Changes in EEG laterality index effects of social inhibition on putting in novice golfers.

Shelley-Tremblay, John F. ; Shugrue, John D. ; Kline, John P. 等

Throughout the last twenty-five years studies have examined changes in electroencephalogram (EEG) laterality as elite athletes have prepared to execute a motor act, such as shooting a bow and arrow or rifle, or performing a golf putt (Bird, 1987; Hatfield, Landers, & Ray, 1984; Hillman, Apparies, Janelle, Hatfield, 2000; see Hatfield, Haufler, Hung, & Spalding, 2004 for a review). However, relatively few studies have examined the EEG correlates of performance for novice athletes during a similar preparatory period. This is largely due to the emphasis placed on research designed to investigate "peak" or "ideal" performance, with its direct application to providing a competitive edge in sports (Williams & Krane, 1998).

In addition, while social facilitation/inhibition is one of the oldest and most researched areas in sport psychology, there are virtually no studies that examine how the presence of an audience affects EEG lateralization patterns within the sports literature (see Saarela, 2000; and see Davidson, Marshall, Tomarken, & Henriques, 2000 for EEG and public speaking). This is particularly surprising given the aforementioned emphasis on researching issues relevant to competitive sport, which has as a hallmark the presence of an audience. The current study seeks to expand on the work of Crews and Landers (1993) by exploring the shifts in EEG laterality that occur prior to a golf putt made by novice golfers who were assigned to putt alone and also in front of an audience.

Electroencephalogram (EEG)

The electroencephalogram represents a record of fluctuations in the electrical activity of the brain recorded from the surface of the scalp. The EEG records potential changes, which are generated by the summed ionic currents of many thousands of cortical neurons. The EEG represents the excitatory and/or inhibitory post-synaptic potentials recorded primarily from the apical dendrites of pyramidal neurons in neocortex. The frequencies of the potentials recorded from the surface of the scalp of a normal human vary from 1 to 50 Hz, and the amplitudes typically range from 20 to 100uV (Neidermeyer & Lopez da Silva, 1999). Four dominant frequency ranges are typically observed: alpha, beta, delta and theta. Theta and delta activity predominate during sleep, and as such, are not reviewed further here. Since the present study focused on participants during a waking state, only alpha and beta were analyzed in the present study and will be reviewed further.

Beta waves are normally seen more diffusely during intense mental activity, and have frequencies ranging from 13 to 30 Hz. Beta waves have the smallest amplitudes of recorded EEG activity (Neidermeyer & Lopez da Silva, 1999). For the present study, beta was broken down into beta 1 (13 to 21 Hz) and beta 2 (21-30), in order to replicate previous research. Beta 2 frequencies seem to be particularly active in schizophrenics and highly anxious performers (Ramos, Cerdan, & Guevara, 2001).

Alpha waves, which range in frequency from 8 to 13 Hz, are sometimes called Berger rhythm, after Hans Berger who first identified them. Alpha waves are generally associated with a state of relaxed wakefulness, especially visible in occipital regions when the eyes are closed. An increase in alpha amplitude in a task has frequently been linked to cortical deactivation (Kimura et al., 2001), especially in the sport psychology literature (Hatfield et al., 1984; Rebert, Low, & Larson, 1984; Crews & Landers, 1993).

Studies of EEG Asymmetry Prior to the Execution of a Motor Act

Several investigations have explored the relationship between EEG asymmetry prior to performing a motor act and subsequent athletic performance. Hatfield and Hillman (2000, p. 362) provide an excellent review of this literature, which begins with the work of Pullum (1977) indicating that better shooting accuracy was associated with an "enhanced" alpha state. Hatfield went on to complete a series of studies to more thoroughly investigate changes in alpha activity as elite marksmen prepared to fire their rifles. Hatfield, Landers, Ray, and Daniels (1982) reported that the left hemisphere showed relatively greater alpha activity than the right hemisphere as the time to pull the trigger approached. This effect was replicated by Hatfield, Landers, and Ray (1984), who extended the prior findings to show that the alpha laterality effect was comprised of right temporal (T4) stability with relative left temporal (T3) alpha increase. They also reported a global "quieting" of the cortex as the time to pull the trigger neared as demonstrated by increased alpha power at temporal and occipital sites.

The interpretation provided by Hatfeld et al. (1984) for these findings was that left hemisphere processes became less important as the trigger pull approached, while right hemisphere processes became relatively more important. The exact processes have yet to be specifed, but Hat field et al. (1984) did include additional tasks designed to selectively engage the marksmens' verbal/analytical abilities or their visual/spatial abilities. They found that while the verbal/analytical tasks led to a relative increase in right hemisphere to left hemisphere alpha (LH activation), the visual/spatial task led to no significant changes in the alpha ratio. The authors interpreted their findings as consistent with the idea that right hemisphere (presumably visual/spatial) processes are critical in successful marksmanship. This general theory has been widely influential in the interpretation of the results of subsequent EEG laterality effects in the sport literature, including the study by Crews et al. (1993) upon which the current study is based.

