Abstract
People excel at acquiring and mastering motor skills to move efficiently. While skill acquisition relies on establishing a new sensorimotor mapping, most daily tasks benefit from refining precision towards ideal movements. However, the mechanisms driving improvements from continued practice remain unclear. Here, we investigate continued practice effects, using known measures from skill acquisition tasks. Participants used a stylus on a digitizing tablet to control a racecar through a track as quickly and accurately as possible, across training in two consecutive days. In day 1, participants (N = 45) quickly achieved high accuracy, with lap times and path lengths improving throughout the session. In day 2, participants (N = 43) re-experienced the same track, and quickly regained prior performance. Moreover, accuracy and path lengths showed further improvements, suggesting offline gains. We then flipped the track by 180°, which initially worsened performance before quickly rebounding. Finally, participants moved through the track in reverse. Although accuracy improved within this block, lap times and path lengths were consistently longer than in previous blocks. This direction effect, however, could be due to movement biases in right-handed participants. Thus, we had a control group of participants (N = 16) train on the reverse track. Lap times and path lengths were generally longer, suggesting that performance depends on movement direction. Overall, these findings demonstrate that continued practice in a continuous motor task yields rapid improvements, offline gains, robust retention, and partial generalization, thereby shedding light on mechanisms that support long-term motor skill refinement beyond initial acquisition.
