAGE 173

173 contends that the active search of the environment beyond a cert ain point provides redundant and often distracting sources of info rmation. However, prolonged fixations, such is the case with the QE period, permit the detailed processing of in formation and even the cortical organization necessary for performance. As such, once a basi c understanding of the re quisite environmental cues and movement sequences have been learned, QE training may facilitate the relative cortical quieting necessary to perform at a higher level. In other words, although deliberate practice is believed to result in the automaticity of moveme nt, systematically training the QE may augment practice effectiveness by allowing the mind to pro cess the visual-spatial characteristics of the task while permitting the organization of the ne ural networks responsible for movement. With specific reference to golf, when addres sing the putt, the typical golfer will spend a brief moment estimating the distance, speed, and lin e of putt to the target. However, at the onset of the putting stroke, most non-expert players revert to conscious processing of the putting stroke. For example, Vickers ( 2004) reported that non-expert golfe rs often track the putter-head as it traverses back and through ba ll contact, a behavior not as evident in highly skilled putters. Arguably, this ineffective behavior can interfere with the visual-spatial cues previously attended. Encouraging and/or training the QE in this case can alleviate such inefficient gaze behavior, thereby rendering the performer more effective. This investigation was the first to assess th e mechanisms responsible for skill-based QE differences. As previously mentioned, the results from this st udy support the motor programming/movement preparation function of the QE duration over the emotion regulation function. Given that this is a seminal investigat ion, replication of thes e findings would lend to the empirical and theoretical support of the QE period for movement preparation and to determine whether these findings are gene ralizable to other self-paced tasks.

PAGE 174

174 The anxiety manipulation used in this invest igation successfully modulated both self-report and physiological indices of anxiet y and arousal. However, the lack of statistical support for the emotion regulation function of the QE period may have been underestimated based on the current results. For example, data trends suggest that as anxiety increased so too did QE, yet the reported differences between high and low anxi ety conditions were not significant. Perhaps future studies should consider including multiple manipulations varying in relative anxiety. For example, the current manipulation included a mone tary reward for perfor ming well, coupled with the anxiety provoking experience of performing in front of a camera. In either case, there was very little at risk for the part icipants. Considering that most people are more adversely affected by the thought of losing money and welcome th e chance to win money, future research may consider implementing a monetary penalty for unsuccessful or poor performance. The expert-near-expert paradigm that was us ed here has proven successful, providing insightful information into the relative skill based differences. However, as previously alluded to, a practical implication of these findings is to tr ain the QE in a manner that would facilitate information processing and sensorimotor efficienc y. Although it is reasonable to expect that the QE and BP differences found here would corr espond with the very best players (i.e., professional), without explicitly testing them, the magnitude and direction of the QE period and relative cortical changes associated with them remain uncertain. Until such observations can be made, the development of a QE training prot ocol should be cauti ously undertaken. Summary The purpose of this investigation was to clarify the role of the QE period in the preparatory process of a self-paced motor skill. The concurrent exploration of the BP and QE period under varying levels of a nxiety was designed to assess the underlying mechanisms that link QE duration and performance. Twenty gol fers were classified by their USGA handicap

PAGE 175

175 rating as either a high handicap (near-expert) or low handicap (expert), to permit skill-based inferences. Participants completed 45 putts in both the low and high a nxiety conditions during which cognitive anxiety, somatic anxiety, hear t rate, QE duration, BP activity, and putting performance were recorded. As expected, the LH groups putting performan ce was superior to the HH groups, with the LH group successfully completing more putts while missing closer in proximity to the target across anxiety conditions. Moreover, the QE pe riod of the LH group was longer than the HH groups, yet this relationship was not maintain ed across anxiety conditions. Furthermore, QE duration did not differ for successful (i.e., hits) and unsuccessful (i.e., misses) putts both across and within participants. As e xpected however, the LH group disp layed greater BP negativity in the right-central, right-lateral, and the right-parietal regions re lative to the HH group and this increase was significantly related to an increase in QE duration. Taken together, these results suggest that expe rt players operate with a greater level of automaticity and less conscious processing of th e requisite movements than their near-expert counterparts. Moreover and paramount to this in vestigation, the data le nd empirical support to the motor programming implications of the QE duration over the arousal regulation function as indicated by the relationships between QE and radial error a nd QE and BP negativity around the primary motor cortex, as well as the lack of a demonstrable relations hip between anxiety and QE.