One of the important differences between the present investigation and that of Crews et al. (1993) is that novice golfers were used instead of experts. Previous investigations that have examined the relationship between an athlete's skill level and the EEG correlates of performance allow us to make predictions about the findings in the present study. One of the general findings from EEG research in this area has been that an increase in alpha activity, commonly seen as skill level increases, is not simply indicative of cortical deactivation, but is indicative of neural reorganization concomitant with the acquisition of more efficient, task-specific cognitive and motor processes (Nunez, 1995; Smith, McEvoy, & Gevins, 1999).

Following this logic, Haufler, Spalding, Santa-Maria, and Hatfield (2000) investigated differences in EEG power between the left and right hemispheres of both novice and expert marksmen. They predicted that since novice shooters should lack task-specific strategies, they should rely more heavily on verbal mediation and effortful processing than the experts in the period prior to firing their rifles. Their prediction that these processing differences between skill groups would manifest in differential EEG laterality was partially born out by their finding that novices showed significantly reduced alpha in the LH, albeit in a restricted range of only 10-11 Hz. Again, as in the Hatfield et al. (1984) study, the RH did not show significant differences in EEG between the experimental conditions. Hatfield and Hillman (2000, p.366) interpreted this EEG pattern as supporting the notion that "true novices are less efficient in their resource allocation." It maybe premature to draw any strong conclusions at this point, as the role of the RH in expert performance is still largely unclear.

Underscoring the idea of differential resource allocation are the findings of Janelle, Hillman, Apparies, Murray, Meili, Fallon, and Hatfield (2000) who found that less experienced marksmen demonstrated increased alpha power in both hemispheres compared to experts. However, they did find differences in the beta range such that both groups showed equivalent power in the LH, but experts showed significantly less RH beta activity. This finding was interpreted to signify a larger degree of hemispheric specificity in the expert group, presumably underlying the development of practice-related strategies. Another informative study was conducted by Landers, Han, Salazar, Petruzzello, Kubitz, and Gannon (1994) who began with relatively inexperienced archers and measured changes in EEG asymmetry as they progressed though a two and a half month training program. As in previous studies, these authors found that increased skill level was associated with increased LH alpha power, with little change in RH activity. Taken in total, the evidence reviewed here seems to suggest that a critical component in expert performance seems to be the silencing of LH-based verbal/analytic, or "self-talk" strategies prior to task initiation.

The current investigation most closely replicates and extends the work of Crews and Landers (1993), who examined hemispheric differences in EEG observed prior to putting in a sample of elite golfers. Golfers completed 40, 12 foot putts in a laboratory setting watched only by the experimenters. The data was divided into 1 second epochs starting 3 seconds prior to contact with the ball. They hypothesized, based on the Hatfield model of LH processes interfering with marksmen's' performance, that LH activity would decrease while RH activity would remain relatively stable as the participants got closer to striking the ball.

In relation to performance, only right hemispheric alpha activity was significantly associated with accuracy. Increased alpha activity in the right hemisphere correlated with increased accuracy. Their EEG results revealed that, for motor cortex (electrodes placed near C3 and C4), the LH showed significantly increased alpha, decreased beta I (13-20 Hz), and no change in beta II (21-30 Hz) compared to the RH as the golfer approached the putt. The RH motor electrodes showed a significant increase in beta II activity. For the temporal cortex (T5 and T6) the beta results were the same, but the alpha results were interesting in that RH power decreased significantly over time compared to an LH alpha increase over time in motor cortex. They discussed their finding of increased beta II activity in terms of its possible linkage to a state of anxiety brought on by the task.

EEG Asymmetry and Emotion

EEG asymmetries, particularly in the alpha band, have been linked in many studies to both state and trait aspects of emotion (see Davidson, 1995; 2004, Allen & Kline, 2004; Kline, Blackhart, & Joiner, 2002 for reviews). In this literature, it is assumed that alpha activity is an inverse index of cortical activity. Appetitive, approach-related motivation and emotion have been shown to relate to relative left frontal and anterior temporal activity, whereas negative, withdrawal-related motivation and emotion have been shown to relate to relative right frontal and anterior temporal activity. This has been documented with state-dependent responses to affective stimuli (e.g. Kline, Blackhart, Woodward, Williams, & Schwartz, 2000; Davidson & Fox, 1982; Fox & Davidson, 1986), as well as in trait-related affective dispositions (Tomarken & Davidson, 1994; Sutton & Davidson, 1997, Harmon-Jones & Allen, 1998). Other work has suggested that relative right posterior activation, i.e. in parieto-temporal regions, is related to the cortical representation of arousal (see Heller, 1993; Allen, Iacono, Depue, and Arbisi, 1992).

The affective component of EEG laterality may also relate to athleticism and athletic performance. Petruzzello and Tate (1997) reported that individuals with relative left frontal activation pre-exercise reported anxiety reduction in response to exercise, whereas individuals with relative right frontal activation pre-exercise showed increases in anxiety. Petruzzello, Hall, and Ekkekakis (2001) found that physically fit individuals who showed relative left frontal activity pre-aerobic exercise showed increased positive affect from pre to post-exercise. Such studies point to the importance of considering the affective component of EEG asymmetries within the context of athletic performance, and may reflect the degree of task-engagement and motivation, in addition to the cognitive aspects of performance.

