Introduction & Purpose During training and competition, athletes must cope with numerous demands that can influence their performance. These include physiological and mental fatigue, psychological stress, and the ability to adapt quickly to varying situations. Controlling the high number of degrees of freedom with which the musculoskeletal system can produce movements is challenging for the central nervous system CNS (Bernstein, 1967), even without the context of a sporting setting. One common theory in the field of motor control posits the existence of muscle synergies (Turpin et al., 2021). These are constituted of synergy vectors within the spinal cord that comprise the relative weightings of different co-active muscles. Synergy vectors are activated by time-varying activation coefficients, which correspond to the central commands from supraspinal areas. Consequently, the CNS is not required to individually time and scale the activation of each muscle independently; rather, it only regulates the activation of a limited number of synergies (Turpin et al., 2021). This simplification strategy has been extensively studied in a variety of settings, including locomotion (Boccia et al., 2018; McGowan et al., 2010) and reaching movements (Scano et al., 2019). However, muscle synergies were rarely studied in sport-like settings which demands fall beyond daily tasks. In this light, we conducted several studies in order to build a deeper understanding of motor control in movements with high demands. This abstract presents a summary of the main findings of four studies. Specifically, SKATE, examining the complex movements of skateboard tricks; LEARN, demonstrating modifications as participants learned to walk on a tightrope; STRESS, investigating alterations under psychological stress during treadmill walking; and FATIGUE, exploring the fatigue strategies of climbers during overhead hanging tasks. The overarching aim of these studies was to enhance our understanding of motor control and adaptation in movements of a high-demand nature. Methods To study muscle synergies, surface electromyography signals from multiple muscles were recorded, filtered, rectified, amplitude- and time-normalised, and concatenated across captured trials. Subsequently, factorisation algorithms were employed to extract the spatial synergy vectors and time-varying activation coefficients for each participant. The required number of synergies to perform the movement was then determined by the total variance accounted for by a given number of synergies (Turpin et al., 2021). SKATESeven recreational skateboarders performed three tricks (Ollie, Kickflip, 360°-flip), which involved different rotations of the board in the air. Each participant was required to land the Ollie, Kickflip and 360°-flip successfully six times. Muscle activity of eight muscles per leg was collected and used to extract muscle synergies of each trick during the take-off phase. The degree of similarity between synergy vectors among tricks was quantified by calculating the Pearson correlation coefficient. Synergies were identified as similar if the correlation exceeded a threshold of 0.623 (Kaufmann et al., 2024). LEARNTen participants were required to walk over a line taped on the floor, a beam, and learn to walk over a tightrope, all within one single data collection session. Muscle activity of thirteen muscles of the right leg was collected for several trials of each walking task and used to extract muscle synergies. The temporal overlap between different synergies within one trial and the trial-to-trial similarity between different trials within one synergy were determined through the Pearson correlation of the activation coefficients (Figure 1A). STRESSEight participants walked on a treadmill at a self-selected, constant velocity under two conditions in a random order. Once walking during a psychological stress condition, induced by the Paced Auditory Serial Addition Task (Gronwall, 1977), and once without stress. The stress level was measured through the tonic skin conductance, and force insoles were used to determine foot contacts. Muscle activity of four to seven muscles on each leg was collected for all steps and used to extract muscle synergies. The trial-to-trial similarity of the activation coefficients was determined by the Pearson correlation coefficient (Figure 1A). The coefficient of variation among steps was employed to quantify the trial-to-trial similarity of the participants’ stance-phase to gait-cycle ratio as a temporal-spatial gait parameter. FATIGUEEleven climbers performed sustained isometric finger hangings until failure. Muscle activity of six arm and trunk muscles per limb was recorded throughout the hang and used to extract muscle synergies. Subsequently, similar synergy vectors among participants were clustered. The mean activation from the activation coefficients of synergies within the same cluster was compared between the first and last 20% of the hang in order to identify changes between a non-fatigued and a fatigued stage. Results SKATEThree to six synergies were required among participants and tricks. At least one similar synergy vector was identified for each pair of trick comparisons. Furthermore, one consistent synergy vector was observed in all three tricks. This synergy vector was primarily formed by the quadriceps muscles and was activated during the initial third of the take-off phase, thereby being responsible for the fundamental task of jumping. LEARNAcross participants, four to eight synergies were required to complete all walking tasks. Statistical analyses revealed a significantly higher trial-to-trial similarity (p < 0.05) alongside a lower temporal overlap of activation coefficients during walking on a line compared to walking over the beam or the tightrope and during beam walking compared to tightrope walking. Furthermore, trial-to-trial similarity was higher and overlap was lower at...