PAGE 176

176 LIST OF REFERENCES Aalto, H., Pyyk, I., Ilmarinen, R., Khknen, E., & Starck, J. (1990). Postural stability in shooters. Journal of Oto-Rhino-Larynology and its Related Specialties, 52, 232-238. Abernethy, B. (1988). The effects of age and ex pertise upon perceptual skill development in a racquet sport. Research Quarterly for Exercise and Sport, 59, 210-221. Abernethy, B. (1989). Expert-novic e differences in perception: Ho w expert does the expert have to be? Canadian Journal of Sports Sciences, 14 (1), 27-30. Abernethy, B. (1990a). Anticipation in squash: Di fferences in advance cu e utilization between expert and novice players. Journal of Sports Sciences, 8 17-34. Abernethy, B. (1990b). Expertise, visual se arch, and information pick-up in squash. Perception, 19 63-77. Abernethy, B. (1991). Visual search strategi es and decision-making in sport. International Journal of Sport Psychology, 22, 189-210. Abernethy, B. (1996). Training the vi sual-perceptual skills of athletes: Insights from the study of motor expertise. The American Journal of Sports Medicine, 24 (6), S-89-S-92. Abernethy, B., Burgess-Limerick, R., & Parks, S. (1994). Contrasting approaches to the study of motor expertise. Quest, 46, 186-198. Abernethy, B., Neal, R.J., & Koning, P. (1994). Visual-perceptual and cognitive differences between expert, intermediate and novice snooker players. Applied Cognitive Psychology, 18 185-211. Abernethy, B., & Russell, D.G. (1987a). Expert -novice differences in an applied selective attention task. Journal of Sport Psychology, 9 326-345. Abernethy, B., & Russell, D.G. (1987b). The relati onship between expertise and visual search strategy in a racquet sport. Human Movement Science, 6 283-319. Abernethy, B., Thomas, K.T., & Thomas, J.T. ( 1993). Strategies for improving understanding of motor expertise [or mistakes we have made and things we have learned!!]. In J.L. Starkes & F. Allard (Eds.), Cognitive issues in motor expertise (pp.317-356) Amsterdam: NorthHolland. Adolphe R., Vickers, J. N., & LaPlante, G. (199 8). The effects of traini ng visual attention on gaze behaviour and accuracy: A pilot study. International Journal of Sports Vision 4 (1), 28-33. Allard, F., & Burnett, N. (1985). Skill in sport. Canadian Journal of Psychology, 39 (2), 294-312.

PAGE 177

177 Allard, F., Graham, S., & Paarsa lu, M.E. (1980). Skill in sport. Canadian Journal of Psychology, 39, 294-312. Anderson, J.R. (1982). Acquisition of a cognitive skill. Psychological Review, 89, 369-406. Andreassi, J.L. (1980). Psychophysiology: Human behavior and physiological response. New York: Oxford University Press. Baddeley, A.D. (1986). Working memory Oxford, UK. Clarendon Press. Bard, C. & Fleury, M. (1976). Analysis of visual search activity during s port problem situations. Journal of Human Movement Studies, 3, 214-227. Bard, C. & Fleury, M. (1987). Considering eye move ment as a predictor of attainment. In I.M. Cockerill & W.W. MacGillivary (Eds.), Vision and sport (pp. 28-41). Cheltenham: Stanley Thornes. Becker, W., Iwase, R., Jurgens, R., & Kor huber, H.H. (1976). Bereitschaftspotential preceding voluntary slow and rapid hand move ments. In W.C. McCallum & J.R. Knott (Eds.), The responsive brain (pp. 99-102) Bristol: John Wright & Sons. Boschert, J., Hink, R.F., & Deecke, L. (1983). Fi nger movement versus toe movement-related potentials: Further evidence for supplementary motor area (SMA) participation prior to voluntary action. Experimental Brain Research, 52 (1), 73-80. Boutcher, S.H., & Zinsser, N.W. (1990). Cardia c deceleration of elite and beginning golfers during putting. Journal of Sport and Exercise Psychology, 12, 37-47. Brunia, C.H.M. (1993). Waiting in readiness: Gaiting in attention and motor preparation. Psychophysiology, 30, 327-229. Brunia, C.H.M., & Van Boxtel, G.J.M. (2000) Motor preparation. In J.T. Cacioppo, L.G. Tassinary, & G.G. Berntson (Eds.), Handbook of psychophysiology (2nd ed., pp. 507-532). Cambridge, UK: Cambridge University Press. Buckholz, E., Prapavessis, H., & Fairs, J. (1988) Advance cues and their use in predicting tennis passing shots. Canadian Journal of Sport Sciences,13, 20-30. Byrnes, A. (2002). Visual search and equestrian riders. Unpublished manuscript. Cauraugh, J., Singer, R.N., & Chen, D. (1993). Vi sual scanning and anticipation of expert and beginner tennis players. In S Serpa (Ed.), Proceedings : VIII Wo rld Congress of Sport Psychology (pp.336-340). Lisbon: International Soci ety of Sport Psychology Chairenza, G.A., Vasille, G., & Villa, M. ( 1990). Goal or near miss! Movement potential differences between adults and ch ildren in skille d performance. International Journal of Psychophysiology 10 (2), 105-115.

