This article was downloaded by: [University of Auckland Library] On: 12 December 2014, At: 01:17 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Sports Sciences Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rjsp20
Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation a
J.J. Adam & R.B. Wilberg
b
a
Department of Movement Sciences , University of Limburg , PO Box 616, Maastricht, 6200‐MD, The Netherlands b
Department of Physical Education , University of Alberta , Edmonton, Alberta, T6G 2H9, Canada Published online: 14 Nov 2007.
To cite this article: J.J. Adam & R.B. Wilberg (1992) Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation, Journal of Sports Sciences, 10:3, 261-273, DOI: 10.1080/02640419208729925 To link to this article: http://dx.doi.org/10.1080/02640419208729925
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Sports Sciences, 1992,10, 261-273
Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation Downloaded by [University of Auckland Library] at 01:17 12 December 2014
J.J. ADAM 1 * and R.B. WILBERG 2 1 Department of Movement Sciences, University of Limburg, PO Box 616, 6200-MD Maastricht, The Netherlands and 2Department of Physical Education, University of Alberta, Edmonton, Alberta T6G 2H9, Canada
Accepted 2 August 1991
Abstract
This study used a backward-masking paradigm to examine individual differences in rate of visual information processing among university basketball, ice hockey and Canadian football players. Displays containing four letters were presented for stimulus durations ranging from 25 to 300 ms. Following stimulus offset, a masking stimulus was presented for 200 ms. The subjects were instructed to write down as many letters as possible from the briefly presented stimulus display on a specially prepared response grid. The results indicated consistent individual differences in rate of visual information processing. More importantly, it was found that rate of visual information processing as indexed by the backward-masking technique, has promising validity for predicting general performance excellence in university ice hockey and basketball players. Individual differences in rate of visual information processing were interpreted as reflecting the operation of attentional factors. Keywords: Individual differences, information processing, skilled performance, team sports, visual perception. Introduction
The ability to process visual information quickly is a factor of crucial importance in many skills, sports and games. Consider, for example, such fast action motor skills as those required when batting in baseball or cricket, goal-keeping in hockey, and punching in boxing. The idea that high-speed visual processing plays a pivotal role in 'open' skills, suggests that performance excellence could be related to a performer's processing capacity and rate. The assumption behind this speculation is that athletes who possess the ability to process more visual information in a shorter period of time than their competitc r s will have an advantage. This advantage can take two forms (Deary and Mitchell, 1989); perceptually quick athletes may instruct their motor effector processes sooner and/or they may provide the decision processes with more detailed information, increasing the chances of an appropriate response. * To whom all correspondence should be addressed. 0264-0414/92
© 1992 E. & F.N. Spon
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
262
Adam and Wilberg
Of specific concern, then, are perceptual processes that come into play when performers are faced with visual displays of very short duration. There is considerable evidence to suggest that people do differ in their ability to process short duration visual displays. This evidence is based on backward-masking studies in which a subject's recognition of a brief target stimulus is impaired by a masking stimulus that follows the initial display. Before discussing this evidence, it seems useful to introduce briefly the concept of short-term visual storage - or iconic memory, as Neisser (1967) termed it since it is regarded by many theorists as a key component in visual information processing theory (but see Haber, 1983). Using the partial report technique, Sperling (1960) demonstrated convincingly that following display offset, subjects have much more information available than they can consciously report. Sperling even suggested that perhaps all of the information in the display is represented in what he called 'visual information storage' and Neisser (1967) later dubbed 'iconic memory'. Even though the nature of the representations in iconic memory is a matter of dispute (for a discussion, see Coltheart, 1980; Kahneman and Treisman, 1984), it is generally agreed that the contents of iconic memory are accessible for further processing for a very short duration (~250 ms), after which they will have decayed away. Following his work on the experimental demonstration of iconic memory, Sperling (1963) went on to investigate the rate at