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.

Modeling biological individuality using machine learning: A study on human gait.

Human gait is a complex and unique biological process that can offer valuable insights into an individual's health and well-being. In this work, we leverage a machine learning-based approach to model individual gait signatures and identify factors contributing to inter-individual variability in gait patterns. We provide a comprehensive analysis of gait individuality by (1) demonstrating the uniqueness of gait signatures in a large-scale dataset and (2) highlighting the gait characteristics that are most distinctive to each individual. We utilized the data from three publicly available datasets comprising 5368 bilateral ground reaction force recordings during level overground walking from 671 distinct healthy individuals. Our results show that individuals can be identified with a prediction accuracy of 99.3% by using the bilateral signals of all three ground reaction force components, with only 10 out of 1342 recordings in our test data being misclassified. This indicates that the combination of bilateral ground reaction force signals with all three components provides a more comprehensive and accurate representation of an individual's gait signature. The highest accuracy was achieved by (linear) Support Vector Machines (99.3%), followed by Random Forests (98.7%), Convolutional Neural Networks (95.8%), and Decision Trees (82.8%). The proposed approach provides a powerful tool to better understand biological individuality and has potential applications in personalized healthcare, clinical diagnosis, and therapeutic interventions.

Comparing the Effects of Differential and Visuo-Motor Training on Functional Performance, Biomechanical, and Psychological Factors in Athletes after ACL Reconstruction: A Randomized Controlled Trial.

Variation during practice is widely accepted to be advantageous for motor learning and is, therefore, a valuable strategy to effectively reduce high-risk landing mechanics and prevent primary anterior cruciate ligament (ACL) injury. Few attempts have examined the specific effects of variable training in athletes who have undergone ACL reconstruction. Thereby, it is still unclear to what extent the variations in different sensor areas lead to different effects. Accordingly, we compared the effects of versatile movement variations (DL) with variations of movements with emphasis on disrupting visual information (VMT) in athletes who had undergone ACL reconstruction. Forty-five interceptive sports athletes after ACL reconstruction were randomly allocated to a DL group (n = 15), VT group (n = 15), or control group (n = 15). The primary outcome was functional performance (Triple Hop Test). The secondary outcomes included dynamic balance (Star Excursion Balance Test (SEBT)), biomechanics during single-leg drop-landing task hip flexion (HF), knee flexion (KF), ankle dorsiflexion (AD), knee valgus (KV), and vertical ground reaction force (VGRF), and kinesiophobia (Tampa Scale of Kinesiophobia (TSK)) assessed before and after the 8 weeks of interventions. Data were analyzed by means of 3 × 2 repeated measures ANOVA followed by post hoc comparison (Bonferroni) at the significance level of p ≤ 0.05. Significant group × time interaction effects, main effect of time, and main effect of group were found for the triple hop test and all eight directions, SEBT, HF, KF, AD, KV, VGRF, and TSK. There was no significant main effect of group in the HF and triple hop test. Additionally, significant differences in the triple hop test and the seven directions of SEBT, HF, KF, KV, VGRF, and TSK were found between the control group and the DL and VMT groups. Between group differences in AD and the medial direction of SEBT were not significant. Additionally, there were no significant differences between VMT and the control group in the triple hop test and HF variables. Both motor learning (DL and VMT) programs improved outcomes in patients after ACL reconstruction. The findings suggest that DL and VMT training programs lead to comparable improvements in rehabilitation.

Resonance Effects in Variable Practice for Handball, Basketball, and Volleyball Skills: A Study on Contextual Interference and Differential Learning.

Effective sports training should be attuned to the athlete's specific conditionings and characteristics. In motor learning research, two often neglected factors that influence this resonance are the learner's athletic background and the structural diversity of exercises (e.g., relative similarity). In the setting of real-word training with higher external validity, this study examines the effects of three learning approaches (i.e., contextual interference (CI), differential learning (DL), and free-play control condition (CO)) on the parallel learning of handball (HB), volleyball (VB), and basketball (BB) skills, considering participants' prior sport backgrounds. Forty-five males (15 HB, 15 VB, and 15 BB players) with a mean age of 22 ± 1.4 years and at least 6 years of experience in the mastered discipline voluntarily participated in this study. A pre-post-retention test design including a 6-week-intervention program was employed. During the intervention period, participants engaged in three training sessions a week, with each one lasting approximately 80 min. Each of the three test sessions involved the execution of ten attempts of BB free-throw shooting, HB three-step goal throwing, and VB underarm passing following a blocked order. In terms of short-term (pre-post) gain, only the DL group significantly improved their performance in both non-mastered disciplines (p = 0.03, ES = 1.58 for the BB free-throw and p = 0.05, ES = 0.9 for the HB shooting tests), with a trend (ES = 0.53) towards an improvement in the performance of the mastered VB underarm-pass skill. In terms of relatively permanent gains, the CI group significantly improved their performances from pre- to retention test only in the non-mastered BB free-throw skill (p = 0.018, ES = 1.17). In contrast, the DL group significantly improved their performance at retention compared to the pre-test in both non-mastered BB (p = 0.004, ES = 1.65) and HB (p = 0.003, ES = 2.15) skills, with a trend (ES = 0.4) towards improvement in the mastered VB test. In both the short-term and relatively long-term, higher composite score gains were observed in DL compared to CI (p = 0.006, ES = 1.11 and 0.049, ES = 1.01) and CO (p = 0.001, ES = 1.73 and <0.0001, ES = 2.67). In conclusion, the present findings provide additional support for the potential advantages of the DL model over those of CI. These findings can serve as the basis for tailored training and intervention strategies and provide a new perspective for addressing various issues related to individual and situational learning.

