EEG Dynamics of Movement Preparation and Error Processing Distinguish Motor Adaptation From De Novo Learning

Abstract

Previous research has distinguished behavioral mechanisms between motor adaptation and de novo learning. However, electroencephalography (EEG) markers dissociating them remain unclear. Here, participants (N = 32, 13 female) performed center-out reaching under a 30{degrees} fixed rotation (adaptation), mirror reversal (de novo learning), or random rotation (errors without learning), while we recorded EEG during movement preparation and post-movement feedback. We explored how perturbation type, training phase, and error magnitude shaped neural dynamics in both temporal and frequency domains. During preparation, the readiness potential (RP) showed opposite training-related modulation in the fixed and random rotations, suggesting increased reliance on updated internal models for adaptation. Mirror reversal training showed no RP change, likely reflecting reliance on effortful, explicit learning. Interestingly, we found that Lateralized Readiness Potentials (LRPs) reflected planned movement direction rather than effector-specific preparation. In the frequency domain, beta attenuation for large errors and alpha synchronization for small errors were more pronounced in the fixed rotation than in the mirror reversal, suggesting greater error-sensitivity during adaptation. During post-movement feedback, P3 amplitude decreased from early to late learning in the fixed and random rotations but remained stable in the mirror reversal, highlighting differences in cognitive demands across perturbations. A sustained late positivity emerged following small errors across perturbations, potentially indexing implicit learning. Random rotations also elicited distinct frontal and central beta modulation, likely reflecting disengagement due to task unpredictability. Together, we show that preparatory and feedback-related EEG signatures differ across perturbation types, revealing distinct neural mechanisms underlying motor adaptation and de novo learning. Significance statementUnderstanding how the brain supports skill acquisition and adaptation is critical for advancing theories and applications of motor learning. Although behavioral differences between motor adaptation and de novo learning are well established, the neural processes that distinguish them remain unclear. By comparing EEG activity during movement preparation and feedback-error processing across multiple perturbation types, we uncover distinct temporal and frequency domain signatures that uniquely characterize each learning process. Our findings show that adaptation engages neural dynamics consistent with updating internal models and error-processing, whereas de novo learning relies on explicit, cognitively demanding strategies that likely unfold over extended practice. These results provide a clear neural framework for distinguishing motor learning mechanisms.