Long-Term Motor Training Enhances Functional Connectivity between Semantic and Motor Regions in an Effector-Specific Manner: Evidence from Elite Female Football Athletes

doi:10.21203/rs.3.rs-3953635/v1

The relation between the action verb semantic processing and sensorimotor experience remains controversial. In this study, we examined whether plasticity changes in brain are specifically related to semantic processing of foot action verbs when long-term motor training is mainly aimed at the foot. To address this question, we acquired resting-state functional magnetic resonance imaging scans and behavioral data from a verb two-choice task from female expertise football players and football novices. We compared the resting-state functional connectivity (rsFC) differences between experts and novices using motor execution regions and general semantic regions (left anterior temporal lobe, lATL) as seed, and explored the neural correlates of behavioral performance. Here, the drift rate (v) parameter of the Drift Diffusion Model (DDM) was used to capture the semantic processing capability. We found experts showed increased correlation between lATL subregions and important brain regions for motor processing, including supplementary motor area (SMA), bilateral paracentral lobule (PL), superior parietal lobule (SPL) and inferior parietal lobule (IPL), in contrast to novices. Further predictive model analysis showed the FC found in rsFC analysis can significantly predict drift rate of foot action verb in both experts and novices, but not drift rate of hand action verb. Our findings therefore establish a connection between effector-related semantic processing and the plasticity changes in brain functional connectivity, attributable to long-term foot-related motor training. This provides evidence supporting the view that semantic processing is fundamentally rooted in the sensorimotor system.

semantic processing

resting-state functional connectivity

Drift Diffusion Model

long-term motor training

embodied theory.

The representation of semantics in the human brain remains a topic of controversy. Embodied theories propose that semantic processing is grounded in the sensory-motor system (e.g. Pulvermüller, 2005, 2013; Gallese & Cuccio, 2018). According to embodied theories, the semantic system can undergo plastic changes due to sensorimotor experiences (Xiong et al. 2023; Beilock et al. 2008). That is, short- or long-term sensorimotor experiences can alter the related semantic representations in the brain. For example, following a 15-day period of head-down tilt bed rest, the brain's representations for foot action verbs in subcortical motor regions and the left anterior temporal region were reduced (Xiong et al. 2023). Similarly, Beilock et al. demonstrated that sports motor experience enhances action-related language understanding by recruiting the left dorsal lateral premotor cortex (Beilock et al. 2008). Nonetheless, it remains to be determined whether plasticity changes induced by long-term motor training correlate with semantic processing in an effector-specific manner.

Somatotopic activity during action verb processing supports the embodied theory of semantic processing. In an event-related fMRI study, Hauk et al. (2004) demonstrated that when participants passively read action words related to face, arm, or leg actions, the premotor and frontal areas were activated in a somatotopic manner. More recently, a single neuron recording study indicated that words share a neural substrate with corresponding sensorimotor representations (Aflalo et al., 2020). However, the presence of somatotopic activity during action word processing alone does not constitute evidence in favor of an embodied theory of action processing. The overlap between the activity in the word and the motor localizer conditions can also be predicted by cognitive theories (Caramazza et al., 2014). Moreover, some studies have not observed consistent somatotopic activation in the primary or premotor cortex during the processing of action words (Kemmerer et al. 2008, Kemmerer & Gonzalez-Castillo 2010). Additionally, Argiris et al. (2020) furnished empirical support for the preservation of action verb processing capabilities in patients despite lesions within the sensorimotor cortex. Complementary findings from a repetitive transcranial magnetic stimulation (rTMS) investigation conducted by Solana et al. (2023) indicated that stimulation of the primary motor cortex (M1-rTMS) did not impede the construction of meaning, thereby challenging theories positing the primary motor cortex's role in deriving semantic content from action language. Therefore, evidence from a broader range of perspectives is needed to address the controversy between cognitive and embodied theories of semantic processing.

In this study, we employed the resting-state functional connectivity (rsFC) method. Compared to task-based fMRI, rsFC can be utilized to examine the plasticity changes in the brain due to training (Demirakca et al., 2016; Wang et al., 2016; Raichlen et al., 2016; Vahdat et al., 2011; Ma et al., 2011; Dayan & Cohen, 2011), learning (Lewis et al. 2009), or rehabilitation (Park et al., 2011). For instance, Vahdat demonstrated that motor learning can induce changes in the functional connectivity of resting-state networks related to the sensorimotor system. Similarly, four weeks of motor skill training modified the resting-state network of the motor system. Meanwhile, rsFC patterns are relatively stable over time, which renders them useful for studying individual differences and the effects of long-term training on brain networks (Guo et al., 2012). Moreover, the resting-state connectivity of the left anterior temporal lobe (lATL) - a region considered critical for semantic cognition - has been explored by examining the overlap of brain regions active during rest and task states (Jackson et al., 2016). Previous studies have identified connectivity between the supplementary motor area and language networks in both healthy controls (Lou et al., 2017) and patients (Bathla et al., 2019). Based on and extending previous research, our study test if long-term motor training was correlated to increased functional connectivity between motor processing regions and language neural networks.

Furthermore, evidence of plastic functional connectivity needs to be linked with behavioral performance in semantic tasks to further explore the underlying mechanisms of semantic processing. To isolate the somatotopic activity during action word processing, we decomposed the behavioral performance into several parameters using the Drift Diffusion Model (DDM, Ratcliff, 1978). This model could facilitate the separation of encoding processes for different effector verbs, potentially shedding light on brain plasticity changes following long-term motor training. Previous studies employing the DDM have demonstrated that the drift rate parameter in the model captures processes operating at the level of lexical semantics. For instance, a higher drift rate was observed for words when compared to random letter strings, but no significant difference in drift rate was found between words and word-like non-words (Ratcliff et al., 2004a, b). Recently, in a lexical-semantic study, both patients with temporal and frontal lesions exhibited an increased drift rate for unrelated word-picture matching pairs (Todorova et al., 2020). Soares et al. correlated grey matter volume with DDM parameters including drift rate, non-decision time, and decision boundary. The results indicated that a lifetime of bilingual language processing may change the structure of the brain in the DLPFC (Soares et al., 2019). These studies suggest that the drift rate may reflect lexical-semantic processing and correlate with brain structure, function, and connectivity.

