Unveiling individuality in the early phase of motor learning: a machine learning approach for analysing weightlifting technique in novices.

Introduction: Despite the growing body of evidence highlighting the individuality in movement techniques, predominant models of motor learning, particularly during the acquisition phase, continue to emphasise generalised, person-independent approaches. Biomechanical studies, coupled with machine learning approaches, have demonstrated the uniqueness of movement techniques exhibited by individuals. However, this evidence predominantly pertains to already stabilised movement techniques, particularly evident in cyclic daily activities such as walking, running, or cycling, as well as in expert-level sports movements. This study aims to evaluate the hypothesis of individuality in whole-body movements necessitating intricate coordination and strength among novice participants at the very beginning of an acquisition phase. Methods: In a within-subject design, sixteen highly active male participants (mean age: 23.1 ± 2.1 years), all absolute novices in the learning task (i.e., power snatch of Olympic weightlifting), participated in randomised snatch learning bouts. These bouts comprised 36 trials across various motor learning models: differential learning contextual interference (serial, sCIL; and blocked, bCIL), and repetitive learning. Kinematic and kinetic data were collected from three standardised snatch trials performed following each motor learning model bout. The time-continuous data were input to a linear Support Vector Machine (SVM). We conducted analyses on two classification tasks: participant and motor learning model. Results: The Support Vector Machine classification revealed a notably superior participant classification compared to the motor learning model classification, with an averaged prediction accuracy of 78% (in average ≈35 out of 45 test trials across the folds) versus 27.3% (in average ≈9 out of 36 test trials across the folds). In specific fold and input combinations, accuracies of 91% versus 38% were respectively achieved. Discussion: Methodically, the crucial role of selecting appropriate data pre-processing methods and identifying the optimal combinations of SVM data inputs is discussed in the context of future research. Our findings provide initial support for a dominance of individuality over motor learning models in movement techniques during the early phase of acquisition in Olympic weightlifting power snatch.

Mechanism for High-Precision Control of Movement at Maximum Output in the Vertical Jump Task.

Human movements are governed by a tradeoff between speed and accuracy. Previous studies that have investigated the tradeoff relationship in sports movements involving whole-body movements have been limited to examining the relationship from the perspective of competition-specific movements, and the findings on whether the relationship is valid have not been unified. Therefore, this study incorporated a vertical jump task with the introduction of a condition in which landing position control was added to evaluate the essence of a sports movement that requires both speed and accuracy. Accuracy was examined using a method that quantifies the coordinates of the landing and takeoff positions using entropy. The mechanism of that tradeoff was then examined by confirming the phenomenon and analyzing the 3D vector trajectories. An increase in accuracy and a decrease in speed were observed when the landing position was the control target, even in the vertical jumping task normally performed at maximum effort, and the 3D velocity vector was characterized by the following: a reduced scalar and a more vertical direction. While the entropy from the takeoff to the landing position seemed to decrease when the accuracy of the landing position improved, the following noteworthy results were obtained given the characteristics of the vertical jump. Unlike traditional feedback control in the entropy reduction in hand movements, the trajectory is predetermined in a feedforward-like manner by controlling the initial velocity vector at takeoff, which allows the landing point to be adjusted.

Integration or separation? Spatial and temporal representations of whole-body movements in visual working memory.

Spatial and temporal information are two major feature dimensions of human movements. How these two types of information are represented in working memory-whether as integrated units or as individual features-influences how much information might be retained and how the retained information might be manipulated. In this study, we investigated how spatial (path/trajectory) and temporal (speed/rhythm) information of complex whole-body movements are represented in working memory under a more ecologically valid condition wherein the spatiotemporal continuity of movement sequences was considered. We found that the spatial and temporal information are not automatically integrated but share the storage capacity and compete for a common pool of cognitive resources. The finding rejects the strong form of object-based representation and supports the partial independence of spatial and temporal processing. Nevertheless, we also found that contextual factors, such as the way movements are organized and displayed, can further modulate the level of object-based representation and spatiotemporal integration.

Effects of spatiotemporal (dis)continuity on working memory for human movements.