PAGE 178

178 Chase, W.G., & Simon, H.A. (1973). Perception in chess. Cognitive Psychology, 4 55-81. Chi, M.T.H., Feltovich, P.J., & Glaser, R. ( 1981). Categorization and representation of physics problems by experts and novices. Cognitive Science, 5, 121-125. Craig, C.M., Delay, D., Grealy, M.A., and Lee, D. N. (2000). Guiding the swing in golf putting. Nature, 405, 295-296. Crews, D.L. (1989). The influence of attentive states on gol f putting as indicated by cardiac and electrocortical activity. Unpublished Doctoral Dissertat ion, Arizona State University, Tempe, AZ. Crews, D.J., & Landers, D.M. (1993). Electroencepha lographic measures of attentional patterns prior to the golf putt. Medicine and Science in Sports and Exercise, 25, 116-126. Cui, Q.R., & Deecke, L. (1999). High-resolution DC -EEG analysis of the Bereitschaftspotential and post movement onset potentials accomp anying uni-or bilatera l voluntary finger movements. Brain Topography, 11, 233-248. Davids, K.W., Rex Pe Palmer, D., & Savelsbergh, G.J.P. (1989). Skill level, peripheral vision and tennis volleying performance. Journal of Human Movement Studies, 16, 191-202. Davidson, R.J., Jackson, D.C., & Larson, C.L. (2000). Human electroencephalography. In J.T. Cacioppo, L.G. Tassinary, & G.G. Berntson (Eds.), Handbook of psychophysiology (2nd ed., pp. 27-52). New York: Cambridge University Press. Deecke, L. (1973). Potential changes associat ed with motor action, reflex responses and readiness. In W.C. McCallum & J.R. Knott, The responsive brain: The proceedings of the third international congress on event-related slow potentials of the brain (pp.91-93) Dorchester, UK: John Wright & Sons Ltd. Deecke, L. (1987). Bereitschaftspotential as an indicator of movement preparation in supplementary motor area and motor cortex. In Motor areas of the cerebral cortex (Ciba Foundation Symposium, no.132, pp.231-250) Chichester, UK: Wiley. Deecke, L., & Kornhuber, H.H. (1978). An electr ical sign of participa tion of the mesial supplementary motor cortex in human voluntary finger movement. Brain Research, 159, 473-476. Deecke, L., & Kornhuber, H.H. (2003). Human freedom, reasoned will and the brain. In M. Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential : Movement related cortical potentials (pp.283-321). New York, NY: Kluwer Academic/Plenum Publishers. Deecke, L., Kornhuber, H.H., Lang, W., Lang, M., & Schreiber, H. (1985). Timing functions of the frontal cortex in sequential motor-and learning tasks. Human Neurobiology, 4, 143-154.

PAGE 179

179 Deecke, L., Scheid, P., & Kornhuber, H.H. ( 1969). Distribution of readiness potential, premotion positivity, and motor potential of th e human cerebral cortex preceding voluntary finger movement. Experimental Brain Research, 7, 158-168. Deeny, S.P., Hillman, C.H., Janelle, C.M., & Hatfield, B.D. (2003). Cortico-cortical communication and superior performance in skilled marksmen: An EEG coherence analysis. deGroot (1978). Though and choice in chess. The Hague: Mouton. Duchowski, A.T. (2002). Eye tracking methodology: Theory and practice, London: Springer. Easterbrook, E.A. (1959). The effect of emoti on on cue utilization a nd the organization of behavior. Psychological Review, 66 (3), 183-201. Ennis, C.D. (1994). Knowledge and belie fs underlying curric ular expertise. Quest, 46, 164-175. Ericsson, K. A., & Delaney, P. F. (1999). Long-te rm working memory as an alternative to capacity models of working memory in everyday skilled performance. In A. Miyake & P. Shah (Eds.), Models of working memory (pp. 257-297). Cambridge, MA: Cambridge University Press. Ericsson, K. A., & Kintsch, W. (1995). Long-term working memory. Psychological Review, 102 (2), 211-245. Ericsson, K.A., Krampe, R.T., & Tesch-Rmer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100, 363-406. Ericsson, K.A., & Lehmann, A.C. (1996). Expert and exceptional performance: Evidence of maximal adaptation to task constraints. Annual Review of Psychology, 47, 273-305. Eysenck, M.W. (1982). Attention and arousal, c ognition and performance New York: SpringerVerlag. Eysenck, M.W., & Calvo, M.G. (1992). Anxiety a nd performance: The processing efficiency theory. Cognition and Emotion, 6 (6), 409-434. Fabiani, M., Gratton, G., & Coles, M.G.H. ( 2000). Event related brai n potentials: Methods, theory, and applications. In J.T. Cacioppo, L.G. Tassinary, & G.G. Berntson (Eds.), Handbook of Psychophysiology (2nd ed., pp.53-84). Ca mbridge, UK: Cambridge University Press. Fitts, P. M. & Posner, M. I. (1967). Human performance. Belmont, CA: Brooks/Cole. French, K.E., Spurgeon, J.H., & Nevett, M.E. (1 995). Expert-novice differences in cognitive and skill execution components of youth baseball performance. Research Quarterly for Exercise and Sport, 66, 194-201.

