Dissociable cognitive strategies for sensorimotor learning

Samuel D Mcdougle,Jordan A Taylor

doi:10.1038/s41467-018-07941-0

Samuel D Mcdougle, Jordan A Taylor

Open Access

https://doi.org/10.1038/s41467-018-07941-0

Copy DOI

Abstract

Computations underlying cognitive strategies in human motor learning are poorly understood. Here we investigate such strategies in a common sensorimotor transformation task. We show that strategies assume two forms, likely reflecting distinct working memory representations: discrete caching of stimulus-response contingencies, and time-consuming parametric computations. Reaction times and errors suggest that both strategies are employed during learning, and trade off based on task complexity. Experiments using pressured preparation time further support dissociable strategies: In response caching, time pressure elicits multi-modal distributions of movements; during parametric computations, time pressure elicits a shifting distribution of movements between visual targets and distal goals, consistent with analog re-computing of a movement plan. A generalization experiment reveals that discrete and parametric strategies produce, respectively, more localized or more global transfer effects. These results describe how qualitatively distinct cognitive representations are leveraged for motor learning and produce downstream consequences for behavioral flexibility.

Highlights

Computations underlying cognitive strategies in human motor learning are poorly understood
We hypothesized that strategies would take two broad forms, either discrete response caching (RC) or parametric mental rotation (MR)
RC is here defined as the maintenance of acquired one-toone associations between a set of stimuli and a set of responses maintained in memory[14,15], perhaps relying on processing in prefrontal cortex[16,17]

Summary

Introduction

Computations underlying cognitive strategies in human motor learning are poorly understood. A common strategy (used in piano pedagogy) to overcome this limitation is to approach it parametrically: She can anchor her thumb on middle C and reference other notes on the lines of the musical staff relative to this key While this strategy affords her the ability to play a more complex melody within a few minutes of practice, it becomes increasingly cumbersome the further a given note is from middle C—echoing the phenomena of mental scanning and mental rotation[2,3]. These two strategies, one a discrete map (caching stimulus–response pairs) and the other a parametric algorithm (computing relative distances from C), offer two ways to approach learning a novel motor skill. As a form of item-based working memory, the efficacy of RC should be subject to load (e.g., the number of items to be stored[14,18])

Methods

Results

Conclusion