Social Inhibition, Stress, and EEG

The question of whether an audience has a positive (social facilitation) or negative (social inhibition) effect on sport performance has long been of great interest (Triplett, 1897-1898). This is clearly a complex issue, and leads to a number of salient questions. For example, the audience may facilitate performance in some sports and impair it in others. Football games are typically raucous events, with loud cheering, wild audience behavior, and expressive announcers. By contrast, audience noise is typically kept to a minimum during golf, and the announcers speak quietly.

The degree of expertise in a sport may moderate the effect of an audience on performance, an effect that represents the expression of a larger emotional phenomenon. In his review of early research on social facilitation, Zajonc (1965, p. 6) stated that, "the emission of well-learned responses is facilitated by the presence of spectators, while the acquisition of new responses is impaired." This finding has held up remarkably well in the literature (Strauss, 2002), and has often lead to the prediction that experts will benefit from an audience while novices' performance will suffer. The explanation offered for this phenomenon is that audiences lead to increased arousal, and that this in turn leads to higher levels of effort. Alternately, the theories of evaluation apprehension (Geen, 1989), and distraction-conflict theory (Baron, 1996) have been used to explain social mediation of general performance levels.

In most cases the theories make the same prediction: For the expert, increased effort should result in better cognitive and motor preparation and task execution, while the reverse should happen for the novice who has yet to achieve any degree of automaticity in the relevant motor acts. The prediction about putting behavior arising from this theory is straightforward: novices should be less accurate in front of an audience. The EEG record should therefore reflect the neural state associated with the stress of social inhibition experienced by the novices. Based on the literature reviewed above, the EEG should exhibit evidence of increased arousal, most likely evidenced by increased beta activity in the audience-present condition. Due to the inconsistencies in the literature reviewed above, it is somewhat more difficult to make a prediction about the effect of social inhibition on the laterality of alpha and beta. An affective/cognitive framework, such as that described in Kline, Blackhart, and Joiner (2002), would make the fairly straightforward prediction that under the audience condition, novice golfers would experience a higher degree of task-related withdrawal motivation (due to the aversive nature of social inhibition) and so would demonstrate an RH alpha laterality shift.

While there appear to be no published studies that have systematically manipulated the presence or absence of an audience during a motor act in order to observe changes in EEG, a prediction can be made based on a number of studies that have looked at how other stressors influence performance-related EEG. Saarela (2000) examined the effects of time pressure on the accuracy and EEG of 12 marksmen who were required to complete 40 shots in a normal 80 minute, and also a rushed 40 minute condition. Using leads placed at frontal (F3--F4 and F7-F8) and temporal sites (T3--T4) Saarela predicted that time pressure would be associated with right frontal hypoactivation/left frontal hyperactivation, as reviewed above (Fox & Davidson, 1986). It was also predicted that time pressure would lead to interference in the normally observed pre-shot reduction in LH activity. This prediction is particularly important because Saarela hypothesized that stress would lead to difficulty in quieting the LH verbal mediation that experts have been shown to suppress when shooting alone. The hypotheses were partially supported, in that the time pressure condition produced significantly RH shifted, log-transformed EEG alpha power and hemispheric asymmetry scores (right alpha power--left alpha power), consistent with negative affect. However, at the temporal sites, a bilateral increase in alpha amplitude was found in the time pressure condition. This result was interpreted by the author as indicating a decrease in allocation of neural resources to the temporal region

Alternately, it could be that the global decrease in alpha was actually an indication of a global arousal increase associated with the stress of the time pressured situation. For the present study, focusing on novice golfers, we predicted that in addition to increased global arousal, the audience would lead to specific increases in LH activation as the putt approached. The novices should both: 1) find the presence of the audience to be an arousing stressor (as opposed to experts who are more used to public competition), and 2) lack automatic, task-specific skills that could be relied upon to complete the putt. Thus, the novice putters should be even less able to reduce LH verbal processing than they normally would.

Methods

Participants

The participants were volunteers from introductory psychology classes at a mid sized southeastern university. They reported having little to no golf experience. Participants received credit in fulfillment of a course requirement. Of the individuals (n = 78) who volunteered for the study, 20 (14 females, 6 males) were randomly selected to putt, and the rest were used as members of an audience. The mean age of the participants was 21.7 years old. All participants who performed the putting task were right handed.