PAGE 180

180 French, K.E., & Thomas, J.R. (1987). The relation of knowledge development to childrens basketball performance. Journal of Sport Psychology, 9 15-32. Goulet, C., Bard, C., & Fleury, M. (1989). Expertis e differences in prepari ng to return a tennis serve: a visual information processing approach. Journal of Sport & Exercise Psychology 11 382-398. Hancock, G.H., Butler, M.S., & Fischman, M.G. (1995). On the problem of two-dimensional error scores: Measures and analyses of accuracy, bias, and consistency. Journal of Motor Behavior, 27 (3), 241-250. Huckabee, M.L., Deecke, L., Cannito, M.P., Goul d, H.J., & Mayr, W. (2003). Cortical control mechanisms in volitional swallowing: The Bereitschaftspotenial. Brain Topography, 16 (1), 3-16. Hardy, L., Mullen, R., & Jones, G. (1996). Know ledge and conscious control of motor actions under stress. British Journal of Psychology, 87, 621-636. Harle, S.H., & Vickers, J.N. ( 2001). Training quiet eye improves accuracy in the basketball free throw. The Sports Psychologist, 15, 289-305. Hatfield, B.D., & Hillman, C.H. (2001). Th e psychophysiology of sport: A mechanistic understanding of the psychology of superior performance. In R.N. Singer, H.A. Hausenblas, & C.M. Janelle (Eds.), Handbook of sport psychology (2nd ed., pp.362-386). New York, NY: John Wiley & Sons, Inc. Hatfield, B.D., Hillman, C.H., Apparies, R.J., Janelle, C.M., & Vickers, J. (1999). Expert performance states: Psychophysiologi cal perspectives on efficiency. Journal of Sport and Exercise Psychology, 21, s5. Hatfield, B.D., Landers, D.M., & Ray, W.J. ( 1984). Cognitive processes during self-paced motor performance: An electroencephalographic profile of skilled marksmen. Journal of Sport Psychology, 6, 42-59. Hatfield, B.D., Landers, D.M., & Ray, W.J. (1987) Cardiovascular-CNS interactions during a self-paced, intentional attentive state. Elite marksmanship performance. Psychophysiology, 24, 542-549. Hatfield, B.D., Landers, D.M., Ray, W.J., & Daniel s, F.S. (1982). An electroencephalographic study of elite rifle shooters. The American Marksman, 7, 6-8. Haufler, A.J., Spalding, T.W., Santa Maria, D.L., & Hatfield, B.D. (2000). Neuro-cognitive activity during a self-pac ed visuospatial task: Comparative EEG profiles in marksmen and novice shooters. Biological Psychology 53, 131-160. Helsen, W., & Pauwels, J.M. (1990). Analysis of vi sual search activity in solving tactical game problems. In D. Brogan (Ed.), Visual search (pp. 177-184) London: Taylor and Francis.

PAGE 181

181 Helsen, W., & Pauwels, J.M. (1992). A cognitive appr oach to visual search in sport. In D. Brogan (Ed.), Visual search II (pp. 177-184). London: Taylor & Francis. Helsen, W., & Pauwels, J.M. (1993). The relations hip between expertise and visual information processing in sport. In J. Starkes & F. Allard (Eds.), Cognitive issues in motor expertise (pp.109-134). Amsterdam: Elsevier. Helsen, W., & Starkes, J. (1999). A multidimensional approach to skilled perception and performance in sport. Applied Cognitive Psychology, 13, 1-27. Hillman, C.H., Apparies, R.J., Janelle, C.M., & Hatfield, B.D. (2000). An electrocortical comparison of executed and reject ed shots in skilled marksmen. Biological Psychology, 52, 71-83. Holyoak, K.J. (1991). Symbolic connectionism: Toward thirdgeneration theories of expertise. In K.A. Ericsson & J. Smith (Eds.), Toward a general theory of expertise: Prospects and limits (pp. 301-335) New York: Cambridge University Press. Houlston, D.R., & Lowes, R. (1993). Anticipator y cue-utilization pro cess amongst expert and non-expert wicketkeep ers in cricket. International Journal of Sport Psychology, 24, 59-73. Ikeda, A., Lders, H.O., Burgess, R.C., & Shibas aki, H. (1993). Moveme nt-related potentials associated with single and repetitive m ovements recorded from human supplementary motor area. Electroencephalography Clinical Neurophysiology, 89, 269-277. Isaac, L., & Finch, A. (1983). Antic ipatory timing of beginning and intermediate tennis players. Perceptual and Motor Skills 57 451-454. Jahanshahi, M., & Hallett, M. (2003). The Bereit schaftspotential: What does it measure and where does it come from. In M. Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential : Movement related cortical potentials (pp.1-17). New York, NY: Kluwer Academic/Plenum Publishers. Janelle, C.M. (1997). Changes in visual search patterns as an indication of attentional narrowing and distraction during a simulated hi gh-speed driving task under increasing levels of anxiety Unpublished Doctoral Dissertation, Univ ersity of Florida, Gainesville, Florida. Janelle, C.M. (2002). Anxiety, arousal and visual attention: A mechanistic account of performance variability. Journal of Sports Sciences, 20, 237-251. Janelle, C.M., & Hillman, C.H. (2003). Expert performance in sport: Current perspective and critical issues. In J.L. Starkes & K.A. Erikson (Eds). Expert Performance in sports: Advances in research on sport expertise (pp. 19-48) Champaign, IL: Human Kinetics. Janelle, C.M., Hillman, C.H., Apparies, R.J., Mu rray, N.P., Meili, L., Fallon, E.A., & Hatfield, B.D. (2000). Expertise differences in corti cal activation and gaze behavior during rifle shooting. Journal of Sport and Exercise Psychology, 22 167-182.

