Eye movements during visuomotor adaptation represent only part of the explicit learning

Abstract

Visuomotor rotations are learned through a combination of explicit strategy and implicit recalibration. However, measuring the relative contribution of each remains a challenge and the possibility of multiple explicit and implicit components complicates the issue. Recent interest has focused on the possibility that eye movements reflect explicit strategy. Here we compared eye movements during adaptation to two accepted measures of explicit learning - verbal report and the exclusion test. We found that while reporting, all subjects showed a match between all three measures. However, when subjects did not report their intention, the eye movements of some subjects suggested less explicit adaptation than what was measured in an exclusion test. Interestingly, subjects whose eye movements did match their exclusion could be clustered into two subgroups: fully implicit learners showing no evidence of explicit adaptation and explicit learners with little implicit adaptation. Subjects showing a mix of both explicit and implicit adaptation were also those where eye movements showed less explicit adaptation than did exclusion. Thus, our results support the idea of multiple components of explicit learning as only part of the explicit learning is reflected in the eye movements. Individual subjects may use explicit components that are reflected in the eyes or those that are not or some mixture of the two. Analysis of reaction times suggests that the explicit components reflected in the eye-movements involve longer reaction times. This component, according to recent literature, may be related to mental rotation.

Significance Statement Visuomotor adaptation involves both explicit and implicit components: aware re-aiming and unaware error correction. Recent studies suggest that eye movements could be used to capture the explicit component, a method that would have significant advantages over other approaches. We show that eye movements capture only one component of explicit adaptation. This component scales with reaction time while the component unrelated to eye movements does not. Our finding has obvious practical implications for the use of eye movements as a proxy for explicit learning. However, our results also corroborate recent findings suggesting the existence of multiple explicit components, and, specifically, their decomposition into components correlated with reaction time and components that are not.