Rugby is one of the most tactically complex sports. According to rugby coaching theory, players are required to deal with a changing environment and to make appropriate decisions during rugby games (18). Specifically, rugby players need to learn the specific motions of rugby from older members to predict the motions of the ellipsoidal ball and the players, to recognize player formations on the field, to identify friends and foes in both the central and peripheral visual fields, and to decide to attack or not; furthermore, they must perform these actions simultaneously during rugby games (25). To perform such actions, various cognitive abilities such as motor learning, motion prediction, visuospatial cognitive ability, inhibition, and simultaneous operations would be necessary for rugby players.
Although rugby coaching theory claims that such cognitive processes are essential, only one study has addressed the cognitive functioning of rugby players. Kasahara et al. (17) revealed that the Wechsler Adult Intelligence Scale Revised Block Design scores of top-level rugby players were better than those of a standardized sample. The scores of the Block Design subtest are related to performance on the Everyday Spatial Activity Test and the Road Map Test (11), which involves tracing routes indicated by dotted lines on a paper road map with a pencil and determining the orientation from a "bird's eye view," without rotating the map (20). From these findings, Kasahara et al. (17) proposed that rugby players have a higher visuospatial cognitive ability because they use a specific strategy that involves taking a bird's eye view. On the basis of previous findings about rugby players, we assumed that a specific brain activation pattern underlying such visuospatial cognitive processing in rugby players would be observable using a functional neuroimaging method.
The purpose of the current study was to examine if there were differential cortical networks related to visuospatial processing among top-level rugby players and control novices. To this end, we compared brain activities during a visuospatial processing task between top-level rugby players and novices using functional magnetic resonance imaging (fMRI). To avoid the possibility that subjects in the two groups would exhibit differential behavioral performance, which would affect brain activation patterns, in the present study, we recruited novices whose visuospatial ability was expected to match that of the rugby players. We hypothesized that top-level rugby players process visuospatial information with equal aptitude but use different cortical networks to do so. In line with the hypothesis, we expected that the differential patterns of brain activation would be apparent despite the absence of differences in the behavioral performance of rugby players and novices.
To detect brain activation associated with visuospatial cognitive processing, we used a 3-D mental rotation task. Within cognitive psychology, visuospatial cognitive processes have often been evaluated by using mental rotation tasks (7,23,28,34). In addition, accuracy scores for the 3-D mental rotation task have been shown to be positively correlated with scores on the Block Design subtest (24), which was used in the study by Kasahara et al. (17). Because these findings indicate that this task and the Block Design subtest can estimate the same visuospatial ability, we used the 3-D mental rotation task for the present study. This task has also been used in functional neuroimaging studies to localize brain regions associated with visuospatial cognitive processing, such as the superior parietal lobe (SPL) and the lateral occipital cortices (LOC) (6,27,38). On the basis of the previously mentioned studies, we expected that differential activation in the SPL and the LOC would be detected between the top-level rugby players and the novices.
Twenty male rugby players from the Akita Northern Bullets (ANB) Rugby Club (the Top group) and 20 novice males (the Novice or control group) participated in this study. No significant difference in age was found between the Top (mean ± SD = 28.2 ± 5.3 yr, range = 24-40 yr) and Novice groups (mean ± SD = 26.8 ± 5.8 yr, range = 20-41 yr). The ANB is ranked in the top 3% of the 1815 teams registered with the Japan Rugby Football Union, so we regarded the ANB as a top-level rugby team. Subjects in the Novice group were matched for age, sex, handedness, educational history, and visuospatial cognitive ability with subjects in the Top group. To control visuo-spatial cognitive ability, the subjects in the Novice group were primarily recruited from the Tohoku University community, a high-level university community in Japan. We expected that the intelligence quotients (IQ) of the Novice group would be higher than those of the standard population and that the scores on the Block Design subtest, representing visuospatial cognitive ability, would also be higher than those of the standard population and comparable to those of the rugby players. We verified that all subjects in the Novice group had no history of playing rugby with an original questionnaire concerning athletic activity. Through the questionnaire, it was also verified that all subjects in the Novice group had a history of doing sports; however, all of them participated only in club activities, which were not elite-level sports. In addition, most of them had finished their sport careers. Therefore, we regarded all subjects in the Novice group as control novices. Potential participants were excluded if they were left-handed, had full-scale IQ (FIQ) scores <80, reported a history of head trauma, had a current or past diagnosis of a neurological or psychiatric disorder, currently used psychoactive medication, or had any condition not compatible with MRI use (e.g., metal in the body). Handedness was evaluated using the Edinburgh Handedness Inventory (21). From these exclusion criteria, four rugby players were excluded. Results and participant characteristics are shown in Table 1. Each participant provided written informed consent before participation. The study was approved by the institutional review board of the Tohoku University School of Medicine.