which information from iconic memory could be 'read out' and thus reported. For different stimulus durations, he presented slides containing six letters, followed by a masking stimulus consisting of random black and white squares. Sperling found that the number of letters reported correctly increased regularly up to 75 ms of stimulus duration, at which point four items could be reported. Additional stimulus duration up to 200 ms resulted in one more letter being reported. In other words, letters seem to be initially read out from iconic memory at a high rate, and subsequently at a low rate. Similar results have been reported by other researchers (for a review, see Coltheart, 1972). The rate at which material in iconic memory can be 'read out' varies considerably from one individual to another. Turvey (1973) and Marcel (1983) reported large inter-individual variations in ability to detect or recognize a word following the use of a pattern mask. Di Lollo et al. (1982) and Walsh (1976) found evidence supporting a drop in processing rates as age increases. Galbraith and Gliddon (1972), Mosley (1981) and Baumeister et al. (1984) found support for intelligence-related differences in visual information processing rates among high- and low-IQ subjects. Importantly, Deary and Mitchell (1989) reported that successful batsmen in cricket were faster at picking up information from briefly presented visual displays than less successful batsmen. Besides studies of the ability to process rapid visual displays, there have also been studies of sport-specific or context-dependent perceptual skills. In general terms, the basic finding is that high-level players have superior perceptual skills in situations particular to their sport only. For example, Allard et al. (1980) compared the performance of basketball players and non-players on a task requiring the recall of slides containing structured and unstructured game information, after a 4-s viewing period. They found that basketball players were superior to non-players in recall of structured (context-dependent) slides only. As pointed out by Allard et al. (1980), similar results have been established for the games of chess (De Groot, 1965; Chase and Simon, 1973), Go (Reitman, 1976) and bridge (Charness, 1979). The interaction between the superior perception of skilled athletes and the structured displays of their sport-specific environment suggests that an 'encoding of structure' is important in a performer's excellence (Allard et al, 1980; Allard and Starkes, 1980).
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Visual information processing rate
263
In contrast to the high-speed visual processing paradigms that employ a backwardmasking stimulus, sport-specific perceptual studies generally employ long periods of display availability, such as 4-5 s of viewing time. The difference between these two approaches suggests that the underlying mechanisms may also be different. The ability to extract visual information from displays of short duration may be a general ability, as opposed to the context-dependent perceptual skill that develops as performers gain situation-specific experience. Proof of such a general factor would require an experimentally significant main effect in faster recognition of non-contextually specific stimulus material between experienced athletes who differ in excellence. The approach taken in this study to demonstrate that speed of visual information processing is an important factor in predicting performance excellence in open skills differs in several respects from the interaction paradigm. First, type of stimulus material is not varied systematically, but is held constant and neutral (i.e. letters of the alphabet). Secondly, stimulus duration is not held constant at a relatively long duration, but varied systematically from 25 to 300 ms. Moreover, following stimulus offset, a masking stimulus is employed to control the functional lifetime of the test stimulus. Thirdly, there are no marked differences between the athletes in the subject pool in terms of experience and general skill level, in that they are all university athletes. This latter requirement takes into account the fact that performance success is undoubtedly determined by a variety of factors. If the groups to be compared are not homogeneous in terms of other potentially important factors influencing performance level, the experiment will be confounded and the effect of rate of processing could be 'swamped' by uncontrolled variance. Fourthly, the subjects are differentiated into two groups (i.e. 'top-ranked' vs 'bottom-ranked' groups) by the head coach, according to the criterion of performer's general excellence. Taken together, these experimental manipulations will make it possible to conclude that when a significant main effect in recognition speed is found between 'top-ranked' and 'bottom-ranked' athletes, rate of visual information processing is an important mediating variable in performance excellence in team sports.
