Abstract
The color wheel continuous-report task has frequently been used to study visual working memory and the nature of its capacity limits. This task requires fine motor control, typically with a mouse, with the precision of responses being interpreted as due to the nature of memory representation, but without regard to the contribution of noise from the response effectors (i.e., the hand). Here, we tested the hypothesis that motor noise contaminates our estimates of visual memory representations in the continuous-report task. To test this hypothesis, subjects were asked to complete the color wheel continuous-report task, using either their non-dominant or dominant hand on different blocks of trials. Subjects reported the color of a single probe object on each trial, while we varied the set size of to-be-remembered colored squares from 1 to 4 randomly across trials. When comparing dominant hand to non-dominant hand performance across set sizes, the only significant difference we discovered was in reaction time. We found that subjects took significantly longer time to complete the task with their non-dominant hand, but this did not affect error rates in reporting. Our findings suggest that this commonly used task to study visual memory is relatively immune to contamination from motor noise at the output stage, thus demonstrating the potential of the task to study memory mechanisms in patient populations.