In this fMRI study, we explored whether long-term specialized football training enhances the brain's functionality in semantic processing in an effector-specific manner. We hypothesized the existence of functional connectivity plasticity between general semantic regions and motor areas, including the primary motor cortex, premotor cortex, and supplementary motor area. Furthermore, we anticipated that brain plasticity changes in functional connectivity resulting from long-term motor training would correlate with behavioral performance in effector-related semantic tasks.

Participants

We recruited twenty-eight active professional female football athletes (excluding goalkeepers) from the Shanghai Greenland Shenhua Football Club and twenty-nine female undergraduate students from Shanghai University of Sport for this study. The decision to exclude goalkeepers from our study was based on the unique nature of their training, which is significantly different from that of other football players in terms of both physical requirements and cognitive processes. None of the novices had any experience with football training. The experts and novices were comparable in terms of age (t = 0.273, p = 0.786) and education level. All participants were right-handed with normal or corrected-to-normal vision and had no history of neurological disorders. Two experts and one novice were excluded from the analyses due to excessive movement (one expert and one novice) and low behavioral accuracy (43.17% in the foot verb condition). All experimental protocols were approved by the Shanghai University of Sport, China, and all participants provided written informed consent.

Stimuli and procedure

In order to evaluate differences between experts and novices in the processing of foot and hand action verbs, we implemented a verb-effector two-choice task. The task used fourteen prominent foot action verbs (e.g. “踢“, /ti1/, to kick; “跑”, /pao1/, to run), all containing the same Chinese character component “足”, and fourteen hand action verbs (e.g. “抓”, /zhua1/, to scratch; “握”, /wo1/, to grip) all containing the character component “扌”. The word frequency between foot and hand action verbs was matched (Mann-Whitney U test: Z = -1.378, p = 0.168).

The task consisted of two identical blocks, with each block containing 70 trials per condition. Each trial began with a 400ms fixation display, followed by the stimulus presented at the center of the screen. Participants were asked to silently read the action verb and determine the body part utilized in the action, pressing "g" on the keyboard for foot verbs and "f" for hand verbs. They were encouraged to respond both accurately and quickly. All stimuli were displayed in white [RGB, 255 255 255] against a black background [RGB: 0 0 0]. Stimuli presentation and response collection were facilitated using a PC computer with Eprime-3.0 software (Psychology Software Tools, Pittsburgh, PA, USA) on a screen with a resolution of 1024 * 768, 60Hz.

Functional Imaging

Imaging data were collected using a Siemens Prisma 3T scanner equipped with a 64-channel phase-array head coil, located at the Imaging Center for MRI Research, Shanghai University of Sport. Participants lay supine, with their heads secured using foam pads to minimize head movement. Resting state functional imaging data consisted of 240 continuous whole brain functional volumes, acquired using a gradient echo-planar imaging sequence (repetition time, 2000ms; echo time, 30ms; 58 slices; voxel size, 2 * 2 *2.4mm³; flip angle, 90°). A T1-weighted anatomical MRI was also collected (repetition time, 2530ms; echo time, 2.98ms; 192 slices; voxel size, 1 * 1 * 1 mm³; flip angle, 7°).

Data analysis

Behavioral data analysis

Accuracy and reaction times (RTs) for each experimental condition were recorded. Responses slower than the mean RT plus 2 standard deviations (SD), faster than the mean RT minus 2 SD, and those faster than 200ms were excluded from our analysis.

Statistical analysis was conducted using IBM SPSS Statistics, Version 26.0 (IBM SPSS, Inc., Chicago, IL, USA). A repeated measures analysis of variance (ANOVA) was performed to examine the accuracy and RTs, with group (experts vs. novices) as a between-subjects factor and condition (foot verbs vs. hand verbs) as a within-subjects factor.

Drift diffusion model analysis

The Drift Diffusion Model posits that every binary decision-making task can be characterized as a stochastic process, where evidence is gathered over time until a final decision is made favoring one of two decision boundaries. This model encompasses four key parameters that define the underlying components of the decision-making process. These parameters include the rate at which evidence is collected over time (drift rate, v), the quantity of evidence needed to reach a decision (decision threshold, a), the initial amount of information gathered before the decision-making process initiates (starting point, z), and the time consumed by non-decision processes (non-decision time, t0), such as motor preparation or stimulus encoding. These parameters collectively define the dynamics of the decision-making process within the DDM framework (Voss et al., 2013a,b).

To distinguish various components of task performance, we applied a DDM. Parameters of DDM were estimated using Fast-dm (version 30.2) software (Voss & Voss, 2007, 2008). We employed the KS statistic to measure the degree of correlation between observed cumulative distribution and predicted cumulative distribution. Fast-dm provides estimates of six parameters (a, v, t0, sa, sv, st0), with the exception of the starting point which was fixed midway between the two response barriers, assuming an unbiased decision-making process in our task.

Resting-state functional connectivity data analysis and MRI acquisition

Preprocessing of the resting-state fMRI data was carried out using the SPM12 and the DPABI toolbox (Yan et al., 2016). The preprocessing steps included: (1) discarding the first ten images of each participant to stabilize the magnetic field and ensure that the participants adapted well to the scanner environment, (2) slice time and head motion correction, (3) bet and T1 coreg to fun, (4) regression of nuisance covariates including the Friston-24 motion parameters and 3 other confounding signals(white matters, cerebrospinal fluid and global signals), (4)spatial normalization to the Montreal Neurological Institude space and resampling to 2*2*2mm³, (5)space smoothing using a 4mm full-width-maximum Gaussian kernal, (6)removing the linear trends, (7)temporal band-pass filtering(0.01-0.08Hz).