Method Subjects A total of 22 male athletes from the University of Alberta served as subjects, 6 of whom were recruited from the university basketball team, 6 from the university football team (Canadian football) and 10 from the university ice hockey team. Selection was random. The basketball players (mean age 20.8 ±1.5 years) averaged 2.7 years of intercollegiate playing experience. The football players (mean age 19.8 + 2.0 years) averaged 2.0 years and the hockey players (mean age 22.2 + 1.8 years) averaged 3.3 years. All of the subjects were unpaid volunteers and had normal or corrected-to-normal vision. Apparatus The subjects stood in a normally lit room and viewed from above a 8.5 x 3.0 cm, horizontally placed, rectangular opaque rear projection glass screen at a distance of about 58 cm. The viewing distance was kept constant by employing a chin support. The test stimulus and masking stimulus were rear-projected tachistoscopically by two identical sets of apparatus,
264
Adam and Wilberg
each consisting of a Kodak slide projector (Ektagraphic), an Opticon Uniblitz electroprogrammable shutter (Model 262) and an Opticon Uniblitz shutter driver and timing unit (Model SD-10). Control of display parameters was established by a PDP-11/10 computer in conjunction with the Uniblitz SD-10 shutter control units.
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Stimuli The target stimuli consisted of 35-mm slides of four different letters arranged in a linear array. The letters for each array were selected randomly from all consonants (except for the letter 'y'), to minimize the possibility of subjects interpreting the arrays as words. A total of 100 target arrays was constructed. Two different sizes of letters were used for the elements within the target arrays. The two outer elements were 1.0 degree in height and 0.8 degree in width, while the two inner elements were 0.5 degree in height and 0.4 degree in width. The inner letters were spaced 0.5 degree apart, while there was a distance of 1.5 degrees between the inner and outer letters. These values were chosen from the guidelines established by Anstis (1974), so as to keep letter resolution above threshold at both retinal locations. The outer letters were printed in uppercase Helvetica Medium script, and the inner letters were printed in upper-case Helvetica Small script. Figure 1 schematically shows a target stimulus.
X
D
N
Fig. 1. A schematic representation of a target stimulus.
The masking stimulus consisted of a pattern of randomly superimposed letters of both sizes, covering the entire field. The target stimuli were presented for six different stimulus durations: 25,50,75,100,150 and 300 ms. Following target offset, the masking stimulus was displayed for 200 ms.
Procedure The subjects were tested individually in a session lasting approximately 20 min. After entering the room, the subjects were informed about the nature of the experimental task. Each subject was then shown a different random sample of 10 target arrays at several stimulus durations to orient them to the task. Following these orienting trials, the subjects received a block of 15 trials for each stimulus duration. The first five trials of each block were considered practice trials, and were not included in the analysis. Each subject received the same order of presentation of the stimulus durations, which started with the longest (300 ms) and ended with the shortest (25 ms). Prior to the experiment, all of the slides were arranged in random order into one block of 10 trials (the orienting trials) and six blocks of 15 trials (the test trials). For each trial, the subjects were instructed to fixate on the centre of the screen and to give a verbal 'ready' signal. Then the experimenter pressed a button, which initiated a warning tone of 50 ms. After a delay of 1 s, the target stimulus appeared, which was followed at its offset by
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Visual information processing rate
265
the masking stimulus. The subjects were instructed to write down as many letters as possible from the briefly presented array on a specially prepared response grid. While the subject wrote down his response, the experimenter advanced the slide projector by remote control. It was emphasized that the letters had to be reported in the correct location and that guessing was allowed if one was uncertain. After testing was completed, the head coaches of the respective sports were asked to rank all their athletes who had participated in the experiment according to the criterion of'general excellence', with players of equal ability being given tied or equal ranks. Based on these evaluations, 'top-ranked' and 'bottom-ranked' groups were formed for each sport. That is, athletes in the top 30% of the rankings were categorized as 'top-ranked' athletes and athletes in the bottom 30% of the rankings were categorized as 'bottom-ranked' athletes. The athletes in the middle 40% of the rankings, therefore, were not included in the comparison between 'top-ranked' and 'bottom-ranked' athletes.
