Divergent effects of pitch feedback on online and offline motor sequence learning

Pauline Ploettner^*Christoph MuehlbergFelix PsurekChristopher FrickeJost-Julian Rumpf^*

• Department of Neurology, University of Leipzig Medical Center, Leipzig, Germany

Introduction: Motor sequence learning - the integration of individual movement elements into coordinated actions - is essential for everyday skills. This process comprises online learning during practice and post-practice offline consolidation. A key mechanism is action–perception coupling, in which motor actions become linked with predictable sensory outcomes. Pitch feedback, which conveys timing and spatial information, may strengthen this coupling and facilitate skill acquisition. Here, we evaluated pitch feedback as a tool to modulate both online and offline motor sequence learning. Methods: We included sixty healthy young non-musicians (mean age: 28.4 ± 4.6 years) who were asked to perform a finger-tapping task on a MIDI keyboard. They were randomly assigned to one of three auditory feedback groups: congruent, fixed, and random pitch feedback. The task involved repeatedly performing an 11-item sequence with the right hand. Pitch feedback was delivered according to group assignment during 14 training blocks of six sequences each. Prior to training, participants completed one block of the task without pitch feedback to assess baseline performance. Retention was tested six hours later under two conditions: seven blocks without pitch feedback (Retest 1) and seven blocks with pitch feedback (Retest 2). Results: Congruent pitch feedback facilitated online learning across the initial training session compared to fixed or random feedback. This advantage of congruent pitch feedback persisted during retesting in the presence of feedback (Retest 2), but did not generalize to task performance in the absence of pitch feedback (Retest 1). Importantly, while online learning and task performance were facilitated by congruent pitch feedback, between-session performance changes were significantly larger in the group that received random pitch feedback during the initial training session compared to the congruent and fixed feedback groups. Conclusion: These findings highlight a dissociation between feedback types that optimize immediate performance and those that promote lasting motor memory formation. While congruent pitch feedback facilitates online skill acquisition compared to fixed or random pitch feedback, unpredictable auditory input may challenge learners to engage internal monitoring mechanisms, leading to more robust, feedback-independent motor memory consolidation. These insights have implications for optimizing auditory feedback in motor learning and neurorehabilitation contexts.