Experimental paradigm: the 3-D mental rotation task.
All fMRI experiments in this study were designed using the block design fMRI method. To measure visuospatial cognitive processing, we adopted a 3-D mental rotation task during MRI scanning. Stimuli for this task consisted of perspective drawings of 3-D objects (Fig. 1), modified from a classical experiment by Shepard and Metzler (28). The stimuli were visually presented through a projector and were back-projected onto a screen within a 5° visual angle from the central fixation point (Fig. 1) so that the eye movements were controlled within the central visual field across all subjects. In this task, subjects were shown 30 pairs of 3-D drawings, which were presented in five blocks of six pairs, and judged whether the drawings were mirror images or the same images by pressing a button. In each trial, subjects viewed a pair of 3-D drawings and made a judgment within 3 s. After an interval of 500 ms, during which time subjects were asked to focus on a central fixation point, the next pair of drawings appeared. Reaction times were recorded in each trial, and accuracy rates on the task were calculated for each subject. Each block lasted 21 s. A central fixation point lasting 14 s was shown between blocks as a baseline condition.
All subjects completed the Wechsler Adult Intelligence Scale-Third Edition (WAIS-III) test battery (35), administered by trained clinical psychologists, to measure IQ. Four subjects in the Novice group did not complete the WAIS-III test for technical reasons, but they completed subtests for calculating performance IQ, including the Block Design subtest. To confirm the relationship between visuospatial cognitive ability and behavioral performance on the 3-D mental rotation task, we performed a correlation analysis between the scores on the Block Design subtest and the accuracy rates on the mental rotation task for each subject.
fMRI scanning and data analysis.
All MRI data acquisition was conducted using a 3-T Philips Achieva scanner. Participants viewed the stimuli via a mirror attached to the head coil of the MRI scanner. Behavioral responses were recorded using a two-button fiber-optic response box (Current Designs, Philadelphia, PA). Scanner noise was reduced with earplugs and head motion was minimized using foam pads and a headband. Functional images were acquired using echo-planar images sensitive to blood oxygenation level-dependent contrast (64 × 64 matrix, TR = 2500 ms, TE = 30 ms, flip angle = 80°, field of view (FOV) = 192 mm, 42 slices, 3-mm slice thickness). The preprocessing and statistical analyses for all images were performed using SPM5 (Wellcome Department of Cognitive Neurology, London, UK) implemented in MATLAB (www.mathworks.com/). In the preprocessing analysis, images were corrected for head motion, spatially normalized to the Montreal Neurological Institute space using an echo-planar template, and spatially smoothed using a Gaussian kernel of 8 mm full width at half maximum. No subject data had excessive head motion (>1.5 mm). Thus, data from all subjects (16 top-level rugby players and 20 control novices) were analyzed.