PAGE 182

182 Janelle, C.M., Hillman, C.H., & Hatfield, B.D. (2000). Concurrent measurement of electroencephalographic and oc ular indices of attention during rifle shooting: An exploratory case study. International Journal of Sports Vision, 6 (1), 21-29. Janelle, C.M., Singer, R.N., & Williams, A.M. (1999). External distraction and attentional narrowing: Visual search evidence. Journal of Sport & Exercise Psychology, 21 (1) 70-91. Jasper, H.H. (1958). Report of the committee on methods of clinical examination in electroencephalography. Journal of Electroencephalography and Clinical Neurophysiology, 10, 370-375. Jones, C., & Miles, J. (1978). Use of advance cues in predicting the flight of a lawn tennis ball. Journal of Human Movement Studies, 4 231-235. Jones, G., & Swain, A.B.J. (1992). Intensity a nd direction dimensions of competitive state anxiety and relationships with competitiveness. Perceptual and Motor Skills, 74 467-472. Kerick, S.E., Douglass, L.W., & Hatfield, B.D. ( 2004). Cerebral cortical ad aptations associated with visuomotor practice. Medicine & Science in Sports & Exercise, 36, 118-129. Konttinen, N., Landers, D.M., & Lyytinen, H. (2000) Aiming routines and their electrocortical concomitants among competitive rifle shooters. Scandinavian Journal of Medicine & Science in Sports, 10 (3), 169-177. Konttinen, N., & Lyytinen, H. (1992). Physiology of preparation: Brain slow waves, heart rate, and respiration preceding triggering in rifle shooting. International Journal of Sport Psychology, 23, 110-127. Konttinen, N., & Lyytinen, H. (1993). Brain sl ow waves preceding time-locked visuo-motor performance. Journal of Sport Sciences, 11, 257-266. Konttinen, N., Lyytinen, H., & Era, P. (1999). Br ain slow potentials and postural sway behavior during sharpshooting performance. Journal of Motor Behavior, 31, 11-20. Konttinen, N., & Lyytinen, H., & Konttinen, R. (1995). Brain slow pot entials reflecting successful shooting performance. Research Quarterly for Exercise and Sport, 66, 64-72. Kornhuber, H.H., & Deecke, L. (1965). Hi rnpotentialnderungen bei Willkrbewegungen und passiven Bewegungen des Menschen: Bereitscha ftspotential und reafferente Potentiale. Pflgers Archiv, 284, 1-17. Krane, V. (1994). The mental readiness form as a measure of competitive state anxiety. The Sport Psychologist 8, 189-202. Lacey, B.C., & Lacey, J.L. (1978). Two-way comm unication between the heart and the brain: Significance of time within the cardiac cycle. American Psychologist, 33, 99-113.

PAGE 183

183 Landers, D.M., Han, M., Salazar, W., Petruzello, S.J., Kubitz, K.A., & Gannon, T.L. (1994). Effects of learning on electroencelephalograp hic and electrocardiographic patterns in novice archers. International Journal of Sport Psychology, 25 313-330. Lawton, G.W., Hung, T.M., Saarela, P., & Hatfie ld, B.D. (1998). Electroencephalography and mental states associated with elite performance. Journal of Sport and Exercise Psychology 20 35-53. Libet, B. (2003). Can conscious experience affect brain activity? Journal of Consciousness Studies, 10 (12), 24-28. Liebert, R.M., & Morris, L.W. (1967). Cognitive a nd emotional components of test anxiety: A distinction and some initial data. Psychological Reports, 20, 975-978. Lopes da Silva, F. (1991). Neural mechanisms underlying brain waves: From neural membranes to networks. Electroencephalography and Clinical Neurophysiology, 79, 81-93. Loveless, N.E., & Sanford, A.J. (1974). Slow potential correlates of preparatory set. Biological Psychology, 1, 303-314. Loze, G.M., Collins, D., & Holmes, P.S. (2001). Pre-shot EEG alpha-power reactivity during expert air-pistol shooting: A comp arison of best and worst shots. Journal of Sports Sciences, 19, 727-733. Lyoka, P.A., & Bressan, E.S. (1998). Elite coach es perceptions of the characteristics of decision-making that discriminate between expert and novice basketball players. Journal For Research in Sport, Physical Education and Recreation, 25 (1), 59-69. Magill, R. (1998). Motor learning: Con cepts and applications (5th ed.). Boston, MA; McGraw Hill. Mann, D.T.Y., William, A.M., Ward, P., & Janelle, C.M. (2006). Perceptual-cognitive expertise in sport: A meta-analysis. Manuscript submitted for publication. Marteniuk, R.G. (1976). Information processing in motor skills. New York: Holt, Rinehart & Winston. Martens, R., Burton, D., Vealey, R.S., Bump, L.A., & Smith, D.E. (1990). Development and validation of the Competitive State Anxiety In ventory-2. In R. Martens, R.S. Vealey, & D. Burton (Eds.), Competitive anxiety in sport (pp.117-190). Champaign, IL: Human Kinetics. Masters, R.S.W. (1992). Knowledge, knerves and know -how: The role of exp licit versus implicit knowledge in the breakdown of a co mplex motor skull under pressure. British Journal of Psychology, 83, 343-358.