Procedure

Volunteers came in groups consisting of three to six people. They were first briefed and asked to sign a consent form. Next, one member of each group was randomly selected to be monitored on the EEG while performing a series of putts. Others in the group acted as the audience. Audience sizes ranged from 2 to 5 observers Volunteers who approached the green and putted left-handed were automatically placed in the audience, because the EEG and putting green were set up for right-handers. The EEG was attached to each participant selected to putt. Each participant was then asked to putt 20 times with no audience, and 20 times while the audience group watched. The audience group was not instructed to evaluate the participant, simply to "watch attentively and quietly" while the participant putted. The experimenter was present for both conditions to operate the EEG equipment. In an effort to counter-balance, participants were randomly assigned to either putt alone first, then in front of the audience, or vice versa. In order to control for pre-performance anxiety, whenever the participants putted by themselves first, they were not informed they would subsequently be putting in front of the audience. Immediately after being chosen, the golfers who were selected to putt were asked to complete a self-report mood measure, the Profile of Mood States (POMS). They were then administered the POMS again at the end of the experiment.

The Profile of Mood States

The POMS (McNair, Lorr and Droppleman, 1971), was developed as a measure of "right now" kinds of mood states, particularly in people undergoing counseling or psychotherapy. However, the POMS quickly gained popularity in sports and exercise (LeUnes, 2000; LeUnes & Burger, 1998). The POMS is a self-report instrument consisting of six scales--Tension-Anxiety, Depression-Dejection, Anger-Hostility, Vigor-Activity, Fatigue-Inertia and Confusion-Bewilderment. A study by Beedie, Terry, and Lane (2000), which performed a meta-analysis of published studies using the POMS to investigate relationships between mood and performance outcome, found that the POMS seems to have utility in the prediction of performance outcome, and was therefore expected to be sensitive to any stress produced by the social inhibition manipulation.

Performance Measures

The accuracy of each putt was measured by the distance in centimeters (cm) between the closest rim of the hole and the nearest edge of the ball after it came to rest. The maximum distance possible was 61cm due to the limitations of the indoor putting surface. If the putt went off the green, 61cm was recorded.

EEG Measures

Biopack acquisition software BSL Pro was used to obtain the raw EEG data. Six 7mm sintered Ag-AgCl leads were used, and a Biopac MP30 was used to amplify and transfer the data to an IBM Think Pad for storage. The sampling rate for all data was 200 Hz. Hardware filters were set at a bandpass of .5 to 38.5 Hz with a filter rolloff of about 12 db/octave, with an additional 60 Hz notch filter enabled. Leads were attached over the temporal lobe at T3 and T4, and also over the motor cortex at C3 and C4, in accordance with the International 10-20 system. The other two leads were a ground placed on the forehead, and a reference placed on the nose. The impedance for each electrode was tested before each putting condition and was kept below 5 kiliohms.

Data Reduction

The EEG data from all four sites (T3, T4, C3 and C4) was originally saved as a raw-ASCII file. The data was converted to a NeuroScan continuous file for further analysis. Each continuous file was epoched using event files and two epoch files were made for each putt--a backswing epoch and a contact epoch. Each epoch file consisted of the 1.275 seconds of data recorded just prior to each event. All of the backswing epoch files for each participant, 40 total, were combined into two separate files---one consisting of the 20 putts performed in front of the audience and the other consisting of the 20 putts performed alone. The same procedure was applied to the contact epoch files. Artifact rejection was performed first by visual inspection of the waveforms. Epochs containing visible movement, EMG, blink, or other artifact were manually rejected. These files were then transformed by linear detrend. Automatic artifact rejection removed any epoch with amplitude of greater than +/-75 microvolts.

Spectral averaging was performed on the artifact-free epochs. The resulting averaged files had a mean of 12 accepted epochs. The average spectral power (amplitude squared) for the alpha, beta 1 and beta 2 bands were computed using a FFT transform. A 10 percent Hamming window was applied to taper the epochs in order to prevent spectral leakage.

Data Analysis

Several multivariate analyses of variance (MANOVAs) were performed on the averaged power band data (alpha, beta 1 and beta 2) for each subject. The repeated measures factors were Audience (no audience/audience), Pre-putt Epoch (backswing/contact), and in some cases, hemisphere (LH/RH).

Laterality coefficients were computed in order to directly compare the relative amount of activity between the hemispheres. The laterality coefficients in this experiment where calculated using the formula:

 Laterality coefficient = Log (left hemisphere power)--Log (right
 hemisphere power), where a positive value corresponds to a larger
 left hemisphere value, and a negative value with a larger right
 hemisphere value.

A paired sample t-test was performed on the putting accuracy data to determine if there was a difference in distance from the hole as a function of the group conditions. Correlational analyses were performed on the putting accuracy and the EEG band power data to determine which, if any, EEG measures were associated with accuracy.

Results

As predicted, the presence of an audience led to decreased putting accuracy (t(19) = -2.09, p = .025). The mean distance in centimeters (cm) from the cup while putting alone was 32.3 (SD = 9.5), compared to 36.7 (SD=9.6) while in the being watched by the audience condition (MEAN = 36.7, SD = 9.6). The presence of an audience resulted in an average 4.4 cm increase in distance from the hole.