Gait Abnormalities in Parkinson's Disease Are Associated with Extracellular Free-Water Characteristics in the Substantia Nigra.

BACKGROUND: Gait impairments are common in Parkinson's disease (PD). The pathological mechanisms are complex and not thoroughly elucidated, thus quantitative and objective parameters that closely relate to gait characteristics are critically needed to improve the diagnostic assessments and monitor disease progression. The substantia nigra is a relay structure within basal ganglia brainstem loops that is centrally involved in gait modulation. OBJECTIVE: We tested the hypothesis that quantitative gait biomechanics are related to the microstructural integrity of the substantia nigra and PD-relevant gait abnormalities are independent from bradykinesia-linked speed reductions. METHODS: Thirty-eight PD patients and 33 age-matched control participants walked on a treadmill at fixed speeds. Gait parameters were fed into a principal component analysis to delineate relevant features. We applied the neurite orientation dispersion and density imaging (NODDI) model on diffusion-weighted MR-images to calculate the free-water content as an advanced marker of microstructural integrity of the substantia nigra and tested its associations with gait parameters. RESULTS: Patients showed increased duration of stance phase, load response, pre-swing, and double support time, as well as reduced duration of single support and swing time. Gait rhythmic alterations associated positively with the free-water content in the right substantia nigra in PD, indicating that patients with more severe neurodegeneration extend the duration of stance phase, load response, and pre-swing. CONCLUSION: The results provide evidence that gait alterations are not merely a byproduct of bradykinesia-related reduced walking speed. The data-supported association between free-water and the rhythmic component highlights the potential of substantia nigra microstructure imaging as a measure of gait-dysfunction and disease-progression.

Acute Effects of Various Movement Noise in Differential Learning of Rope Skipping on Brain and Heart Recovery Analyzed by Means of Multiscale Fuzzy Measure Entropy.

In search of more detailed explanations for body-mind interactions in physical activity, neural and physiological effects, especially regarding more strenuous sports activities, increasingly attract interest. Little is known about the underlying manifold (neuro-)physiological impacts induced by different motor learning approaches. The various influences on brain or cardiac function are usually studied separately and modeled linearly. Limitations of these models have recently led to a rapidly growing application of nonlinear models. This study aimed to investigate the acute effects of various sequences of rope skipping on irregularity of the electrocardiography (ECG) and electroencephalography (EEG) signals as well as their interaction and whether these depend on different levels of active movement noise, within the framework of differential learning theory. Thirty-two males were randomly and equally distributed to one of four rope skipping conditions with similar cardiovascular but varying coordinative demand. ECG and EEG were measured simultaneously at rest before and immediately after rope skipping for 25 mins. Signal irregularity of ECG and EEG was calculated via the multiscale fuzzy measure entropy (MSFME). Statistically significant ECG and EEG brain area specific changes in MSFME were found with different pace of occurrence depending on the level of active movement noise of the particular rope skipping condition. Interaction analysis of ECG and EEG MSFME specifically revealed an involvement of the frontal, central, and parietal lobe in the interplay with the heart. In addition, the number of interaction effects indicated an inverted U-shaped trend presenting the interaction level of ECG and EEG MSFME dependent on the level of active movement noise. In summary, conducting rope skipping with varying degrees of movement variation appears to affect the irregularity of cardiac and brain signals and their interaction during the recovery phase differently. These findings provide enough incentives to foster further constructive nonlinear research in exercise-recovery relationship and to reconsider the philosophy of classical endurance training.

Deficient Interhemispheric Connectivity Underlies Movement Irregularities in Parkinson's Disease.