Then, seed-based voxel-wise rsFC analysis were performed by the DPABI. We defined two types of ROIs: (1) Motor execution regions were 6*6*6 sphere with peak MNI coordinate when did foot or hand motor execution location tasks (Xiong et al. 2023). (2) General semantic system, left ATL, was defined anatomically as a union of the following six anterior temporal regions in the Harvard-Oxford Atlas (probability > 0.25, following Xu et al. 2018): left temporal pole(TP); left anterior superior temporal gyrus(aSTG); left anterior middle temporal gyrus(aMTG); left inferior temporal gyrus(aITG); left anterior parahippocampal gyrus (aPHG); left anterior temporal fusiform cortex(aTFC).

The seed-based voxel-wise rsFC analysis entailed three components: (1) the time courses of all voxels within each seed region were extracted and averaged; (2) the Pearson correlation was calculated between the mean time courses and those of all other voxels over the whole brain (e.g. rsFC maps); (3) all rsFC maps were converted into z-maps using a Fisher’s transformation to improve the normality and enhance the quality of the statistical analysis.

Subsequently, spm12 were used for group-level analyses. GLMs estimating were used to contrast rsFC between expert and novice for each mask. Age and TIV were controlled as covariates of noninterest. Results were threshold voxel-wise at p < 0.001 (uncorrected), cluster-level at p < 0.05 by FWE-corrected and cluster size > 50 voxels.

Predictive model analysis

We used the brain regions that showed differences between experts and novices in the group-level analysis as mask to extract and average rsFC values, multiple linear regression analyzes were conducted to test if the brain regions showed differences in group-level analysis can predict the parameters of DDM for both experts and novices, and further examine the independent contributions of the rsFC in explaining individual variations. Independent variables included rsFC values obtained previously. The dependent variable was the drift rate of foot or hand verb for expert or novice. We then conducted an analysis using the non-decision time as the dependent variable through the same multiple linear regression method.

Post-hoc rsFC analysis

According to the results of rsFC analysis, we found MFG/SMA may be an important node for the connectivity between semantic neural network and motor neural network. To further explore this phenomenon, we conducted a subsequent rsFC analysis using the MFG/SMA brain regions that exhibited group differences in the functional connectivity analysis as seeds. We then implemented one sample t-test to inspect the connectivity pattern originating from the MFG/SMA in both experts and novices. To achieve reliable results, we applied a stringent family-wise error (FWE) multiple correction (p = 0.01) and established a large cluster size (k = 200).

Behavioral results

The accuracy and reaction times (RTs) were subjected to a repeated measures ANOVA with 'group' (expert vs. novice) as the between-subject factor and 'condition' (foot vs. hand verbs) as the within-subject factor. In terms of accuracy, there were no significant main effects for either group (F = 0.073, p = 0.788) or condition (F = 0.645, p = 0.426) the interaction was not significant (F = 0.461, p = 0.500). For RTs, significant main effects were observed for both group (F = 4.693, p = 0.035, η_p² = 0.083) and condition (F = 6.462, p = 0.014, η_p² = 0.111), however, their interaction was not significant. As illustrated in Fig. 1, the multiple t tests showed experts had faster RTs than novices in both foot (t = 2.104, p = 0.040) and hand (t = 2.138, p = 0.037) effector conditions.

Parameters of the drift diffusion model

Results for the DDM parameters are illustrated in Fig. 1. For each parameter, a repeated measures ANOVA was performed with 'group' as the between-subject factor and 'condition' as the within-subject factor. For response criterion, the main effects of group and condition were not significant (group, F = 0.061, p = 0.805; condition, F = 0.001, p = 0.982), and the interaction was not significant (F = 0.148, p = 0.702). For drift rate, the main effects of group and condition were not significant (group, F = 2.249, p = 0.140; condition, F < 0.001, p = 0.993), and the interaction was not significant (F = 0.009, p = 0.926). For non-decision time, the main effects of group and condition were significant (group, F = 8.439, p = 0.005, η_p² = 0.140; condition, F = 8.439, p = 0.005, ηp2 = 0.140), but the interaction was not significant (F = 0.372, p = 0.544). The multiple t tests showed non-decision time for experts was lower than that for novices in both the foot effector condition (t = 2.463, p = 0.017) and hand effector condition (t = 2.894, p = 0.006).

(Fig. 1)

ROI seed-based rsFC

Motor execution areas

We initially conducted a seed-based functional connectivity analysis to examine the FC differences in foot and hand execution areas between experts and novices. For the seed in the foot execution region, we observed an enhanced correlation with the right middle temporal gyrus in expert football players compared to novices. However, when using the hand execution region seed, we detected no significant FC differences between experts and novices.

2. General semantic regions: lATL

We then investigated the FC differences between experts and novices derived from the general semantic regions. Utilizing the TP seed, we noticed decreased correlations with a cluster in the right superior frontal gyrus and a cluster encompassing the right superior temporal gyrus and right middle temporal gyrus in experts compared to novices. However, no significant differences were identified between the two groups when using the aSTG seed. In contrast to novices, experts demonstrated an increased correlation between the aMTG seed and a cluster in the right anterior cingulate, as well as a cluster including the left medial frontal gyrus and bilateral paracentral lobule. Furthermore, an enhanced correlation was detected between aMTG and the right superior frontal gyrus in experts. For the aITG seed, experts showed an increased functional connectivity with a cluster in the left cingulate gyrus and with the right paracentral lobule. For the aPHG seed, experts displayed a reduced functional connectivity with a cluster in the right middle temporal gyrus. Lastly, from the aTFC seed, we noted increased correlations to a cluster comprising the left superior parietal lobule and left precuneus, a cluster including the left supramarginal gyrus and left inferior parietal lobule, and a cluster in the left postcentral gyrus in experts compared to novices.