Also as predicted, the presence of an audience resulted in a relatively global increase in beta 1 and beta 2 band power across the cortex. The means of the alone condition in both beta 1 and beta 2 were lower than the means for the group condition and therefore supported the hypothesis. This can be seen in Table 1. An Epoch (backswing/putt) X Audience (alone/ audience) X Hemisphere (LH/RH) ANOVA revealed a significant main effect of audience for both beta 1 (F(1, 14) = 7.464,p = .016) and beta 2 (F(1, 14) = 8.790,p = .010) in temporal regions (T3 and T4). The same was true for the motor cortex (C3 and C4) for beta 1 (F(1, 14) = 4.278, p = .048) and beta 2 (F(1, 14) = 5.938,p = .029). There were no significant effects for the alpha band.

Data from the pre-putt epoch was analyzed to test the hypothesis that the novice participants would exhibit greater LH than RH activity overall. An ANOVA with Audience (alone/ audience) and Hemisphere (LH/RH) as within-subject factors was also computed on the mean alpha power values for both temporal and central sites. No effect of Audience or interaction with Audience and Hemisphere emerged (p < .05), but there were main effects of Hemisphere for both central (F(1, 15) = 8.654,p = .010) and parietal (F(1, 15) = 9.480,p = .008) sites. Thus, support was provided for the hypothesis.

In order to test the hypothesis that the presence of an audience would increase EEG hemispheric asymmetry in the beta range, an ANOVA was first performed on the mean amplitudes of the LH and RH alpha, beta1 and beta2 bands with factors Epoch, Hemisphere, and Audience. This ANOVA yielded a marginally significant interaction of Epoch and Hemisphere for the beta1 band (F(1, 14) = 4.034,p = .06). Also, a marginally significant interaction of Epoch, Hemisphere, and Audience was also seen for the alpha band (F(1, 14) = 4.107,p = .062). An examination of Table 1 reveals the finding that there is an ERD/increase in Alpha power/cortical deactivation greater on the right side than left side as the participants get closer in time to the putt. Note that the mean difference in alpha power is about -.08 (right -left) for temporal and motor cites in the backswing epoch, in the pre-putt epoch the difference is 1.129 and .554, respectively.

In order to isolate the relative activity while deemphasizing absolute amplitude differences between channels, laterality coefficient (see above) analysis was then performed. Two separate 2 X 2 (Epoch X Group) ANOVAs were performed on the laterality coefficients of each band for both the temporal and motor cortex. The temporal lobe ANOVA yielded a significant main effect of being watched in the univariate test of the beta 1 band only (F(1, 14) = 5.835,p = .030) (See Figure 1). The ANOVA for the motor cortex had no significant findings. The means for the temporal lobe indicated a relative increase in left side activity when the participants were being watched.

[FIGURE 1 OMITTED]

The final hypothesis to be tested was that putting accuracy would correlate positively with RH alpha activity over the central cortex in the epoch just prior to putting. Mean power scores for all bands in the pre-putt epoch were correlated with the average putting distance from the hole for each participant. The results can be seen in Table 2. When participants were putting alone, their accuracy was significantly correlated with both beta1 and beta2 activity at central and right temporal sites, but not at left temporal regions. Interestingly, the same pattern of central and right temporal correlations held when individuals were putting in front of an audience, but only in the beta1 band. The positive correlations in all cases indicate that greater beta activity was correlated with increased distance from the hole, and thus less accuracy. Alpha revealed no significant relationship to putting accuracy (p's > .05). An analysis was conducted to determine whether changes in EEG activity from backswing to putting epoch were predictive of putting accuracy, and no significant correlations were obtained. Also, a correlational analysis was undertaken to determine whether the difference in accuracy between the alone and group conditions correlated with difference between those conditions for any EEG band, which produced no significant correlations.

Discussion

The first hypothesis examined the effect of an audience on performance. It was predicted that being watched would result in social inhibition for the novice putters, leading to a decrease in the accuracy of putting. The putting accuracy data showed a significant 4.41 cm decrease in accuracy when participants putted in front of an audience, compared to when putting alone. This result supported the first hypothesis and suggested that subjects, on average, experienced social inhibition. The POMS data was suggestive of the idea that people found putting in front of a group of strangers to be negatively arousing. A significant drop in TA was noted after the participants finished the experiment (p < .05). Figure 6 shows the mean POMS Tension-Anxiety(TA) subscale scores for the participants in both the alone and audience conditions, as a function of which condition they received first. Note that the difference between the alone and audience conditions is larger in the "group last" condition. The logic of this design was that participants who had finished putting in front of an audience last should show the effects of that experience in a more pronounced way than those who had just putted alone. While the means are consistent with this notion, no significant interaction emerged. It is likely that the situation, while producing some social inhibition, did not produce the type of negative feelings commonly detectable on the POMS.

[FIGURE 2 OMITTED]

The second hypothesis predicted that the presence of an audience would increase beta 1 and beta 2 band power across the cortex. This hypothesis was supported, and thus provides some evidence for the idea that the neural correlates of social inhibition are detectable and quantifiable in the EEG record. As indicated above, the literature has been inconsistent in regards to interpretation of the functional significance of beta activity. It is likely that the audience results in an increase in global cortical arousal, as beta has often been linked to increased cortical activity. Crews and Landers (1993) speculated that their beta II increase was a marker of task-related anxiety, based on the PET scan results of Reiman et al. (1989). This interpretation, while still tenable, is made more complicated by the finding that the data used in that study was likely confounded by EMG activity (Drevets, Videen, & MacLeod, 1992).