A conventional two-step approach (10) for block design fMRI data was adopted using SPM5. In the first step, the parameter estimates for each individual subject were detected by a multiple regression analysis of the time course data of each voxel, with a boxcar function representing the task condition blocks against the resting conditions. The onset of the boxcar function was set to the beginning of each 3-D mental rotation task. In the second step, statistical inference was performed on contrasts of parameter estimates from the first step analysis for the between-subject (random effects) model using a two-sample t-test on the conventional subtraction analysis to examine differential activation maps between the Top and the Novice groups (contrasts of Top > Novice and Novice > Top). In addition, to investigate the independence of differential brain activation in the two groups and the effect of the visuospatial cognitive ability, we also performed an ANCOVA using the Block Design subtest scores of the WAIS-III from each subject as covariates. The statistical threshold was set at P < 0.001 (uncorrected for multiple comparisons) with a minimum cluster size of 10 voxels for whole-brain analysis. This height threshold is typical for neuroimaging studies of cognitive and/or physical expertise, such as acupuncture physicians (5), expert meditators (3), and ski jumpers (22). We also used a more lenient threshold of P < 0.005 (uncorrected for multiple comparisons) with a minimum cluster size of 10 voxels within the SPL and the LOC on the basis of the a priori hypothesis from the review of the mental rotation tasks (38). The SPL and the LOC were defined using the Wake Forest University PickAtlas (http://www.fmri.wfubmc.edu) and the Anatomical Automatic Labeling Region of Interest package (32).
Psychological and behavioral results.
The FIQ and WAIS-III Block Design subtest scores showed no significant difference between the two groups (Table 1). No statistical difference in accuracy rates or reaction times on the 3-D mental rotation tasks was observed between the two groups (Table 2), as we expected. A significant positive correlation between the accuracy rates on the 3-D mental rotation task and the Block Design subtest scores was found for all subjects (r = 0.47, P < 0.005, Fig. 2), replicating the previous study (24).
Statistically greater activation in the Top group than in the Novice group (contrast of Top > Novice) was observed in the left middle temporal gyrus (MTG) from whole-brain analysis and in the right SPL and right LOC from the ROI analysis (Fig. 3A, Table 3). The activation profiles of these regions showed that activity in the right SPL and right LOC were significantly enhanced compared with the baseline condition in the Top group, whereas activity in the left MTG was not significantly enhanced in the Top group but was reduced in the Novice group (Fig. 3A). Statistically greater deactivation in the Top group than in the Novice group (contrast of Top < Novice) was observed in the right medial prefrontal cortex (MPFC; Fig. 3B, Table 3) from the whole-brain analysis. The activation profiles of the right MPFC showed that activity in this region was reduced compared with the baseline condition in the Novice group but was reduced more in the Top group (Fig. 3B). From the ANCOVA, we replicated similar differential activation patterns with those from conventional subtraction analyses. Statistically significant differential activation between the Top and the Novice groups was observed in the right SPL, the right LOC, and the left MTG (contrast of Top > Novice) and in the right MPFC (contrast of Top < Novice), respectively.
To examine if there were differential cortical networks related to a visuospatial processing among top-level rugby players and control novices, we compared brain activities during a 3-D mental rotation task between the Top and Novice groups. We found differential brain activation between the Top and Novice groups in the SPL and the LOC from the ROI analysis and between the MTG and the MPFC from the whole-brain analysis. As expected, no significant difference was found between the Top and Novice groups for psychological and behavioral outputs, such as the Block Design subtest of the WAIS-III, reaction times, and accuracy rates on the 3-D mental rotation task (Tables 1 and 2).
The discrepancy between the psychobehavioral outputs and the fMRI results was interpreted as an indication that differential strategies exist between the two groups. The absence of differences in the psychobehavioral outputs implies that the differences in brain activation could not be attributed to behavioral performance. This inconsistency suggested that the different patterns of brain activation were caused by the differences in strategy. This interpretation was congruent with previous studies reporting that the use of a different strategy between groups is reflected in different patterns of brain activation, despite the absence of differences in behavioral performance (13,15,29). Even if no statistically significant difference was found, it might be argued that the scores on the Block Design subtest in our Novice group were relatively high compared with those of the Top group. This difference would confound differences in fMRI data. The response to such criticism lies in the results of the ANCOVA using the scores on the Block Design subtest as covariates. The results demonstrated that differential activations between the two groups were independent of visuospatial cognitive ability. The fact that the activation patterns from conventional subtraction analyses were replicated by ANCOVA suggested that the differential activation patterns between the two groups were robust.