Human movements are dynamic and continuous in nature. However, how the spatiotemporal continuity influences working memory for movements is still unclear. Specifically, spatiotemporal continuity of movements may facilitate integrative processing ("integration") and enhance memory performance by optimizing the encoding process, but it may also diminish memory benefits from distinctive processing ("separation"). In this study, we manipulated the continuity state (continuous/discontinuous) (Experiment 1) and its predictability (Experiment 2) of whole-body movement sequences and tested participants' working memory for observed movements with a single-probe recognition task. We formulated potential influence from spatiotemporal (dis)continuity by two opposite forces - integration vs. separation, and demonstrated a conflict between these two processes across space and time. Moreover, we found that the seemingly stimulus-driven perceptual effects from spatiotemporal (dis)continuity might be supported by a prediction-based mechanism, which guided the selection of an optimal processing strategy. Overall, our finding illustrates an interweaving relationship between spatial and temporal processing during action observation and highlights the importance of considering the dynamic and continuous nature of human movements in visual perception and working memory research.

Limb Preference and Skill Level Dependence During the Imagery of a Whole-Body Movement: A Functional Near Infrared Spectroscopy Study.

In the past years motor imagery (MI) turned out to be also an innovative and effective tool for motor learning and improvement of sports performance. Whereas many studies investigating sports MI focusing on upper or lower limbs involvement, knowledge about involved neural structures during whole-body movements is still limited. In the present study we investigated brain activity of climbers during a kinesthetic motor imagery (KMI) climbing task with different difficulties by means of functional near infrared spectroscopy (fNIRS). Twenty healthy participants were split into two groups according to their climbing skill level. The aim of the current study is investigating neural correlates of a whole-body sports MI task with an additional focus on skill level dependency. Climbing experts and non-experts imagined bouldering an "easy" and "difficult" route from a first-person perspective while hemodynamic responses were recorded simultaneously. We found significant differences between the two climbing routes, easy and difficult within participants as well as between the two groups of different climbing skill levels. Overall beginners showed increased hemodynamic responses compared to experts in all defined regions of interest (ROI) supporting the claim of the neural efficiency hypothesis (NEH). Even though climbing is a complex, coordinated movement of upper and lower limbs we found a stronger activation focus of the upper limbs, especially of the dominant hand-area, while the foot area seems to be deactivated or inhibited simultaneously. Summarizing, these findings provide novel insights into brain activation during the imagery of a whole-body movement and its relation to climbing expertise.

Attention modulates incidental memory encoding of human movements.

Attention has been shown to enhance the processing of task-relevant information while suppressing the processing of task-irrelevant information. However, it is less clear whether this attentional modulation exists when there is an intrinsic dependence between task-relevant and task-irrelevant information, such as the dependence of temporal processing on spatial information. In this study, we used complex whole-body movement sequences to investigate the extent to which the task-irrelevant spatial information (trajectory) is processed when only the temporal information (rhythm) is in focus. Moreover, we examined, if the task-irrelevant spatial information is "co-selected" with the target temporal information as predicted by the intrinsic spatiotemporal dependence, whether task-driven attention that is actively directed to spatial information provides extra benefits. Through a two-phase experiment (an incidental encoding phase followed by a surprise memory test phase), we found that the task-irrelevant spatial information was not only perceived but also encoded in memory, providing further evidence in support of a relatively automatic co-selection of spatial information in temporal processing. Nevertheless, we also found that movements whose trajectories were intentionally attended to during the encoding phase were recognized better in the test phase than those that were not, indicating a further modulation from attention on incidental memory encoding and information processing.

Comparison of whole-body sensorimotor skill learning between strength athletes, endurance athletes and healthy sedentary adults.