(Table 1)

(Fig. 2)

Following the correlation analysis, we performed a predictive model analysis to further evaluate whether the functional connectivity demonstrating plasticity changes could predict the drift rate in each group for each condition.

Predictive model analysis

We identified differences in functional connectivity between experts and novices in both the motor execution regions and the lATL area. Subsequently, we tested whether these differing functional connectivities could predict the drift rate of foot and hand verbs in both the expert and novice groups. The results from the multiple linear regression analysis, which are summarized in Table 2, indicate that the rsFC differences between the groups have a strong predictive power for the drift rate of foot verbs in both the expert (adjusted R² = 0.409, F = 2.574, p = 0.049) and novice (adjusted R² = 0.408, F = 2.692, p = 0.035) groups.

The coefficients of the predictive factors are presented in the supplementary materials. For experts, the functional connectivities FC4 (β = 0.759, p = 0.019) and FC8 (β = 0.787, p = 0.002) were positively associated with the drift rate of foot verbs. Although no significant effects were found for the predictive model of the drift rate of hand verbs in both experts and novices, FC8 (β = 0.617, p = 0.014) was positively associated with the drift rate of hand verbs in experts..

(Table 2)

(Fig. 3)

Analysis of the predictive model for t0 indicated that the functional connectivity differences between experts and novices were not able to predict t0 for either foot or hand verbs in either group (experts: foot verb, p = 0.641; hand verb, p = 0.397; novices: foot verb, p = 0.86; hand verb, p = 0.506). These results can be found in Supplementary materials

Post-hoc rsFC analysis

Based on the results from the correlation and predictive model analysis, we found that FC4 was positively correlated with the drift rate of foot verbs, but it did not show a significant correlation with the drift rate of hand verbs. To further explore this phenomenon, we performed an additional rsFC analysis, selecting the MFG/SMA brain regions that showed group differences as seed regions. As presented in Table 3, in experts, the MFG was correlated with the bilateral precentral gyrus, whereas in novices, the MFG showed correlation with the right paracentral gyrus and right precentral gyrus.

(Table 3)

In this study, we suggest plasticity changes of functional connectivity induced by long-term motor training are correlated with semantic processing in an effector-specific manner. Firstly, building upon and expanding the findings of previous studies (Hervais-Adelman et al., 2015; Amad et al., 2017), our results indicate that long-term expertise in football training enhances the functional connectivity between the general semantic system and the motor-related cortex. Secondly, we observed that the plasticity changes in brain functional connectivity correlated with the behavioral performance in semantic tasks. Our findings thus provide further supporting evidence for the notion that semantic processing is rooted in the sensory-motor system.

The results presented in this study bolster the hypothesis that the neural systems instrumental in forming and retrieving semantic knowledge overlap with those essential for perceiving various sensory modalities or executing actions. Our findings suggest that long-term motor training increases the functional connectivity between semantic processing regions and motor processing regions. Specifically, we observed an increased correlation between the foot execution region and the right MTG in expert football athletes compared to novices. The right MTG has been associated with semantic processing (Jung et al., 2017). Furthermore, experts showed an increase in functional connectivity between the left aMTG and the ACC, Supplementary Motor Area (SMA), bilateral Paracentral Lobule (PL), and right medial Superior Frontal Gyrus (SFG). These areas (ACC, SMA, PL, and medial SFG) are intricately involved in motor control and coordination, particularly for complex, voluntary movements and movements involving the lower extremities (Paus, 2001; Wenderoth et al., 2005; Tanji, 2001; La Fougère et al., 2010; Rizzolatti & Luppino, 2001).On the other hand, the left aMTG plays a crucial role in many domains of language processing (Xu et al., 2019; Visser et al., 2012), and its connectivity with primary and secondary sensory-motor areas has been associated with semantic processing (Riccardi et al., 2019). Moreover, the analysis revealed that experts also showed increased functional connectivity between the left aITG and the Mid-Cingulate Cortex (MCC) and PL. From the aTFC seed, experts demonstrated increased functional connectivity with the left Superior Parietal Lobule (SPL), left Inferior Parietal Lobule (IPL), and left Postcentral Gyrus (PoG). The IPL has been associated with the mirror neuron system, thought to be involved in understanding and imitating the actions of others (Rizzolatti & Sinigaglia, 2010). A recent study suggests that the IPL plays a critical role when visuo-motor coordination is required for action (Johari et al., 2022). As shown in Table 1 and Fig. 2, novices showed increased connectivity between TP and right MTG and between aPHG and right MTG. Both of these FCs are within the general semantic system, suggesting that novices may have superior general semantic processing capabilities than football athletes. The advantages of football athletes are reflected in semantic processing of concrete action verbs, especially those related to foot actions. Taken together, our results indicate that long-term foot-related motor training may correlate to plasticity changes in brain functional connectivity between semantic processing and motor processing areas, further enhancing our understanding of the connection between physical training and cognitive functions.