Next, we discuss the nature of the difference in strategy between the two groups. We interpreted our fMRI results to be an indication that the Top group took a bird's eye view during the 3-D mental rotation task. The SPL has been reported to be associated with the spatial integration of visual features (36) and with mental imagery (19). The LOC is a part of the extrastriate visual cortex (30,33), which has been reported to be related to global motion processing (2). In addition, these two regions are among the brain areas related to spatial transformation to a third-person perspective (16). These cognitive features may allow one to take another viewpoint and keep up with the global motion of multiple objects, that is, to take a bird's eye view. In fact, this strategy may allow rugby players to look down on the field and to imagine the players' formation and the motion of the rugby game. The results support the hypothesis that rugby players use a specific strategy that involves taking a bird's eye view, as was proposed by Kasahara et al. (17).
It is necessary to clarify the interpretation of quantitative differential activation in the right SPL and the right LOC in the two groups. According to the activation profiles (Fig. 3A), these regions were significantly activated compared with the baseline condition in both top-level rugby players and in novices. Thus, differential brain activation did not support the interpretation of a qualitatively different pattern of brain activation but rather supported an interpretation of a quantitative difference between the two groups. The interpretation of the quantitative differential activation is still controversial. Two possible interpretations for the increased activation are available. One interpretation was that increased quantitative activation represents the recruitment of a strategy that demands greater use of the cognitive functions of those activated brain areas. In previous studies of sex-specific strategies in visuospatial processing, the increased quantitative activation of particular brain areas was interpreted as evidence for the existence of a particular strategy demanding the greater use of cognitive functions in the brain area in question (13,15,29). The other interpretation was that increased quantitative activation represents the lack of an ability related to the cognitive function of a brain area. This interpretation is based on the effect of habituation, practice, and learning of a particular task (4,12,14), which improves behavioral outputs and reduces brain activation in nature. Therefore, expertise in a familiar task is associated with reduced quantitative activation in the task-relevant brain areas (1,5) and vice versa. In the current study, the subjects of both groups were unfamiliar with the 3-D mental rotation task. This task was difficult for the subjects of both groups because the accuracies of both groups were barely 60%, despite the chance level being 50%. In addition, there was no significant difference in the behavioral outputs of both groups. These behavioral outcomes suggest that although the latter interpretation is not relevant to our case, the former interpretation is appropriate for supporting the increased quantitative activation of our results. Consequently, the Top group used the SPL and the LOC whose functions are related to the strategy of taking a bird's eye view during the difficult task.
It is unclear why we found deactivation in the right MPFC. It is possible that deactivation of this region supports the existence of a specific strategy among the Top group that differed from that among the Novice group during the 3-D mental rotation task. In previous neuroimaging studies, the MPFC has been reported to be part of the default mode network, which is deactivated when cognitive load is high but activated in a resting state (9,26). Given the previous findings, the greater deactivation of the right MPFC in the Top group compared with the Novice group (Fig. 3B) suggests two alternative interpretations. One is that the top-level rugby players were experiencing a higher cognitive load than the novices during the 3-D mental rotation task. The other is that the top-level rugby players experienced lower cognitive load than the novices during the rest condition. In either case, deactivation in the MPFC could indicate a difference in strategy between the two groups. Although deactivation in the MPFC was not sufficient as direct evidence for our hypothesis, our standpoint was based on the hypothesis that top-level rugby players have a specific strategy (17). From this standpoint, the results could suggest that the additional quantitative cognitive load of the top-level rugby players might be related to that strategy. If this suggestion is correct, the findings may support our claim that differential strategies between the two groups exist.
