With or Without Feedback: Implicit sequence learning relies on visuospatial coding regardless of the presence of auditory feedback during practice


Contingently mapped auditory response-effects may enhance sequence learning in the implicit Serial Reaction Time Task (SRTT). While sound appears to improve sequence-specific learning, it is not clear if sound changes how the sequence is encoded. Thus, the present study investigated if spatially contingent sounds influence if the sequence is represented in motor or visuospatial coordinates. Sixteen neurotypical righthanded adults (18-35 years-old) practiced the SRTT with their preferred hand by reaching to an array of four coloured target outlines on a touchscreen, either with spatially contingent auditory feedback, or without sound. On day 1, participants practiced a 10-item sequence (i.e., targets filled in) that repeated 10 times throughout practice blocks 1-6 and 8; while block 7 presented pseudorandom stimuli. On day 2, participants completed a retention test followed by four transfer tests to evaluate sequence performance with the preferred and non-preferred hands in both parallel (spatially congruent) and mirrored (motor congruent) conditions. Both groups improved significantly across practice blocks on day 1. Analysis of the transfer tasks revealed that both groups performed the sequence faster in the parallel transfer tests, indicating that visuospatial coding had occurred regardless of the presence of auditory feedback during practice. Further, when compared to the auditory feedback group, the no sound group took significantly longer to execute the sequence in the auditory feedback transfer conditions. In contrast, the auditory feedback group performed as well as the no sound group in the no sound transfer conditions, indicating an asymmetrical specificity of practice effect for auditory feedback.