The third hypothesis predicted that regardless of audience condition participants should show relatively greater LH than RH activity in the pre-putt epoch, as demonstrated by higher amplitude RH than LH alpha activity. This hypothesis was supported, replicating the basic finding of Landers et al. (1994) that novice athletes performing a bimanual task have higher levels of relative LH than RH cortical activity.

The last major hypothesis was that the presence of an audience would increase hemispheric asymmetry in the beta range, such that the difference between the alone and audience conditions will be greater in the LH. This hypothesis was supported by the finding that the laterality coefficients were greater in the audience condition than the alone condition. This result is compatible with the interpretation that social inhibition leads to increased arousal, and that this arousal in turn interfered with the normal quieting of LH processes, such as self-talk. These results support the original hypothesis of Saarela (2000), who predicted that a stressor would lead to greater relative LH activation. It may be that the subjects selected for the present investigation, who were true novices, were better suited to reveal this effect because they would likely be more sensitive to the audience manipulation, as an audience should be more novel and thus more arousing.

As mentioned above, temporal lobe activity, like frontal activity, has been linked to cognitive/emotional states (Kline, Blackhart, & Joiner, 2002). In such theories, the presence of an aversive stressor (audience) would be predicted to be accompanied by a relative increase in hemispheric asymmetry, such that LH alpha would be greater than RH alpha. This type of result has often been interpreted as relative LH cortical deactivation, which would be a cortical correlate of a negative mood state. In order to examine the possibility of the above pattern in the present data set, the results were analyzed in a follow-up ANOVA that revealed an ERD/ increase in Alpha power/cortical deactivation greater on the right side than left side as novice prepares to putt. Thus for this study, the straightforward prediction of negative mood being linked to relatively greater RH activation was not supported. It is entirely possible that the more commonly measured frontal electrodes would have revealed such a difference, as found by Saarela (2000), and future studies should record from motor, temporal, and frontal sites in order to allow for a true examination of separate cognitive, emotional, and motor preparatory components of sport behavior.

The pattern of correlational results found in this study is different from that reported by Crews and Landers (1993). Those authors found that an increase in RH alpha was correlated with an improvement in accuracy. They interpreted their finding as evidence that the RH was experiencing decreased cortical activity. This is plausible in that only the motor cortex, not the central cortex site demonstrated a correlation with accuracy. In contrast to their results, this study found that lower levels of beta power at bilateral central and RH motor areas were correlated with accuracy. However, like Crews and Landers, the LH temporal cortex was not relevant for accuracy. The most obvious explanation for the difference in findings was the skill level of the participants, as the pattern in the present study was similar in both the alone and audience conditions. In support of this explanation of the results, Deeny, Hilman, Janelle, and Hatfield (2003) found that expert marksmen showed greater inter-site EEG coherence than skilled shooters. The participants in the current study demonstrated a more defuse pattern of cerebral activation associated with putting performance than the experts in the Crews and Landers or the Deeny et al. studies. Deeny et al. offered the explanation that skilled athletes require less cortico-cortical communication to execute a task than novices, reflecting the development of efficient task-specific strategies that display a relatively high level of automaticity.

Based on the results of the current study, several additional investigations are warranted. First, the current paradigm should be replicated using both novice and expert golfers in order to determine directly whether the presence of an audience affects them differentially. Second, additional mood measures should be employed to lend corroborating evidence about the functional significance of the beta elevations associated with an audience. In addition, the paradigm should be extended to other sport situations, including comparisons of both dominant-hand-controlled bimanual (rifle, billiards) and more balanced (golf) manual acts. Crews et al. (1993) pointed out that it is unclear to what extent the right-dominant nature of the shooting tasks, which are the most prevalent in the literature, account for the asymmetries reported in the literature.

It should be noted that the use of an audience is different in this study than it would be during competitive play during any spectator sport. While the typical audience at a sporting event is large, physically removed, and composed of largely anonymous figures (unidentifiable by the athlete), the present study subjected the putters to close inspection by a group of academic peers. This situation more closely resembles a golf class (academic or private), or a game of recreational golf played with new business associates or round-robin group at a local course. A possible implication of this situation is that the results of this study may generalize better to recreational golf situations than professional golf situations. One could speculate that in the case of a collegiate or professional tournament, the athlete might be subjected to a similar level of scrutiny by competitors at a putting green by a busy hole. In any event, an attempt should be made in future investigations to replicate the current results under a range of different athletic situations.

Perhaps the most important finding in the present study was that the beta band is sensitive to the presence of an audience during execution of a motor act. This finding may open the door to a number of additional investigations. Considering that every professional sport, by its definition, is observed by spectators, it is surprising how little is known about the changes in brain physiology that accompany athletes' positive and negative reactions to being watched.