Although deactivation in the left MTG could potentially reflect different strategies, we could not identify any specific strategy used by the novices. From the activation profile, the neural response in the left MTG was not significantly activated in the Top group but was deactivated in the Novice group. Because the MTG is thought to be associated with lexical processing (37) and meaningful action observation (8) and because it neighbors the visual area V5/middle temporal area that is related to motion recognition (31), deactivation in this region might indicate a strategy related to these functions. However, the current study was designed to investigate the strategy of the top-level rugby players, not of the novices. In addition, no qualitative difference was found, as is reflected in differential brain areas, which did not show any particular activation only in the novices when compared with the baseline condition. Thus, the neural response in the left MTG of the novices cannot be discussed meaningfully. Therefore, we cannot specify a strategy among the novices that differed from that among the rugby players.
In the present study, we found quantitative differences in brain activations during visuospatial processing rather than qualitative ones. The results provide evidence for the existence of a strategy regarding visuospatial cognitive processing for top-level rugby players that differs from that among novices. Furthermore, the results imply a bird's-eye-view strategy. However, from this study, we could not identify whether this strategy is unique to rugby players or is more generally characteristic of physically elite players of team sports. To address this issue, further studies of physically elite players in other team sports that require such a strategy for playing, such as soccer, football, basketball, ice hockey, baseball, and so on, are necessary. In addition, a longitudinal study would be the most effective means of clarifying whether the differential neural bases of physically elite players are derived from inherent talent or are the result of the training, practice, and play of specific sports. We hope that the current study can open up entirely new areas of research that use neuroimaging methods to measure the aptitude of potential physically elite players or to validate the effectiveness of sport skill training.
The authors thank Ms. Y. Yamada for operating the MRI scanner; the participants, including the ANB Rugby Club, for their contribution to the study; and our colleagues at the Institute of Development, Aging and Cancer, Tohoku University, for their support.
Part of this work was carried out under the Cooperative Research Project Program of the Institute of Development, Aging and Cancer, Tohoku University, and Japan Science and Technology/Research Institute of Science and Technology for Society and Japan Science and Technology/Core Research for Evolutional Science and Technology.
This study was also supported by a grant-in-aid (grant 21934018) to S.K. from the Ministry of Education, Culture, Sports, Science and Technology, a grant to S.K. for project study from the Fukushima Medical University, and a grant to S.K. for young investigators from the Fukushima ψ21 Plan Group.
There are no conflicts of interest to declare.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Berkowitz AL, Ansari D. Expertise-related deactivation of the right temporoparietal junction during musical improvisation. Neuroimage
2. Braddick OJ, O'Brien JM, Wattam-Bell J, Atkinson J, Hartley T, Turner R. Brain areas sensitive to coherent visual motion. Perception
3. Brefczynski-Lewis JA, Lutz A, Schaefer HS, Levinson DB, Davidson RJ. Neural correlates of attentional expertise in long-term meditation practitioners. Proc Natl Acad Sci U S A
4. Breiter HC, Etcoff NL, Whalen PJ, et al. Response and habituation of the human amygdala during visual processing of facial expression. Neuron
5. Cheng Y, Lin CP, Liu HL, et al. Expertise modulates the perception of pain in others. Curr Biol
6. Cohen MS, Kosslyn SM, Breiter HC, et al. Changes in cortical activity during mental rotation. A mapping study using functional MRI. Brain
7. Crucian GP, Barrett AM, Burks DW, et al. Mental object rotation in Parkinson's disease. J Int Neuropsychol Soc
8. Decety J, Grèzes J, Costes N, et al. Brain activity during observation of actions. Influence of action content and subject's strategy. Brain
. 1997;120 (Pt 10):1763-77.