PAGE 184

184 McAdam, D.W., & Rubin, E.H. (1971). Readiness potential and vertex positive wave and contingent negative variation and accuracy of perception. Electroencephalography & Clinical Neurophysiology, 30, 511-517. McAdam, D.W., & Seales, D.M. (1969). Bereit schaftspotential enhancement with increased level of motivation. Electroencephalography & Clin ical Neurophysiology, 27, 73-75. McPherson, S.L. (1993). The influence of player experience on problem so lving during batting preparation in baseball. Journal of Sport & Exercise Psychology, 15, 304-325. McPherson, S.L. (1994). The development of spor t expertise: Mapping th e tactical domain. Quest, 46, 223-240. McPherson, S.L. (1999). Tactical differences in problem representati ons and solutions in collegiate varsity and beginner women tennis players. Research Quarterly for Exercise and Sport, 70, 369-284. McPherson, S.L. (2000). Expert-novice differences in planning strategi es during collegiate singles tennis competition. Journal of Sport & Exercise Psychology, 22, 39-62. Milton, J., Solodkin, A., Hlusk, P., & Small, S. (2 007). The mind of expert motor performance is cool and focused. NeuroImage, 35, 804-813. Molander, B., & Backman, L. (1989). Age differences in heart rate pattern s during concentration in a precision sport: Implications for a ttentional functioning. Journal of Gerontology and Psychological Science, 44, 80-87. Murray, N.M., & Janelle, C.M. (2003). Anxiety a nd performance: A visual search examination of the processing efficiency theory. Journal of Sport and Exercise Psychology, 25, 171187. Neisser, U. (1967). Cognitive psychology New York: Appleton, Century, Crofts. Niinimaa, V., & McAvoy, T. (1983). Influence of exercise on body sway in the standing rifle shooting position. Canadian Journal of Applied Sport Sciences, 8, 30-33. Nunez, P.L. (1995). Neuromodulation of neur ocortical dynamics. In P.L. Nunez (Ed.), Neurocortical dynamics and human EEG rhythms (pp. 591-627). New York: Oxford University Press. Orgogozo, J.M., & Larsen, B. (1979). Activation of the supplementary motor area during voluntary movement suggests it works as a supramotor area. Science, 206, 847-850. Papakostopoulos, D. (1978). Electr ical activity of the brain associated with skilled performance. In D.A. Otto (ed.), Multi-disciplinary perspectives in event-related brain potentials research (pp.134-137). Washington, DC: U.S. Environmen tal Protection Agency, Office of Research and Development.

PAGE 185

185 Parent, A. (1996). Carpenters human neuroanatomy (9th ed). Baltimore, MD: Williams and Wilkins. Paul, G., & Glencross, D. (1997). Expert perception and decision making in baseball. International Journal of Sport Psychology, 28, 35-56. Pelz, D. (2000). Putting bible: The complete guide to mastering the green New York: Doubleday. Pfurtscheller, G. (1992). Event-related synchroniza tion (ERS): An electro physiological correlate of cortical areas at rest. Journal of Electroencephalogr y and Clinical Neurophysiology 83, 62. Pfurtschelle, G. & Neuper, C. ( 2003). Movement and ERD/ERS. In M. Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential : Movement related cortical potentials (pp.191-205). New York, NY: Kluwer Academic/Plenum Publishers. Pivik, R. T., Broughton, R.J., Coppola, R., Davids on, R. J., Fox, N., & Nuwer, M. R. (1993). Guidelines for the recording and quantitative analysis of electroencephalographic activity in research contexts. Psychophysiology 30 547. Posner, M.I., Nissen, M.J., & Klein, R.M. (1976). Vi sual dominance: An information processing account of its origins and significance. Psychological Review, 83, 157-171. Posner, M.I., & Raichle, M. (1991). Images of mind. New York: Scientific American Books. Radlo, S.J., Janelle, C.M., Barba, D.A., Frehlich, S.G. (2001). Perceptual decision making for baseball pitch recognition: Using P300 la tency and amplitude to index attentional processing. Research Quarterly for Exercise and Sport, 72, 22-31. Reagan, D. (1989). Human brain electrophysiology: Evoked potentials and evoked magnetic fields in science and medicine. New York: Elsevier. Reeve, T.G., Fischman, M.G., Christina, R.W ., & Cauraugh, J.H. (1994). Using one-dimensional task error measures to assess performa nce on two-dimensional tasks: Comments on attentional control, distract ors, and motor performance. Human Performance, 7 (4), 315319. Rektor, I. (2003). Intracerebral recordings of the bereitschaftspotential. In M. Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential : Movement related cortical potentials (pp.61-77). New York, NY: Kluwer Acad emic/Plenum Publishers. Renshaw, I., & Fairweather, M.M. (2000). Cric ket bowling and the discrimination ability of professional and amateur batters. Journal of Sports Sciences, 18 951-957.