References

Allen, J. J. B. & Kline, J. P. (2004). Frontal EEG asymmetry, emotion, and psychopathology: The first, and the next, twenty-five years. Biological Psychology, 67, 1-5.

Allen, J. J., Iacono, W. G., Depue, R. A., & Arbisi, P. (1993). Regional electroencephalographic asymmetries in bipolar seasonal affective disorder before and after exposure to bright light. Biological Psychiatry, 33, 642-646.

Baron, R. A., (1996). Motivation and emotion. In S. Badger (Ed.), Essentials of Psychology (pp. 333-368). Needham Heights, MA: A Simon & Schuster.

Beedie, C.J., Terry, P.C., & Lane, A.M. (2000) The profile of mood states and athletic performance: Two meta-analyses. Journal of Applied Sport Psychology, 12, 49-68.

Bird, E. I. (1987) Psychophysiological processes during rifle shooting. International Journal of Sport Psychology, 18, 9-18.

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.

Davidson, R. J. (1995). Cerebral asymmetry, emotion, and affective style. In R.J. Davidson & K. Hugdahl, (Eds.). Brain Asymmetry (pp. 361-387). Cambridge: MIT Press.

Davidson, R. J., & Fox, N. A. (1982). Asymmetrical brain activity discriminates between positive and negative affective stimuli in human infants. Science, 218, 1235-1237.

Davidson, R. J., Marshall, J. R., Tomarken, A. J., & Henriques, J. B. (2000). While a phobic waits: Regional brain electrical and autonomic activity in social phobics during anticipation of public speaking. Biological Psychiatry, 47, 85-95.

Davidson, R.J. (2004) What does the prefrontal cortex "do" in affect: Perspectives on frontal EEG asymmetry research. Biological Psychology, 67, 219-233.

Deeny, S.P., Hillman, C.H., & Janelle, C.M. (2003) Cortico-cortical communication and superior performance in skilled marksmen: An EEG coherence analysis. Journal of Sport & Exercise Psychology, 25, 188-204.

Drevets, W. C., Videen, T. O., & MacLeod, A. K. (1992) PET images of blood flow changes during anxiety: Correction. Science, 256, 1696.

Fox, N. A., & Davidson, R. J. (1986). Taste-elicited changes in facial signs of emotion and the asymmetry of brain electrical activity in human newborns. Neuropsychologia, 24, 417-422.

Geen, R.G. & Bushman, B.J. (1989) The arousing effects of social presence. In Wagner, H., & Manstead, A. (Eds.), Handbook of social psychophysiology (pp. 261-281). Oxford, England: John Wiley & Sons.

Harmon-Jones, E. & Allen, J. J. B. (1998). Anger and frontal brain activity: EEG asymmetry consistent with approach motivation despite negative affective valence. Journal of Personality and Social Psychology, 74, 1310-1316.

Hatfield, B.D., & Hillman, C.H. (2000) The psychophysiology of sport: A mechanistic understanding of the psychology of superior performance. In Singer, R.N., Hausenblas, H.A., & Janelle, C.M. (Eds.) Handbook of sport psychology (2nd ed.). New York, NY, US: John Wiley & Sons, Inc, 2001. xix, 876 pp.

Hatfield, B.D., Haufler, A.J., Hung, T., & Spalding, T.W. (2004) Electroencephalographic studies of skilled psychomotor performace. Journal of Clinical Neurophysiology, 21, 144-156.

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.

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.

Haufler, A.J., Spalding, T.W., SantaMaria, D.L., & Hatfield, B.D. (2002). Neuro-cognitive activity during a self-paced visuospatial task: comparative EEG profiles in marksmen and novice shooters. Biological Psychology, 59, 87-88.

Heller, W. (1993). Gender differences in depression: Perspectives from neuropsychology. Journal of Affective Disorders, 29, 129-143.

Hillman, C.H., Apparies, R.J., Janelle, C.M., & Hatfield, B.D. (2000). An electrocortical comparison of executed and rejected shots in skilled marksmen. Biological Psychology, 52, 71-83.

Janelle, C.J., Hillman, C.H., Apparies, R.A., Murray, N.P., Meili, L., Fallon, E.A., Hatfield, B.D. (2000). Expertise differences in cortical activation and gaze behavior during rifle shooting. Journal of Sport and Exercise Psychology, 22, 167-182.

Kimura, M., Mori, T., & Suzuki, H. (2001). Journal of International Society of Life Information Science, 19, 271-274.

Kline, J. P., Blackhart, G. C., & Joiner, T. E. (2002). Sex, lie scales, and electrode caps: An interpersonal context for defensiveness and anterior electroencephalographic asymmetry. Personality and Individual Differences, 33, 459-478.

Kline, J. P., Blackhart, G. C., Woodward, K. M., Williams, S. R., & Schwartz, G. E. R. (2000). Anterior electroencephalographic asymmetry changes in elderly women in response to a pleasant and an unpleasant odor. Biological Psychology, 52, 241-250.