9. Fox MD, Raichle ME. Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat Rev Neurosci
10. Friston KJ, Frith CD, Frackowiak RS, Turner R. Characterizing dynamic brain responses with fMRI: a multivariate approach. Neuroimage
11. Groth-Marnat G, Teal M. Block design as a measure of everyday spatial ability: a study of ecological validity. Percept Mot Skills
12. Henson RN. Neuroimaging studies of priming. Prog Neurobiol
13. Hugdahl K, Thomsen T, Ersland L. Sex differences in visuo-spatial processing: an fMRI study of mental rotation. Neuropsychologia
14. Jansma JM, Ramsey NF, Slagter HA, Kahn RS. Functional anatomical correlates of controlled and automatic processing. J Cogn Neurosci
15. Jordan K, Wüstenberg T, Heinze HJ, Peters M, Jäncke L. Women and men exhibit different cortical activation patterns during mental rotation tasks. Neuropsychologia
16. Kaiser S, Walther S, Nennig E, et al. Gender-specific strategy use and neural correlates in a spatial perspective taking task. Neuropsychologia
17. Kasahara S, Mashiko H, Niwa SI. Superior performance in WAIS-R block design among top-level rugby players. Br J Sports Med
19. Mellet E, Petit L, Mazoyer B, Denis M, Tzourio N. Reopening the mental imagery debate: lessons from functional anatomy. Neuroimage
20. Money J, Alexander D, Walker HT. A Standardized Road-Map Test of Direction Sense; Manual
. Baltimore (MD): Johns Hopkins Press; 1965. p. 32.
21. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia
22. Olsson CJ, Jonsson B, Larsson A, Nyberg L. Motor representations and practice affect brain systems underlying imagery: an FMRI study of internal imagery in novices and active high jumpers. Open Neuroimag J
23. Ozel S, Larue J, Molinaro C. Relation between sport activity and mental rotation: comparison of three groups of subjects. Percept Mot Skills
. 2002;95(3 pt 2):1141-54.
24. Parsons TD, Larson P, Kratz K, et al. Sex differences in mental rotation and spatial rotation in a virtual environment. Neuropsychologia
25. Passos P, Araujo D, Davids K, Shuttleworth R. Manipulating constraints to train decision making in rugby union. Int J Sports Sci Coach
26. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proc Natl Acad Sci U S A
27. Richter W, Ugurbil K, Georgopoulos A, Kim SG. Time-resolved fMRI of mental rotation. Neuroreport
28. Shepard RN, Metzler J. Mental rotation of three-dimensional objects. Science
29. Thomsen T, Hugdahl K, Ersland L, et al. Functional magnetic resonance imaging (fMRI) study of sex differences in a mental rotation task. Med Sci Monit
30. Tootell RB, Mendola JD, Hadjikhani NK, et al. Functional analysis of V3A and related areas in human visual cortex. J Neurosci
31. Tootell RB, Reppas JB, Dale AM, et al. Visual motion aftereffect in human cortical area MT revealed by functional magnetic resonance imaging. Nature
32. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage
33. Van Essen D, Orbach HS. Optical mapping of activity in primate visual cortex. Nature
34. Vandenberg SG, Kuse AR. Mental rotations, a group test of three-dimensional spatial visualization. Percept Mot Skills
35. Wechsler D. Technical Manual for the Wechsler Adult Intelligence Scale-Third Edition
. Tokyo (Japan): Nihon Bunka Kagakusha; 2006. 217 p. Japanese.
36. Wilkinson DT, Halligan PW, Henson RN, Dolan RJ. The effects of interdistracter similarity on search processes in superior parietal cortex. Neuroimage
37. Yokoyama S, Miyamoto T, Riera J, et al. Cortical mechanisms involved in the processing of verbs: an fMRI study. J Cogn Neurosci
38. Zacks JM. Neuroimaging studies of mental rotation: a meta-analysis and review. J Cogn Neurosci
Keywords:©2011The American College of Sports Medicine
RUGBY PLAYERS; VISUOSPATIAL PROCESSING; MENTAL ROTATION; fMRI; BIRD'S EYE VIEW