Interestingly, our results also showed that the FCs that showed differences between groups could significantly predict the drift rate of foot verbs in both experts and novices, but not the drift rate of hand verbs. This consistency across groups indicates that FCs exhibiting plastic changes is indeed linked to the semantic processing of foot action verbs. Within our predictive models, we found that FCs between aMTG and MFG/SMA, and between aPHG and right MTG, were significant predictors for the drift rate of foot verbs in experts. However, for the drift rate of hand verbs in experts, only the connectivity between aPHG and right MTG emerged as a significant predictor. These results imply that the FC between aMTG and MFG/SMA might be influenced by motor experience, impacting the semantic understanding of foot action verbs. In contrast, the connectivity between aPHG and right MTG, which remains a constant predictor across verb categories, suggests its pivotal role in the broader comprehension of action language, independent of motor experience. SMA plays a crucial role for linking cognition to action (Nachev et al., 2008). SMA is crucial for motor execution and motor plan (Hardwick et al., 2018), moreover, it is also considered to play a significant role in language processing (Hertrich et al., 2016). The current study discovered that the connectivity between the SMA and MTG brain regions may be modulated by motor experience, and that this connectivity is associated with the semantic processing process, thereby corroborating the findings of previous research. Further rsFC analysis using SMA as seed showed the SMA was correlated with the bilateral precentral gyrus in experts, whereas in novices, the MFG showed correlation with the right paracentral gyrus and right precentral gyrus. The results further support the vies that the SMA is a key node linking language and action. In sum, these findings align with previous studies suggesting that semantic processing of concrete action verbs is based, at least in part, on motor experience (Fernandino et al., 2022). They further emphasize the crucial roles of FC between aMTG and MFG/SMA and FC between aPHG and MTG in embodied semantic processing (Hauk et al., 2004; Courson et al., 2017; Zhang et al., 2020; Fu et al., 2023).

Notably, we did not find a difference in the drift rate between groups or conditions, likely due to the task's difficulty. However, we found significant differences in response time and non-decision time between groups. Non-decision time, which encompasses the period for encoding and response generation processes (Ratcliff et al., 2016), presents a dual explanation for these findings: either experts possess a superior motor processing ability, or they have an advantage in encoding efficiency. The analysis from our predictive models focusing on non-decision time indicated that FC variations between groups failed to significantly predict the drift rate for foot action verbs among both experts and novices. This suggests that the observed advantages among experts may not stem from motor processing capabilities.

In summary, this study adopted a hypothesis-driven approach to assess the impacts of long-term motor training on effector-specific action verb semantic processing in expert football players. Additionally, it sought to identify potential underlying mechanisms that might elucidate these observed effects. Our findings provide a novel perspective, suggesting that long-term motor training can enhance semantic processing in an effector-specific manner. This enhancement appears to be mediated by induced plasticity changes in the functional connectivity between the motor and semantic systems. This study supports the embodied view of semantic processing. Future study could repeat current study by male and female participants to explore if it is a general phenomenon in both genders, and could further determine the embodied semantic processing using fMRI decoding analysis.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Jian Wang, Qingcan Zhou and Yanzhang Chen. The first draft of the manuscript was written by Jian Wang and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Shanghai University of sport (Date 13.09.2022/No.102772022RT082).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Acknowledgments

We would like to express our deepest gratitude to our colleagues and mentors at Shanghai University of Sport for their continuous support, insightful comments, and hard questions, all of which have significantly contributed to the improvement of this research work.

