Visual afferent information dominates other sources of afferent information during mixed practice of a video-aiming task.

The goal of the present study was to determine whether learning of a manual video-aiming task that would have to be performed in a target-only condition (transfer test) would be improved, unaffected or hindered by alternating practice between a normal vision and a target-only condition. The results of the present study indicate that a mixed schedule of practice hindered learning of the task. Specifically, participants who practiced the task in the mixed schedule showed larger direction constant error in transfer than participants who practiced only in the target-only condition. The results suggest that participants in the mixed condition continued to rely on the rapidly decaying information that was extracted in the normal vision condition to plan and control their movement in the target-only condition, which ultimately led to biased movement endpoints.