BACKGROUND: Movement execution is impaired in patients with Parkinson's disease. Evolving neurodegeneration leads to altered connectivity between distinct regions of the brain and altered activity at interconnected areas. How connectivity alterations influence complex movements like drawing spirals in Parkinson's disease patients remains largely unexplored. OBJECTIVE: We investigated whether deteriorations in interregional connectivity relate to impaired execution of drawing. METHODS: Twenty-nine patients and 31 age-matched healthy control participants drew spirals with both hands on a digital graphics tablet, and the regularity of drawing execution was evaluated by sample entropy. We recorded resting-state fMRI and task-related EEG, and calculated the time-resolved partial directed coherence to estimate effective connectivity for both imaging modalities to determine the extent and directionality of interregional interactions. RESULTS: Movement performance in Parkinson's disease patients was characterized by increased sample entropy, corresponding to enhanced irregularities in task execution. Effective connectivity between the motor cortices of both hemispheres, derived from resting-state fMRI, was significantly reduced in Parkinson's disease patients in comparison to controls. The connectivity strength in the nondominant to dominant hemisphere direction in both modalities was inversely correlated with irregularities during drawing, but not with the clinical state. CONCLUSION: Our findings suggest that interhemispheric connections are affected both at rest and during drawing movements by Parkinson's disease. This provides novel evidence that disruptions of interhemispheric information exchange play a pivotal role for impairments of complex movement execution in Parkinson's disease patients.

Fatigue-Related and Timescale-Dependent Changes in Individual Movement Patterns Identified Using Support Vector Machine.

The scientific and practical fields-especially high-performance sports-increasingly request a stronger focus be placed on individual athletes in human movement science research. Machine learning methods have shown efficacy in this context by identifying the unique movement patterns of individuals and distinguishing their intra-individual changes over time. The objective of this investigation is to analyze biomechanically described movement patterns during the fatigue-related accumulation process within a single training session of a high number of repeated executions of a ballistic sports movement-specifically, the frontal foot kick (mae-geri) in karate-in expert athletes. The two leading research questions presented for consideration are (1) Can characteristics of individual movement patterns be observed throughout the entire training session despite continuous changes, i.e., even as fatigue-related processes increase? and (2) How do intra-individual movement patterns change as fatigue-related processes increase throughout a training session? Sixteen expert karatekas performed 606 frontal foot kicks directed toward an imaginary target. The kicks were performed in nine sets at 80% (K-80) of the self-experienced maximal intensity. In addition, six kicks at maximal intensity (K-100) were performed after each of the nine sets. Between the sets, the participants took a 90-s break. Three-dimensional full-body kinematic data of all kicks were recorded with 10 infrared cameras. The normalized waveforms of nine upper- and lower-body joint angles were classified using a supervised machine learning method (support vector machine). The results of the classification revealed a disjunct distinction between the kinematic movement patterns of individual athletes. The identification of unique movement patterns of individual athletes was independent of the intensity and the degree of fatigue-related processes. In other words, even with the accumulation of fatigue-related processes, the unique movement patterns of an individual athlete can be clearly identified. During the training session, changes in intra-individual movement patterns could also be detected, indicating the occurrence of adaptations in individual movement patterns throughout the fatigue-related accumulation process. The results suggest that these adaptations can be modeled in terms of changes in patterns rather than increasing variance. Practical consequences are critically discussed.

Fatigue-related changes in technique emerge at different timescales during repetitive training.

Training consisting of numerous repetitions performed as closely as possible to ideal techniques is common in sports and every-day tasks. Little is known about fatigue-related technique changes that emerge at different timescales when repeating complex actions such as a karate front kick. Accordingly, 15 karatekas performed 600 kicks (1 pre-block and 9 blocks). The pre-block comprised 6 kicks (3 with each leg) at maximum intensity (K-100%). Each block comprised 60 kicks (10 with each leg) at 80% of their self-perceived maximum intensity (K-80%) plus 6 K-100%. In between blocks, the participants rested for 90 seconds. Right leg kinematics (peak joint angles, peak joint angular velocities, peak joint linear resultant velocities, and time of occurrence of peaks) and kick duration corresponding to the K-80% were measured resulting in numerous variations with fatigue. At the timescale of tens of seconds, the changes involved variables that were related to velocity of execution (slowed down), while variables related to movement form were hardly affected. At the timescale of tens of minutes, the opposite results were observed. These findings challenge the long-standing rationale underlying repetitive training, suggesting instead that such involuntary variations in technique might play a crucial role in motor skill training.

Invited commentary: Differential learning is different from contextual interference learning.

There has been renewed interest in the detailed structure of what is learned and the boundary conditions that foster motor learning. The accompanying article by Hossner et al. (2016), particularly their findings about augmented feedback in the context of different levels of additional noise, is consistent with this focus. Unfortunately, the findings from Hossner and colleagues appear to be based on incorrect interpretations of the differential learning (DL) approach. Essential discrepancies in the experimental conditions suggest the basis for the deviating results obtained in comparison to those of the original DL experiments. In this comment, it is also shown that the author's assumptions and interpretations underlying CI and the DL approaches obscure crucial problems and contradictions of classical learning theory.