Aflalo, T., Zhang, C. Y., Rosario, E. R., Pouratian, N., Orban, G. A., & Andersen, R. A. (2020). A shared neural substrate for action verbs and observed actions in human posterior parietal cortex. Science Advances, 6(43), eabb3984. https://doi.org/10.1126/sciadv.abb3984
Amad, A., Seidman, J., Draper, S. B., Bruchhage, M. M. K., Lowry, R. G., Wheeler, J., Robertson, A., Williams, S. C. R., & Smith, M. S. (2017). Motor Learning Induces Plasticity in the Resting Brain—Drumming Up a Connection. Cerebral Cortex, 27(3), 2010–2021. https://doi.org/10.1093/cercor/bhw048
Argiris, G., Budai, R., Maieron, M., Ius, T., Skrap, M., & Tomasino, B. (2020). Neurosurgical lesions to sensorimotor cortex do not impair action verb processing. Scientific Reports, 10(1), 523. https://doi.org/10.1038/s41598-019-57361-3
Bathla, G., Gene, M. N., Peck, K. K., Jenabi, M., Tabar, V., & Holodny, A. I. (2019). Resting State Functional Connectivity of the Supplementary Motor Area to Motor and Language Networks in Patients with Brain Tumors. Journal of Neuroimaging, 29(4), 521–526. https://doi.org/10.1111/jon.12624
Beilock SL, Lyons IM, Mattarella-Micke A, Nusbaum HC, Small SL. (2008). Sports experience changes the neural processing of action language. Proc Natl Acad Sci. 105(36):13269–13273
Caramazza, A., Anzellotti, S., Strnad, L., & Lingnau, A. (2014). Embodied Cognition and Mirror Neurons: A Critical Assessment. Annual Review of Neuroscience, 37(1), 1–15. https://doi.org/10.1146/annurev-neuro-071013-013950
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Erlbaum.
Courson, M., Macoir, J., & Tremblay, P. (2017). Role of medial premotor areas in action language processing in relation to motor skills. Cortex, 95, 77–91. https://doi.org/10.1016/j.cortex.2017.08.002
Dayan, E., & Cohen, L. G. (2011). Neuroplasticity Subserving Motor Skill Learning. Neuron, 72(3), 443–454. https://doi.org/10.1016/j.neuron.2011.10.008
Demirakca, T., Cardinale, V., Dehn, S., Ruf, M., & Ende, G. (2016). The Exercising Brain: Changes in Functional Connectivity Induced by an Integrated Multimodal Cognitive and Whole-Body Coordination Training. Neural Plasticity, 2016, 1–11. https://doi.org/10.1155/2016/8240894
Di, X., Zhu, S., Jin, H., Wang, P., Ye, Z., Zhou, K., ... & Rao, H. (2012). Altered resting brain function and structure in professional badminton players. Brain connectivity, 2(4), 225-233.
Fernandino, L., Tong, J.-Q., Conant, L. L., Humphries, C. J., & Binder, J. R. (2022). Decoding the information structure underlying the neural representation of concepts. Proceedings of the National Academy of Sciences, 119(6), e2108091119. https://doi.org/10.1073/pnas.2108091119
Fu, Z., Wang, X., Wang, X., Yang, H., Wang, J., Wei, T., Liao, X., Liu, Z., Chen, H., & Bi, Y. (2023). Different computational relations in language are captured by distinct brain systems. Cerebral Cortex, 33(4), 997–1013. https://doi.org/10.1093/cercor/bhac117
Gallese, V., & Cuccio, V. (2018). The neural exploitation hypothesis and its implications for an embodied approach to language and cognition: Insights from the study of action verbs processing and motor disorders in Parkinson’s disease. Cortex, 100, 215–225. https://doi.org/10.1016/j.cortex.2018.01.010
Gardini, S., Venneri, A., Sambataro, F., Cuetos, F., Fasano, F., Marchi, M., Crisi, G., & Caffarra, P. (2015). Increased Functional Connectivity in the Default Mode Network in Mild Cognitive Impairment: A Maladaptive Compensatory Mechanism Associated with Poor Semantic Memory Performance. Journal of Alzheimer’s Disease, 45(2), 457–470. https://doi.org/10.3233/JAD-142547
Guo CC, Kurth F, Zhou J, Mayer EA, Eickhoff SB, Kramer JH, Seeley WW. (2013) One-year test–retest reliability of intrinsic connectivity network fMRI in older adults. NeuroImage. 1;61(4):1471-83.
Han, Y., Yang, H., Lv, Y.-T., Zhu, C.-Z., He, Y., Tang, H.-H., Gong, Q.-Y., Luo, Y.-J., Zang, Y.-F., & Dong, Q. (2009). Gray matter density and white matter integrity in pianists’ brain: A combined structural and diffusion tensor MRI study. Neuroscience Letters.
Hardwick, R. M., Caspers, S., Eickhoff, S. B., & Swinnen, S. P. (2018). Neural correlates of action: Comparing meta-analyses of imagery, observation, and execution. Neuroscience & Biobehavioral Reviews, 94, 31–44. https://doi.org/10.1016/j.neubiorev.2018.08.003
Hauk, O., Johnsrude, I., & Pulvermüller, F. (2004). Somatotopic representation of action words in human motor and premotor cortex. Neuron, 41(2), 301-307.
Hervais-Adelman, A., Moser-Mercer, B., & Golestani, N. (2015). Brain functional plasticity associated with the emergence of expertise in extreme language control. NeuroImage, 114, 264–274. https://doi.org/10.1016/j.neuroimage.2015.03.072
Hertrich, I., Dietrich, S., & Ackermann, H. (2016). The role of the supplementary motor area for speech and language processing. Neuroscience & Biobehavioral Reviews, 68, 602–610. https://doi.org/10.1016/j.neubiorev.2016.06.030
Jackson, R. L., Hoffman, P., Pobric, G., & Lambon Ralph, M. A. (2016). The Semantic Network at Work and Rest: Differential Connectivity of Anterior Temporal Lobe Subregions. Journal of Neuroscience, 36(5), 1490–1501. https://doi.org/10.1523/JNEUROSCI.2999-15.2016
Jeannerod, M. (2001). Neural Simulation of Action: A Unifying Mechanism for Motor Cognition. NeuroImage, 14(1), S103–S109. https://doi.org/10.1006/nimg.2001.0832
Johari, K., Riccardi, N., Malyutina, S., Modi, M., & Desai, R. H. (2022). HD-tDCS of primary and higher-order motor cortex affects action word processing. Frontiers in Human Neuroscience, 16. https://www.frontiersin.org/articles/10.3389/fnhum.2022.959455
Johnson, P. B., Ferraina, S., Bianchi, L., & Caminiti, R. (1996). Cortical Networks for Visual Reaching: Physiological and Anatomical Organization of Frontal and Parietal Lobe Arm Regions. Cerebral Cortex, 6(2), 102–119. https://doi.org/10.1093/cercor/6.2.102
Jung, J., Cloutman, L. L., Binney, R. J., & Lambon Ralph, M. A. (2017). The structural connectivity of higher order association cortices reflects human functional brain networks. Cortex, 97, 221–239. https://doi.org/10.1016/j.cortex.2016.08.011
Kemmerer, D., Castillo, J. G., Talavage, T., Patterson, S., & Wiley, C. (2008). Neuroanatomical distribution of five semantic components of verbs: Evidence from fMRI. Brain and Language, 107(1), 16–43. https://doi.org/10.1016/j.bandl.2007.09.003
Kemmerer, D., & Gonzalez-Castillo, J. (2010). The Two-Level Theory of verb meaning: An approach to integrating the semantics of action with the mirror neuron system. Brain and Language, 112(1), 54–76. https://doi.org/10.1016/j.bandl.2008.09.010
La Fougère, C., Zwergal, A., Rominger, A., Förster, S., Fesl, G., Dieterich, M., Brandt, T., Strupp, M., Bartenstein, P., & Jahn, K. (2010). Real versus imagined locomotion: A [18F]-FDG PET-fMRI comparison. NeuroImage, 50(4), 1589–1598. https://doi.org/10.1016/j.neuroimage.2009.12.060