Motor sequences represent an integral part of human motor ability. Apart from simple movement sequences, complex coordinated movement sequences are the building blocks for peak athletic performance. Accordingly, optimized temporal and spatial coordination of muscle action across multiple limbs may be a distinguishing feature between athletes and non-athletes in many sports. In the present study, we aimed to assess differences between strength and endurance athletes and non-athletes during learning of a complex whole-body serial reaction time task (CWB-SRTT). For this purpose, 26 nonathletes (NAG) and 25 athletes (AG) learned the CWB-SRTT over 2 days separated by 7 days. Mean response times of participants were recorded and statistically analyzed for sequence-specific and non-sequence-specific improvements, as well as differences in learning rates and retention. Furthermore, AG was subdivided into strength (SG) and endurance (EG) athletes, and all analysis steps were repeated. Our results show a better mean response time of AG compared to NAG. However, we could not detect differences in sequence-specific or non-sequence-specific learning, as well as different retention rates between NAG and AG or SG and EG. We assume here that a potential lack of motor transfer between general athletic abilities and the specific complex motor sequence mainly accounts for our findings.

Cerebellar Transcranial Direct Current Stimulation Improves Maximum Isometric Force Production during Isometric Barbell Squats.

Maximum contraction force (MVC) is an important predictor of athletic performance as well as physical fitness throughout life. Many everyday life activities involve multi-joint or whole-body movements that are determined in part through optimized muscle strength. Transcranial direct current stimulation (tDCS) has been reported to enhance muscle strength parameters in single-joint movements after its application to motor cortical areas, although tDCS effects on MIVC in compound movements remain to be investigated. Here, we tested whether anodal tDCS and/or sham stimulation over primary motor cortex (M1) and cerebellum (CB) improves maximum isometric contraction force (MIVC) during isometric barbell squats (iBS). Our results provide novel evidence that CB stimulation enhances MIVC during iBS. Although this indicates that parameters relating to muscle strength can be modulated through anodal tDCS of the cerebellum, our results serve as an initial reference point and need to be extended. Therefore, further studies are necessary to expand knowledge in this area of research through the inclusion of different tDCS paradigms, for example investigating dynamic barbell squats, as well as testing other whole-body movements.

Characterizing cortical hemodynamic changes during climbing and its relation to climbing expertise.

Bouldering is a special form of climbing without rope that requires coordinated whole-body movements. While physical performance parameters such as condition have been well studied, the knowledge on neural activity during climbing still remains sparse. Functional near-infrared spectroscopy (fNIRS) allows to measure brain activation while performing sportive actions due to its relative robustness against motion artefacts. In the current study, hemodynamic response alterations of 13 advanced climbers were investigated during boulder performance using fNIRS measurements. Simple and moderate climbing routes were compared regarding their level of cortical activation mainly in the sensorimotor area. Our results show that repetitively climbing a set of boulders activates almost all areas of the sensorimotor system including the bilateral premotor and supplementary motor cortex, bilateral primary motor cortex as well as the bilateral gyrus supramarginalis and somatosensory cortex. This result was found in both simple and moderate climbing routes with no effect of task complexity on the level of cortical activity. Correlation analysis (uncorrected for multiple comparisons) revealed a negative association between the level of expertise and the hemodynamic response in the supplementary-motor region, suggesting that gaining expertise in climbing is associated with a decrease in secondary motor areas, which is an indicator of motor automaticity. In summary, the present study provides first proof of concept that fNIRS is capable of assessing hemodynamic response alterations within the human motor system during the execution of complex whole-body climbing movements.

Sensorimotor Strategies in Individuals With Poststroke Hemiparesis When Standing Up Without Vision.

This study investigated the sensorimotor strategies for dynamic balance control in individuals with stroke by restricting sensory input that might influence task accomplishment. Sit-to-stand movements were performed with restricted vision by participants with hemiparesis and healthy controls. The authors evaluated the variability in the position of participants' center of mass and velocity, and the center-of-pressure position, in each orthogonal direction at the lift-off point. When vision was restricted, the variability in the mediolateral center-of-pressure position decreased significantly in individuals with hemiparesis, but not in healthy controls. Participants with hemiparesis adopted strategies that explicitly differed from those used by healthy individuals. Variability may be decreased in the direction that most requires accuracy. Individuals with hemiparesis have been reported to have asymmetrical balance deficits, and that meant they had to prioritize mediolateral motion control to prevent falling. This study suggests that individuals with hemiparesis adopt strategies appropriate to their characteristics.