Landers, D.M., Han, M., & Salazar, W., Petruzello, S.J., Kubitz, KA., & Gannon, T.L. (1994). Effects of learning on electroencephalographic andelectrocardiogrpahic patterns in novice archers. International Journal of Sport Psychology, 25, 313-330.

LeUnes, A. (2000). An updated bibliography on the Profile of Mood States in sport and exercise psychology research. Journal of Applied Sport Psychology Research, 12, 110-113.

LeUnes, A., & Burger, J. (1998). Bibliography on the Profile of Mood States in sport and exercise psychology research, 1971-1995. Journal of Sport Behavior; 21, 53-70.

McNair D, Lorr M, Droppleman LF. Manual for the Profile of Mood States. San Diego: EdITS, 1981.

Niedermeyer, E., & Lopes da Silva, F.H. (1999). Electroencephalography: Basic principles, Clinical Applications and Related Fields. Williams and Wilkins, Baltimore, MD.

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

Petruzzello, S.J., & Tate, A.K.(1997). Brain activation, affect, and aerobic exercise: An examination of both state-independent and state-dependent relationships. Psychophysiology, 34, 527-533.

Petruzzello, S.J., Hall, E.E., & Ekkekakis, P. (2001). Regional brain activation as a biological marker of affective responsivity to acute exercise: Influence of fitness. Psychophysiology, 38, 99-106.

Pullum, B. (1977). Psychology of Shooting. Schiessportschule Dialogues, 1, 1-17.

Ramos, J., Cerdan, L.F., & Guevara, M.A. (2001). Abnormal EEG patterns in treatment-resistant schizophrenic patients. International Journal of Neuroscience, 109, 47-59.

Rebert, C.S., Low, D.W., & Larsen, F. (1984). Differential hemispheric activation during complex visuomotor performance: Alpha trends and theta. Biological Psychology, 19, 159-168.

Reiman, E.M., Fusselman, M.J., Fox, P.T., & Raichle, M.E. (1989). Neuroanatomical correlates of anticipatory anxiety. Science, 243, 1071-1072.

Saarela, P.I. (2000). The effects of mental stress on cerebral hemispheric asymmetry and psychomotor performance in skilled marksmen. Dissertation Abstracts International: Section B: The Sciences & Engineering, 61, 580.

Smith, M.E., McEvoy, L.K., & Gevins, A. (1999). Neurophysiological indices of strategy development and skill acquisition. Cognitive Brain Research, 7, 389-404.

Strauss, B. (2002) Social facilitation in motor tasks: A review of research and theory. Psychology of Sport & Exercise, 3, 237-256.

Sutton, S.K., & Davidson, R.J. (1997). Prefrontal brain asymmetry: A biological substrate of the behavioral approach and inhibition systems. Psychological Science, 8, 204-210.

Tomarken, A. J., & Davidson, R. J. (1994). Frontal brain activation in repressors and nonrepressors. Journal of Abnormal Psychology, 103, 339-349.

Triplett, N. (1898). The dynamogenic factors in pacemaking and competition. American Journal of Psychology, 9, 507-533.

Williams, J.M., & Krane, V. (1998). Psychological characteristics of peak performance. In J.M. Williams (Ed.), Applied Sport Psychology. (pp. 158-170). Mountain View, CA. Mayfield.

Zajonc, R.B. (1965). Social facilitation. Science, 149, 269-274.

John F. Shelley-Tremblay John D. Shugrue and John P. Kline

University of South Alabama

Address Correspondence To: John F. Shelley-Tremblay, Ph.D., Department of Psychology, University of South Alabama, Mobile, AL. 36688;jstremblay@usouthal.edu.

Table 1.
Mean EEG Power by Epoch, Audience Condition, Spectral Band and
Electrode

Epoch Audience Band T3 C3 C4 T4

Backswing Alone Alpha 2.391 2.567 2.487 2.303
 Beta1 1.170 1.102 1.056 1.062
 Beta2 0.976 0.758 0.684 0.788

 Group Alpha 2.290 2.495 2.505 2.403
 Beta1 1.565 1.316 1.223 1.294
 Beta2 1.544 1.026 0.916 1.150

Putt Audience Band T3 C3 C4 T4

 Alone Alpha 2.779 3.459 4.013 3.908
 Beta1 1.570 1.527 1.542 1.592
 Beta2 1.029 0.834 0.827 0.951

 Group Alpha 2.893 3.405 3.884 3.735
 Beta1 1.866 1.679 1.685 1.777
 Beta2 1.471 1.075 1.034 1.307

Table 2.
Pearson Correlations between Putting Accuracy in the Alone
Condition and EEG Power

Epoch Audience Band Electrode R p

Putt Alone Beta1 C3 .635 0.015
 C4 .697 0.006
 T4 .737 0.003

 Alone Beta2 C3 .584 0.028
 C4 .680 0.007
 T4 .682 0.007

 Group Beta1 C3 .591 0.026
 C4 .620 0.018
 T4 .583 0.029