PAGE 186

186 Richman, H.B., Gobet, F., Staszewski, J.J., & Simon, H.A. (1996). Perceptual and memory processes in the acquisition of expert performance. In K.A. Ericsson (Ed.), The road to excellence: The acquisition of expert perform ance in the arts and sciences, sports, and games (pp.167-187). Mahwah, NJ: Lawrence Erlbaum Associates. Ripoll, H., Kerlirzin, Y., Stein, J.F. & Reine, B. (1995). Analysis of information processing, decision making, and visual search strategies in complex problem solving sport situations. Human Movement Science 14, 325-349. Roland, P.E. (1984). Organization of motor control by the human brain. Human Neurobiology, 2, 205-216. Roland, P.E., Larsen, B., Lassen, N.A., & Ski nhoj, E. (1980). Supplementary motor area and other cortical areas in organiza tion of voluntary movements in man. Journal of Neurophysiology, 43, 118-136. Salazar, W., Landers, D.M., Petruzello, S.J., & Han, M. (1990). Hemisphe ric asymmetry, cardiac response, and performan ce in elite archers. Research Quarterly for Exercise and Sport, 61, 351-359. Savelsbergh, G.J.P., Williams, A.M., Van Der Ka mp, J., & Ward, P. (2002). Visual search, anticipation, and expertise in soccer goalkeepers. Journal of Sport Sciences, 20, 279-287. Schmid, R., & Zambarbieri, D. (1991). Strategies of eye-head coordination. In R. Schmid & D Zambarbieri (Eds.), Oculomotor control and cognitive processes (pp.229-248). Amersterdam: North-Holland. Shank, M.D., & Haywood, K.M. (1987). Eye move ments while viewing a baseball pitch. Perceptual and Motor Skills, 64, 1191-1197. Shibasaki, H., Barrett, G., Halliday, E., & Hallid ay, A.M. (1980). Components of the movementrelated cortical and their scalp topography. Electroencephalography and Clinical Neurophysiology, 49, 213-226. Shim, J., & Carlton, L.G. (1999). Anticipation of movement outcome through perception of movement kinematics. Journal of Sport and Exercise Psychology, 21, S98. Siedentop, D., & Eldar, E. (1989). Expert ise, experience, and effectiveness. Journal of Teaching in Physical Education, 8 (3), 254-260. Singer, R.N. (2000). Performance and human f actors: Considerations about cognition and attention for self-paced and externally paced events. Ergonomics, 43 (10), 1661-1680. Singer, R., Cauraugh, J., Chen, D., Sternberg, G. M., & Frehlich, S. G. (1996). Visual search, anticipation, and reactive comparisons between highly-skilled and beginning tennis players. Journal of Applied Sport Psychology, 8 (1), 9-26.

PAGE 187

187 Slobounov, S., Tutwiler, R., Rearick, M., & Challis, J.H. (1999). EEG correlates of finger movements with different inertial load condi tions as revealed by averaging techniques. Clinical Neurophysiology, 110, 1764-1773. Smith, M.E., McEvoy, L.K., & Gevins, A. ( 1999). Neurophysiological indices of strategy development and skill acquisition. Cognitive Brain Research, 7 389-404. Spielberger, C.D., Gorsuch, R.L., Lushene, R., Vagg, P.R., & Jacobs, G.A. (1983). State-Trait Anxiety Inventory (Form Y). Redwood City, CA: Mind Garden. Starkes, J.L., (1987). Skill in field hockey. The nature of the cognitive advantage. Journal of Sport Psychology, 9, 146-160. Starkes, J.L. (1993). Motor experts: Opening t houghts. In J.L. Starkes & F. Allard (Eds.), Cognitive issues in motor expertise (pp.3-16). Amsterdam: Elsevier. Starkes, J.L., & Allard, F. (Eds.). (1993). Cognitive issues in motor expertise. Amsterdam: Elsevier. Starkes, J.L., Edwards, P., Dissanayake, P., & Dunn.T. (1995). A new t echnology and field test of advance cue usage in volleyball. Research Quarterly for Exercise and Sport, 65, 1-6. Taylor. M. (1978). Bereitschaftspotential dur ing the acquisition of a skilled motor task. Electroencephalography and Clinical Neurophysiology, 45, 568-576. Tenenbaum, G., Levy-Kolker, N., Sade, S., Lieb ermann, D.G., & Lidor, R. (1996). Anticipation and confidence of decisions re lated to skilled performance. International Journal of Sport Psychology, 27, 293-307. Thomas, K.T. (1994). The development of s port expertise: From Leeds to MVP legend. Quest, 46, 199210. Thomas, K.T. & Lee, C. (1991). A description of skill, knowledge, selected fitness measures and golf play in women over 50 years of age. Paper presented at the annual meeting of the North American Society for Psychology of Sport and Physical Activity, Pittsburgh, PA. Thomas, J.R., Thomas, K.T., & Gallagher, J.D. (1993). Developmental considerations in skill acquisitions. In R.N. Singer, M. Murphey, & L.K. Tennant (Eds.), Handbook of research on sport psychology (pp.73-105). New York: Macmillan. Toni, I., & Passingham, R.E. (2003). Movement preparation: Neuroima ging studies. In M. Jahanshahi & M. Hallett (Eds.), The Bereitschaftspotential : Movement related cortical potentials (pp.269-281). New York, NY: Kluwer Academic/Plenum Publishers. Tremayne, P., & Barry, R.J. (1990). Repr ession of anxiety and its effects upon psychophysiological responses to relevant and ir relevant stimuli in competitive gymnasts. Journal of Exercise and Sport Psychology, 12, 333-352.

