Gymnastics is a competitive sport that requires precise motor control, balance, power, and attention during the execution of the motions. Our previous studies of world class gymnasts have shown changes in the topological properties of the brain networks (Wang, et al., 2013; Wang, et al., 2015), and in the gray matter volume (Huang, et al., 2013). Therefore, alterations in cortical thickness may be expected to occur after long-term intensive training. In this study, our goal was to detect the cortical thickness changes in the world class gymnasts using vertex-wise and ROI-wise analysis, and to investigate the correlation between the cortical thickness and the years of training.

Thirteen world class gymnasts (M/F 6/7, aged 17–26 year, mean±std=20.5± 3.2 years) and fourteen age- and gender-matched controls (M/F 7/7, aged 19–28 years, mean ± std = 22.3 ± 2.7 years) were recruited in this study. Written informed consent was obtained from each participant prior to this study. The protocol was approved by the Research Ethics Committee of the Institute of Cognitive Neuroscience and Learning at Beijing Normal University.

MR images were obtained using a Siemens Trio Tim 3T MR scanner. High-resolution brain structural images were acquired using the T1-weighted 3D MP-RAGE sequence (TR/TE/FA=1900 ms/3.44 ms/8°, FOV=256×256mm, matrix=256×256, slice thickness=1mm, and 176 sagittal slices).

The 3D brain structural images were analyzed using FreeSurfer (Dale, Fischl et al. 1999; Fischl, Sereno et al. 1999). First, we reconstructed the cortical surfaces with the following steps: 1) segmentation of the white matter, 2) tessellation of the gray/white matter boundary, 3) inflation of the folded surface tessellation, and 4) automatic correction of topological defects. And the cortical thickness was measured by calculating the shortest distance from the gray/white boundary to the gray/CSF boundary at each vertex. Second, all of the reconstructed cortical surfaces were morphed and registered to an average spherical surface. Then, we resampled the cortical thickness data for each subject. At last, a Gaussian kernel with a full-width half maximum (FWHM) of 10mm was used to smooth the cortical thickness maps.

At each vertex, we used a general linear model (GLM) to detect significant difference in cortical thickness between the gymnasts and the controls. The left and right hemispheres were analyzed separately. A Monte Carlo simulation cluster analysis with 10,000 iterations and a cluster threshold of p < 0.05 was performed to correct for multiple comparisons. Furthermore, we extracted the mean thickness of significant clusters, in which we analyzed the correlation between the thickness and the years of training using Pearson’s correlation coefficient.

In addition, we calculated the mean thickness of the regions of interesting (ROI) determined with the Destrieux atlas (Destrieux, et al., 2010; Fischl, et al., 2004). Each ROI in the Destrieux atlas contains information about either gyral or sulcal structures in the human brain. For the ROI-wise analysis, we utilized a nonparametric permutation test (10,000 times) and then applied false discovery rate (FDR) correction to determine significant between-group differences in the mean cortical thickness for each ROI (p < 0.05, FDR correction).

For the vertex-wise analysis, we found clusters with significantly increased cortical thickness in the left superior parietal, left lateral occipital, and right superior frontal cortex in the gymnasts compared to the controls (Table 1 and Fig. 1). We also detected that the mean thickness was significantly negatively correlated with the years of training in the right superior frontal cortex (r = -0.644, p = 0.018) (Fig. 2).

For the ROI-wise analysis, we found that the ROIs with significantly increased cortical thickness were located in the left orbital gyri, left superior occipital sulcus and transverse occipital sulcus, left inferior part of the precentral sulcus, left subparietal sulcus, and right orbital part of the inferior frontal gyrus (Table 2 and Fig. 3).

Using vertex-wise and ROI-wise analyses, we detected the cortical thickness alterations in world class gymnasts for the first time. The increased cortical thickness in the frontal, parietal, and occipital cortices in gymnasts may provide a structural basis for understanding their abilities of motion, decision making and execution, as well as visuospatial abilities. Thus, this study seems to support the point that acquisition and execution of the complex motor skill with long-term intensive gymnastic training can induce neuroanatomical plasticity in the cortical thickness. Our findings of world class gymnasts may provide useful information to understand the neural mechanism of motor skill acquisition and training in the professional athlete.