Results and discussion
Figure 2 displays the principal results of this experiment. The mean number of letters reported in the correct position is shown for each individual in his respective sport as a function of stimulus duration. An analysis of the experimental data based on the more lenient criterion of letters reported correctly, irrespective of correct location, yielded the same overall pattern of results. No inferential statistical analysis was performed on these data. Visual inspection of Fig. 2 reveals important performance differences between athletes on the backward masking task for all three sports. The athletes differed both in rate of information gain and in final amount gained, at least for the stimulus durations employed in the present experiment. The overall shape of the functions relating amount reported to stimulus duration, i.e. an initial steep slope followed by a relatively shallow slope, seems to hold for each individual and resembles closely those reported by other investigators (for a description, see Coltheart, 1972). The subjects within each type of sport were then divided into two groups based upon the total number of letters reported correctly. These median split divisions produced two groups in all three sports: (1) a 'fast' group, consisting of those athletes with the higher total number of letters reported correctly, and (2) a 'slow' group, consisting of those athletes with the lower total number of letters reported correctly. The result of this manipulation of the individual performance curves shown in Fig. 2 resulted in the group curves presented in Fig. 3. Figure 4 summarizes Fig. 3, and shows the mean performance curve of the 'fast' visual information processors ( n = l l ; mean age 20.5 years, range 18-23 years; mean span of intercollegiate playing experience 2.3 years, range 1-4 years) and the 'slow' visual information processors (n = l l ; mean age 21.8 years, range 19-25 years; mean span of intercollegiate playing experience 2.8 years, range 1-4 years) averaged over type of sport. The data underlying the performance curves of Fig. 4 were entered in a 2 (groups) x 6 (exposure duration) mixed factor analysis of variance (ANOVA), with repeated measures on the factor exposure duration. This analysis yielded the following significant effects: group (F 1 2 O = 74.6,P
Journal of Sports Sciences Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rjsp20
Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation a
J.J. Adam & R.B. Wilberg
b
a
Department of Movement Sciences , University of Limburg , PO Box 616, Maastricht, 6200‐MD, The Netherlands b
Department of Physical Education , University of Alberta , Edmonton, Alberta, T6G 2H9, Canada Published online: 14 Nov 2007.
To cite this article: J.J. Adam & R.B. Wilberg (1992) Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation, Journal of Sports Sciences, 10:3, 261-273, DOI: 10.1080/02640419208729925 To link to this article: http://dx.doi.org/10.1080/02640419208729925
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Sports Sciences, 1992,10, 261-273
Individual differences in visual information processing rate and the prediction of performance differences in team sports: A preliminary investigation Downloaded by [University of Auckland Library] at 01:17 12 December 2014
J.J. ADAM 1 * and R.B. WILBERG 2 1 Department of Movement Sciences, University of Limburg, PO Box 616, 6200-MD Maastricht, The Netherlands and 2Department of Physical Education, University of Alberta, Edmonton, Alberta T6G 2H9, Canada
Accepted 2 August 1991
Abstract
This study used a backward-masking paradigm to examine individual differences in rate of visual information processing among university basketball, ice hockey and Canadian football players. Displays containing four letters were presented for stimulus durations ranging from 25 to 300 ms. Following stimulus offset, a masking stimulus was presented for 200 ms. The subjects were instructed to write down as many letters as possible from the briefly presented stimulus display on a specially prepared response grid. The results indicated consistent individual differences in rate of visual information processing. More importantly, it was found that rate of visual information processing as indexed by the backward-masking technique, has promising validity for predicting general performance excellence in university ice hockey and basketball players. Individual differences in rate of visual information processing were interpreted as reflecting the operation of attentional factors. Keywords: Individual differences, information processing, skilled performance, team sports, visual perception. Introduction