PAGE 188

188 Tremayne, P., & Barry, R.J. (2001). Elite pistol shooters: Physiological patterning of best vs. worst shots. International Journal of Psychophysiology, 41, 19-29. Van Wynsberghe D.M., Noback, C., & Carola, R. (1995). Human anatomy and physiology (3rd ed.). New York: McGraw-Hill, Inc. Vickers, J.N. (1988). Knowledge stru ctures of expert-novice gymnasts. Human Movement Science, 7, 47-72. Vickers, J.N. (1992). Gaze control in putting. Perception, 21, 117-132. Vickers, J.N. (1996a). Visual cont rol while aiming at a far target. Journal of Experimental Psychology: Human Perception and Performance, 22, 342-354. Vickers, J.N. (1996b). Control of visual at tention during the basketball free throw. The American Journal of Sports Medicine 24 (6), S93-S96. Vickers, J.N. (2004). The quiet-eye. Its the difference between a good putter and a poor one. Heres proof. Golf Digest, 1, 96-101 Vickers, J.N., & Adolphe, R.M. (1997). Gaze be havior during a ball tracking and aiming skill. International Journal of Sports Vision, 4 (1), 18-27. Vickers, J.N., Williams, A.M., Rodrigues, S., Hi llis, F., & Coyne, G. (1999). Eye movements of elite biathlon shooters during rested and fatigues states. Journal of Sport and Exercise Psychology, 21, s116. Ward, P., & Williams, A. M. (2003). Perceptual a nd cognitive skill development in soccer: The multidimensional nature of expert performance. Journal of Sport & Exercise Psychology, 25 (1), 93-111. Ward, P., Williams, A.M., & Bennett, S.J., ( 2002). Visual search and biological motion perception in tennis. Research Quarterly fo r Exercise and Sport, 73 (1), 107-112. Warren, C., & Karrer, R. (1984). Movement-relate d potentials in childr en: A replication of waveforms, and their relationships to ag e, performance and cognitive development. Annals of the New York Ac ademy of Sciences, 425, 489-495. Williams, A.M. (2000). Perceptual skill in soccer: Implications for talent identification and development. Journal of Sports Sciences, 18 (9), 737-750. Williams, A.M., & Davids, K. (1995). Declar ative knowledge in sport: A by-product of experience or a character istic of expertise? Journal of Sport and Exercise Psychology, 17, 259-275. Williams, A.M., & Davids, K. (1998). Visual search strategy, selective atten tive, and expertise in soccer. Research Quarterly for Exercise and Sport, 69, 111-128.

PAGE 189

189 Williams, A.M., Davids, K., Burwitz, L. & Williams J.G. (1992). Perception and action in sport. Journal of Human Movement Studies, 22, 147-204. Williams, A.M., Davids, K., Burwitz, L. & Willia ms, J.G. (1993). Visual search and sports performance. The Australian Journal of Scie nce and Medicine in Sport, 25 (2), 147-204. Williams, A.M., Davids, K., Burwitz, L., & Williams J.G. (1994). Visual search strategies in experienced and inexperienced soccer players. Research Quarterly for Exercise and Sport, 6 (2), 127-135. Williams, A.M., Davids, K., & Williams, J.G. (1999). Visual perception and action in sport, London: E & FN Spon. Williams, A.M., & Elliot, D. (1999). Anxiety, expe rtise and visual search strategy in karate. Journal of Sport & Exercise Psychology 21, 361-374 Williams, A.M, Singer, R., & Frehlich, S. (2002) Quiet eye duration, expertise, and task complexity in near and far aiming tasks. Journal of Motor Behavior, 34 197207. Williams, M.A., Vickers, J., & Rodrigues, S. (2002). The effects of anxiety on visual search, movement kinematics, and performance in ta ble tennis: A test of Eysenck and Calvo's processing efficiency theory. Journal of Sport & Exercise Psychology, 24 (4), 438-455. Woo Sohn, Y., & Doane, S.M. (2003). Roles of working memory capacity and long-term working memory skill in complex task performance. Memory & Cognition, 31 (3), 458-466. Wright, D.L., Pleasants, F., & Go mez-Meza, M. (1990). Use of adva nced visual cue sources in volleyball. Journal of Sport and Exercise Psychology, 12, 406-414. Wulf, G., & Shea, C.H. (2002). Principles deri ved from the study of simple skills do not generalize to complex skill learning. Psychonomic Bulletin & Review, 9 (2), 185-211.

PAGE 190

190 BIOGRAPHICAL SKETCH Born in 1973, in York, Ontario, Canada, Dere k Thomas Yonge Mann, brother to Darlene, and son of Diane and Daniel Ma nn, equally pursued both athletic and academic endeavors. After graduating from Richview Collegi ate Institute in Etobicoke, Onta rio, Derek earned the Bachelor of Arts degree in psychology from York Univer sity while playing thr ee years for the York Yeomen ice hockey team. After graduating from York in 1998, Derek left Canada to pursue a masters degree in sport psychology at San Dieg o State University under the guidance of Dr. Brent Rushall. Upon completion of his masters degree, Derek was accepted into the doctoral program at the University of Florida in sport and exercise psychology under the direction of Dr. Christopher M. Janelle. Specializing in the percep tual-cognitive advantage of experts in sport, Derek completed his dissertation and was awarded his Ph.D. degree in December 2007.