Abstract
INTRODUCTION: Variability in task execution, target properties, and recording procedures has limited the development of standardized upper-limb surface electromyography (sEMG) reference patterns. Establishing reproducible muscle activation profiles is essential for advancing sEMG-based assessment and motor control applications. METHODS: This study investigated the reproducibility of muscle activation during a highly standardized reach-to-grasp task performed under controlled setup conditions in twenty-seven healthy adults. Participants performed repeated executions of the task under controlled conditions. sEMG signals were processed using time-normalized activation envelopes and cross-correlation analyses allowing for small temporal shifts. Both intra-subject and inter-subject reproducibility were quantified, and a descriptive composite reproducibility measure, combining these two descriptors, was used to summarize overall activation stability. RESULTS: High reproducibility was observed within subjects, with correlation values ranging from 0.75 to 0.89, and across subjects, with correlations between 0.80 and 0.83. The composite reproducibility analysis highlighted the anterior deltoid, trapezius, lateral deltoid, and biceps brachii as the most reliable contributors to a reproducible activation pattern observed during the reach-to-grasp task. DISCUSSION: These findings indicate that a reduced subset of muscles can reliably represent the reach-to-grasp movement, supporting the design of simplified electrode configurations that minimize redundancy while preserving essential neuromuscular information. Overall, this work provides a methodological step toward the development of standardized upper-limb activation reference profiles and contributes to the development of efficient sEMG-based assessment, monitoring, and motor control strategies.