The ability to process visual information quickly is a factor of crucial importance in many skills, sports and games. Consider, for example, such fast action motor skills as those required when batting in baseball or cricket, goal-keeping in hockey, and punching in boxing. The idea that high-speed visual processing plays a pivotal role in 'open' skills, suggests that performance excellence could be related to a performer's processing capacity and rate. The assumption behind this speculation is that athletes who possess the ability to process more visual information in a shorter period of time than their competitc r s will have an advantage. This advantage can take two forms (Deary and Mitchell, 1989); perceptually quick athletes may instruct their motor effector processes sooner and/or they may provide the decision processes with more detailed information, increasing the chances of an appropriate response. * To whom all correspondence should be addressed. 0264-0414/92
© 1992 E. & F.N. Spon
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
262
Adam and Wilberg
Of specific concern, then, are perceptual processes that come into play when performers are faced with visual displays of very short duration. There is considerable evidence to suggest that people do differ in their ability to process short duration visual displays. This evidence is based on backward-masking studies in which a subject's recognition of a brief target stimulus is impaired by a masking stimulus that follows the initial display. Before discussing this evidence, it seems useful to introduce briefly the concept of short-term visual storage - or iconic memory, as Neisser (1967) termed it since it is regarded by many theorists as a key component in visual information processing theory (but see Haber, 1983). Using the partial report technique, Sperling (1960) demonstrated convincingly that following display offset, subjects have much more information available than they can consciously report. Sperling even suggested that perhaps all of the information in the display is represented in what he called 'visual information storage' and Neisser (1967) later dubbed 'iconic memory'. Even though the nature of the representations in iconic memory is a matter of dispute (for a discussion, see Coltheart, 1980; Kahneman and Treisman, 1984), it is generally agreed that the contents of iconic memory are accessible for further processing for a very short duration (~250 ms), after which they will have decayed away. Following his work on the experimental demonstration of iconic memory, Sperling (1963) went on to investigate the rate at which information from iconic memory could be 'read out' and thus reported. For different stimulus durations, he presented slides containing six letters, followed by a masking stimulus consisting of random black and white squares. Sperling found that the number of letters reported correctly increased regularly up to 75 ms of stimulus duration, at which point four items could be reported. Additional stimulus duration up to 200 ms resulted in one more letter being reported. In other words, letters seem to be initially read out from iconic memory at a high rate, and subsequently at a low rate. Similar results have been reported by other researchers (for a review, see Coltheart, 1972). The rate at which material in iconic memory can be 'read out' varies considerably from one individual to another. Turvey (1973) and Marcel (1983) reported large inter-individual variations in ability to detect or recognize a word following the use of a pattern mask. Di Lollo et al. (1982) and Walsh (1976) found evidence supporting a drop in processing rates as age increases. Galbraith and Gliddon (1972), Mosley (1981) and Baumeister et al. (1984) found support for intelligence-related differences in visual information processing rates among high- and low-IQ subjects. Importantly, Deary and Mitchell (1989) reported that successful batsmen in cricket were faster at picking up information from briefly presented visual displays than less successful batsmen. Besides studies of the ability to process rapid visual displays, there have also been studies of sport-specific or context-dependent perceptual skills. In general terms, the basic finding is that high-level players have superior perceptual skills in situations particular to their sport only. For example, Allard et al. (1980) compared the performance of basketball players and non-players on a task requiring the recall of slides containing structured and unstructured game information, after a 4-s viewing period. They found that basketball players were superior to non-players in recall of structured (context-dependent) slides only. As pointed out by Allard et al. (1980), similar results have been established for the games of chess (De Groot, 1965; Chase and Simon, 1973), Go (Reitman, 1976) and bridge (Charness, 1979). The interaction between the superior perception of skilled athletes and the structured displays of their sport-specific environment suggests that an 'encoding of structure' is important in a performer's excellence (Allard et al, 1980; Allard and Starkes, 1980).