Lewis CM, Baldassarre A, Committeri G, Romani GL, Corbetta M. Learning sculpts the spontaneous activity of the resting human brain. Proceedings of the National Academy of Sciences. 2009 Oct 20;106(42):17558-63.
Lou W, Peck KK, Brennan N et al (2017) Left-lateralization of resting state functional connectivity between the presupplementary motor area and primary language areas. Neuroreport 28:545–550. https ://doi.org/10.1097/WNR.0000000000000783
Ma L, Narayana S, Robin DA, Fox PT, Xiong J. Changes occur in resting state network of motor system during 4 weeks of motor skill learning. NeuroImage. 2011 Aug 1;58(1):226-33.
Nachev, P., Kennard, C., & Husain, M. (2008). Functional role of the supplementary and pre-supplementary motor areas. Nature Reviews Neuroscience, 9(11), Article 11. https://doi.org/10.1038/nrn2478
Park, I.S., Lee, K.J., Han, J.W., Lee, N.J., Lee, W.T., Park, K.A., and Rhyu, I.J. (2009). Experience-dependent plasticity of cerebellar vermis in basketball players. Cerebellum 8, 334–339.
Park CH, Chang WH, Ohn SH, Kim ST, Bang OY, Pascual-Leone A, Kim YH. Longitudinal changes of resting-state functional connectivity during motor recovery after stroke. Stroke. 2011 May;42(5):1357-62.
Paus, T. (2001). Primate anterior cingulate cortex: Where motor control, drive and cognition interface. Nature Reviews Neuroscience, 2(6), 417–424. https://doi.org/10.1038/35077500
Pulvermüller, F. (2005). Brain mechanisms linking language and action. Nature reviews neuroscience, 6(7), 576-582.
Pulvermüller F. 2013. How neurons make meaning: brain mechanisms for embodied and abstract-symbolic semantics. Trends Cognit Sci. 17:458–470.
Ratcliff, R. (1978). A theory of memory retrieval. Psychological review, 85(2), 59.
Ratcliff, R., Gomez, P., & McKoon, G. (2004). A diffusion model account of the lexical decision task. Psychological review, 111(1), 159.
Ratcliff, R., Thapar, A., Gomez, P., & McKoon, G. (2004). A diffusion model analysis of the effects of aging in the lexical-decision task. Psychology and aging, 19(2), 278.
Ratcliff, R., & McKoon, G. (2008). The diffusion decision model: theory and data for two-choice decision tasks. Neural computation, 20(4), 873-922.
Ratcliff, R., Smith, P. L., Brown, S. D., & McKoon, G. (2016). Diffusion Decision Model: Current Issues and History. Trends in Cognitive Sciences, 20(4), 260–281. https://doi.org/10.1016/j.tics.2016.01.007
Raichlen, D. A., Bharadwaj, P. K., Fitzhugh, M. C., Haws, K. A., Torre, G.-A., Trouard, T. P., & Alexander, G. E. (2016). Differences in Resting State Functional Connectivity between Young Adult Endurance Athletes and Healthy Controls. Frontiers in Human Neuroscience, 10. https://doi.org/10.3389/fnhum.2016.00610
Riccardi, N., Yourganov, G., Rorden, C., Fridriksson, J., & Desai, R. H. (2019). Dissociating action and abstract verb comprehension post-stroke. Cortex, 120, 131–146. https://doi.org/10.1016/j.cortex.2019.05.013
Rizzolatti, G., & Luppino, G. (2001). The Cortical Motor System. Neuron, 31(6), 889–901. https://doi.org/10.1016/S0896-6273(01)00423-8
Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the parieto-frontal mirror circuit: Interpretations and misinterpretations. Nature Reviews Neuroscience, 11(4), 264–274. https://doi.org/10.1038/nrn2805
Soares, S. M. P., Ong, G., Abutalebi, J., Del Maschio, N., Sewell, D., & Weekes, B. (2019). A diffusion model approach to analyzing performance on the Flanker task: The role of the DLPFC. Bilingualism: Language and Cognition, 22(5), 1194–1208. https://doi.org/10.1017/S1366728918000974
Solana, P., Casasanto, D., Chica, A. B., & Santiago, J. (2023). No support for a causal role of primary motor cortex in construing meaning from language: An rTMS study [Preprint]. PsyArXiv. https://doi.org/10.31234/osf.io/bnyqt
Tanji, J. (2001). Sequential Organization of Multiple Movements: Involvement of Cortical Motor Areas. Annual Review of Neuroscience, 24(1), 631–651. https://doi.org/10.1146/annurev.neuro.24.1.631
Todorova, L., Neville, D. A., & Piai, V. (2020). Lexical-semantic and executive deficits revealed by computational modelling: A drift diffusion model perspective. Neuropsychologia, 146, 107560. https://doi.org/10.1016/j.neuropsychologia.2020.107560
Vahdat S, Darainy M, Milner TE, Ostry DJ. Functionally specific changes in resting-state sensorimotor networks after motor learning. Journal of Neuroscience. 2011 Nov 16;31(46):16907-15.
Visser, M., Jefferies, E., Embleton, K. V., & Lambon Ralph, M. A. (2012). Both the Middle Temporal Gyrus and the Ventral Anterior Temporal Area Are Crucial for Multimodal Semantic Processing: Distortion-corrected fMRI Evidence for a Double Gradient of Information Convergence in the Temporal Lobes. Journal of Cognitive Neuroscience, 24(8), 1766–1778. https://doi.org/10.1162/jocn_a_00244
Voss, A., & Voss, J. (2007). Fast-dm: A free program for efficient diffusion model analysis. Behavior Research Methods, 39(4), 767–775. https://doi.org/10.3758/BF03192967
Voss, A., & Voss, J. (2008). A fast numerical algorithm for the estimation of diffusion model parameters. Journal of Mathematical Psychology, 52(1), 1-9.
Voss, A., Nagler, M., & Lerche, V. (2013). Diffusion Models in Experimental Psychology: A Practical Introduction. Experimental Psychology, 60(6), 385–402. https://doi.org/10.1027/1618-3169/a000218
Wang, J., Lu, M., Fan, Y., Wen, X., Zhang, R., Wang, B., Ma, Q., Song, Z., He, Y., Wang, J., & Huang, R. (2016). Exploring brain functional plasticity in world class gymnasts: A network analysis. Brain Structure and Function, 221(7), 3503–3519. https://doi.org/10.1007/s00429-015-1116-6
Wenderoth, N., Debaere, F., Sunaert, S., & Swinnen, S. P. (2005). The role of anterior cingulate cortex and precuneus in the coordination of motor behaviour. European Journal of Neuroscience, 22(1), 235–246. https://doi.org/10.1111/j.1460-9568.2005.04176.x
Xiong, Z., Tian, Y., Wang, X., Wei, K., & Bi, Y. (2023). Gravity matters for the neural representations of action semantics. Cerebral Cortex, bhad006. https://doi.org/10.1093/cercor/bhad006
Xu, J., Lyu, H., Li, T., Xu, Z., Fu, X., Jia, F., Wang, J., & Hu, Q. (2019). Delineating functional segregations of the human middle temporal gyrus with resting‐state functional connectivity and coactivation patterns. Human Brain Mapping, 40(18), 5159–5171. https://doi.org/10.1002/hbm.24763
Yan, C.-G., Wang, X.-D., Zuo, X.-N., & Zang, Y.-F. (2016). DPABI: Data Processing & Analysis for (Resting-State) Brain Imaging. Neuroinformatics, 14(3), 339–351. https://doi.org/10.1007/s12021-016-9299-4
Zhang, X., Yang, J., Wang, R., & Li, P. (2020). A neuroimaging study of semantic representation in first and second languages. Language, Cognition and Neuroscience, 35, 1–16. https://doi.org/10.1080/23273798.2020.1738509