Distinct functional roles of the mirror neuron system and the mentalizing system.

Movements can inform us about what people are doing and also about how they feel. This phenomenologically evident distinction has been suggested to correspond functionally with differential neural correlates denoted as mirror neuron system (MNS) and mentalizing system (MENT). To separate out the roles of the underlying systems we presented identical stimuli under different task demands: character animations showing everyday activities (mopping, sweeping) performed in different moods (angry, happy). Thirty-two participants were undergoing functional magnetic resonance imaging (fMRI) while asked to identify either the performed movement or the displayed mood. Univariate GLM analysis revealed the expected activation of either in MNS or MENT depending on the task. A complementary multivariate pattern-learning analysis based on the "social brain atlas" confirmed the expected recruitment of both systems. In conclusion, both approaches converge onto clearly distinct functional roles of both social neural networks in a novel dynamic social perception paradigm.

A reduced muscle model and planar musculoskeletal model fit for the simulation of whole-body movements.

Musculoskeletal models are made to reflect the capacities of the human body in general, and often a specific subject in particular. It remains challenging to both model the musculoskeletal system and then fit the modelled muscles to a specific human subject. We present a reduced muscle model, a planar musculoskeletal model, and a fitting method that can be used to find a feasible set of active and passive muscle parameters for a specific subject. At a minimum, the fitting method requires inverse dynamics data of the subject, a scalar estimate of the peak activation reached during the movement, and a plausible initial estimate for the strength and flexibility of that subject. While additional data can be used to result in a more accurate fit, this data is not required for the method solve for a feasible fit. The minimal input requirements of the proposed fitting method make it well suited for subjects who cannot undergo a maximum voluntary contraction trial, or for whom recording electromyographic data is not possible. To evaluate the model and fitting method we adjust the musculoskeletal model so that it can perform an experimentally recorded stoop-lift of a 15 kg box.

Changes in brain activity during action observation and motor imagery: Their relationship with motor learning.

Many studies have demonstrated that training utilizing action observation and/or motor imagery improves motor performance. These two techniques are widely used in sports and in the rehabilitation of movement-related disorders. Motor imagery has also been used for brain-machine/computer interfaces (BMI/BCI). During both action observation and motor imagery, motor-related regions such as the premotor cortex and inferior parietal lobule are activated. This is common to actual execution and are involved with the underlying mechanisms of motor learning without execution. Since it is easier to record brain activity during action observation and motor imagery than that during actual sport movements, action observation, and motor imagery of sports skills or complex whole body movements have been utilized to investigate how neural mechanisms differ across the performance spectrum ranging from beginner to expert. However, brain activity during action observation and motor imagery is influenced by task complexity (i.e., simple vs complex movements). Furthermore, temporal changes in brain activity during actual execution along the long time course of motor learning are likely nonlinear and would be different from that during action observation or motor imagery. Activity in motor-related regions during action observation and motor imagery is typically greater in experts than in nonexperts, while the activity during actual execution is often smaller in experts than in nonexperts.

Influence of gait mode and body orientation on following a walking avatar.

Regulating distance with a moving object or person is a key component of human movement and of skillful interpersonal coordination. The current set of experiments aimed to assess the role of gait mode and body orientation on distance regulation using a cyclical locomotor tracking task in which participants followed a virtual leader. In the first experiment, participants moved in the backward-forward direction while the body orientation of the virtual leader was manipulated (i.e., facing towards, or away from the follower), hence imposing an incongruence in gait mode between leader and follower. Distance regulation was spatially less accurate when followers walked backwards. Additionally, a clear trade-off was found between spatial leader-follower accuracy and temporal synchrony. Any perceptual effects were overshadowed by the effect of one's gait mode. In the second experiment we examined lateral following. The results suggested that lateral following was also constrained strongly by perceptual information presented by the leader. Together, these findings demonstrated how locomotor tracking depends on gait mode, but also on the body orientation of whoever is being followed.