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Visual information processing rate
263
In contrast to the high-speed visual processing paradigms that employ a backwardmasking stimulus, sport-specific perceptual studies generally employ long periods of display availability, such as 4-5 s of viewing time. The difference between these two approaches suggests that the underlying mechanisms may also be different. The ability to extract visual information from displays of short duration may be a general ability, as opposed to the context-dependent perceptual skill that develops as performers gain situation-specific experience. Proof of such a general factor would require an experimentally significant main effect in faster recognition of non-contextually specific stimulus material between experienced athletes who differ in excellence. The approach taken in this study to demonstrate that speed of visual information processing is an important factor in predicting performance excellence in open skills differs in several respects from the interaction paradigm. First, type of stimulus material is not varied systematically, but is held constant and neutral (i.e. letters of the alphabet). Secondly, stimulus duration is not held constant at a relatively long duration, but varied systematically from 25 to 300 ms. Moreover, following stimulus offset, a masking stimulus is employed to control the functional lifetime of the test stimulus. Thirdly, there are no marked differences between the athletes in the subject pool in terms of experience and general skill level, in that they are all university athletes. This latter requirement takes into account the fact that performance success is undoubtedly determined by a variety of factors. If the groups to be compared are not homogeneous in terms of other potentially important factors influencing performance level, the experiment will be confounded and the effect of rate of processing could be 'swamped' by uncontrolled variance. Fourthly, the subjects are differentiated into two groups (i.e. 'top-ranked' vs 'bottom-ranked' groups) by the head coach, according to the criterion of performer's general excellence. Taken together, these experimental manipulations will make it possible to conclude that when a significant main effect in recognition speed is found between 'top-ranked' and 'bottom-ranked' athletes, rate of visual information processing is an important mediating variable in performance excellence in team sports.
Method Subjects A total of 22 male athletes from the University of Alberta served as subjects, 6 of whom were recruited from the university basketball team, 6 from the university football team (Canadian football) and 10 from the university ice hockey team. Selection was random. The basketball players (mean age 20.8 ±1.5 years) averaged 2.7 years of intercollegiate playing experience. The football players (mean age 19.8 + 2.0 years) averaged 2.0 years and the hockey players (mean age 22.2 + 1.8 years) averaged 3.3 years. All of the subjects were unpaid volunteers and had normal or corrected-to-normal vision. Apparatus The subjects stood in a normally lit room and viewed from above a 8.5 x 3.0 cm, horizontally placed, rectangular opaque rear projection glass screen at a distance of about 58 cm. The viewing distance was kept constant by employing a chin support. The test stimulus and masking stimulus were rear-projected tachistoscopically by two identical sets of apparatus,
264
Adam and Wilberg
each consisting of a Kodak slide projector (Ektagraphic), an Opticon Uniblitz electroprogrammable shutter (Model 262) and an Opticon Uniblitz shutter driver and timing unit (Model SD-10). Control of display parameters was established by a PDP-11/10 computer in conjunction with the Uniblitz SD-10 shutter control units.
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Stimuli The target stimuli consisted of 35-mm slides of four different letters arranged in a linear array. The letters for each array were selected randomly from all consonants (except for the letter 'y'), to minimize the possibility of subjects interpreting the arrays as words. A total of 100 target arrays was constructed. Two different sizes of letters were used for the elements within the target arrays. The two outer elements were 1.0 degree in height and 0.8 degree in width, while the two inner elements were 0.5 degree in height and 0.4 degree in width. The inner letters were spaced 0.5 degree apart, while there was a distance of 1.5 degrees between the inner and outer letters. These values were chosen from the guidelines established by Anstis (1974), so as to keep letter resolution above threshold at both retinal locations. The outer letters were printed in uppercase Helvetica Medium script, and the inner letters were printed in upper-case Helvetica Small script. Figure 1 schematically shows a target stimulus.
X
D
N
Fig. 1. A schematic representation of a target stimulus.