Table 1

**Between-group differences in the rsFC from motor execution area and lATL.** Note: rsFC, resting-state functional connectivity; lATL, left anterion temporal lobe; MNI, Montreal Neurological Institute; TP, temporal pole; aMTG, anterior middle temporal gyrus; aITG, anterior inferior temporal gyrus; aPHG, anterior parahippocampal gyrus; aTFC, anterior temporal fusiform cortex.
Predominant regions in cluster	Cluster size	Peak T-value	Cluster-level	MNI coordinates
Predominant regions in cluster	Cluster size	Peak T-value	P_FWE−corr	x	y	z
Functional connectivity from the foot motor execution area
Contrast: Experts > Novices
FC1: Right middle temporal gyrus	55	4.108	0.042	58	-40	-22
Functional connectivity from the TP
Contrast: Novices > Experts
FC2: Right superior temporal gyrus	188	-6.070	< 0.001	52	-48	8
Right middle temporal gyrus				62	-62	10
Functional connectivity from the aMTG
Contrast: Experts > Novices
FC3: Right anterior cingulate	64	6.533	0.019	2	14	-6
FC4: Left medial frontal gyrus	129	5.761	< 0.001	-2	-18	58
Left paracentral lobule				-6	-20	50
Right paracentral lobule				6	-20	48
FC5: Right superior frontal gyrus	61	4.735	0.024	10	-10	68
Functional connectivity from the aITG
Contrast: Experts > Novices
FC6: Left cingulate gyrus	67	5.136	0.013	-8	-6	42
FC7: Right paracentral lobule	53	4.724	0.043	2	-18	48
				0	-32	52
Functional connectivity from the aPHG
Contrast: Novices > Experts
FC8: Right middle temporal gyrus	60	-4.45	0.021	56	-58	12
				54	-52	4
Functional connectivity from the aTFC
Contrast: Experts > Novices
FC9: Left superior parietal lobule	86	4.572	0.002	-30	-68	44
Left precuneus				-32	-70	36
FC10: Left supramarginal gyrus	71	4.943	0.007	-40	-42	38
Left inferior parietal lobule				-40	-48	44
FC11: Left postcentral gyrus	51	4.640	0.043	-46	-34	50

Table 2

Model summary of linear regression models with rsFC.
Models	R²	Adjusted R²	Std.	F	Sig.
Experts
Drift rate of foot verbs	0.669	0.409	0.767	2.574	0.049
Drift rate of hand verbs	0.602	0.290	1.030	1.926	0.124
Novices
Drift rate of foot verbs	0.649	0.408	0.791	2.692	0.035
Drift rate of hand verbs	0.532	0.283	1.099	0.573	0.824

Note: R², the amount of the variance explained by the identified features. Adjusted R², a modification of the R-squared that adjusts for the number of predictors in a model (Cohen, 2003).

Table 3

Between-group differences in the rsFC from MFG.
Predominant regions in cluster	Cluster size	Peak T-value	Cluster-level	MNI coordinates
Predominant regions in cluster	Cluster size	Peak T-value	P_FWE−corr	x	y	z
Experts
Left medial frontal gyrus	4454	21.084	< 0.001	-2	-18	56
Right precentral gyrus				24	-26	56
Left precentral gyrus				-20	-28	68
Novices
Left medial frontal gyrus	7034	26.046	< 0.001	-2	-18	56
Right paracentral lobule				4	-32	52
Right superior frontal gyrus				15	-14	70
Note: MFG: medial frontal gyrus. FEW, p = 0.01, cluster size = 200

No competing interests reported.

SupplementraryMaterial.docx

Long-Term Motor Training Enhances Functional Connectivity between Semantic and Motor Regions in an Effector-Specific Manner: Evidence from Elite Female Football Athletes

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and Methods

Participants

Stimuli and procedure

Functional Imaging

Data analysis

Behavioral data analysis

Drift diffusion model analysis

Resting-state functional connectivity data analysis and MRI acquisition

Predictive model analysis

Post-hoc rsFC analysis

Results

Behavioral results

Parameters of the drift diffusion model

Predictive model analysis

Post-hoc rsFC analysis

Discussion

Conclusion

Declarations

Funding

Competing Interests

Author contributions

DATA AVAILABILITY STATEMENT

Ethics approval

Consent to participate

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1