The masking stimulus consisted of a pattern of randomly superimposed letters of both sizes, covering the entire field. The target stimuli were presented for six different stimulus durations: 25,50,75,100,150 and 300 ms. Following target offset, the masking stimulus was displayed for 200 ms.
Procedure The subjects were tested individually in a session lasting approximately 20 min. After entering the room, the subjects were informed about the nature of the experimental task. Each subject was then shown a different random sample of 10 target arrays at several stimulus durations to orient them to the task. Following these orienting trials, the subjects received a block of 15 trials for each stimulus duration. The first five trials of each block were considered practice trials, and were not included in the analysis. Each subject received the same order of presentation of the stimulus durations, which started with the longest (300 ms) and ended with the shortest (25 ms). Prior to the experiment, all of the slides were arranged in random order into one block of 10 trials (the orienting trials) and six blocks of 15 trials (the test trials). For each trial, the subjects were instructed to fixate on the centre of the screen and to give a verbal 'ready' signal. Then the experimenter pressed a button, which initiated a warning tone of 50 ms. After a delay of 1 s, the target stimulus appeared, which was followed at its offset by
Downloaded by [University of Auckland Library] at 01:17 12 December 2014
Visual information processing rate
265
the masking stimulus. The subjects were instructed to write down as many letters as possible from the briefly presented array on a specially prepared response grid. While the subject wrote down his response, the experimenter advanced the slide projector by remote control. It was emphasized that the letters had to be reported in the correct location and that guessing was allowed if one was uncertain. After testing was completed, the head coaches of the respective sports were asked to rank all their athletes who had participated in the experiment according to the criterion of'general excellence', with players of equal ability being given tied or equal ranks. Based on these evaluations, 'top-ranked' and 'bottom-ranked' groups were formed for each sport. That is, athletes in the top 30% of the rankings were categorized as 'top-ranked' athletes and athletes in the bottom 30% of the rankings were categorized as 'bottom-ranked' athletes. The athletes in the middle 40% of the rankings, therefore, were not included in the comparison between 'top-ranked' and 'bottom-ranked' athletes.
Results and discussion
Figure 2 displays the principal results of this experiment. The mean number of letters reported in the correct position is shown for each individual in his respective sport as a function of stimulus duration. An analysis of the experimental data based on the more lenient criterion of letters reported correctly, irrespective of correct location, yielded the same overall pattern of results. No inferential statistical analysis was performed on these data. Visual inspection of Fig. 2 reveals important performance differences between athletes on the backward masking task for all three sports. The athletes differed both in rate of information gain and in final amount gained, at least for the stimulus durations employed in the present experiment. The overall shape of the functions relating amount reported to stimulus duration, i.e. an initial steep slope followed by a relatively shallow slope, seems to hold for each individual and resembles closely those reported by other investigators (for a description, see Coltheart, 1972). The subjects within each type of sport were then divided into two groups based upon the total number of letters reported correctly. These median split divisions produced two groups in all three sports: (1) a 'fast' group, consisting of those athletes with the higher total number of letters reported correctly, and (2) a 'slow' group, consisting of those athletes with the lower total number of letters reported correctly. The result of this manipulation of the individual performance curves shown in Fig. 2 resulted in the group curves presented in Fig. 3. Figure 4 summarizes Fig. 3, and shows the mean performance curve of the 'fast' visual information processors ( n = l l ; mean age 20.5 years, range 18-23 years; mean span of intercollegiate playing experience 2.3 years, range 1-4 years) and the 'slow' visual information processors (n = l l ; mean age 21.8 years, range 19-25 years; mean span of intercollegiate playing experience 2.8 years, range 1-4 years) averaged over type of sport. The data underlying the performance curves of Fig. 4 were entered in a 2 (groups) x 6 (exposure duration) mixed factor analysis of variance (ANOVA), with repeated measures on the factor exposure duration. This analysis yielded the following significant effects: group (F 1 2 O = 74.6,P
