Abstract
Many studies have shown that visual feedback of the hand is used to monitor or adjust ongoing movements. It is still unknown, however, whether vision is used to steer the hand along a desired trajectory or to guide the hand towards a desired endpoint. Even though these two functions of visual feedback appear similar on the surface, they are different from both a conceptual and a computational point of view.
Here we tested if visual feedback is used to steer the hand along a desired trajectory or towards a desired endpoint. We manipulated how visual information relevant for moving was presented to subjects (Endpoint vs. Trajectory task) and the availability of visual feedback of the moving hand (no feedback vs. feedback). We tested both direct and tool mediated movements (i.e. computer-mouse mediated cursor movements). In addition, we compared performance between free viewing and fixation of a peripheral target. Finally, we investigated whether or not performance changes when the visual information that specifies the desired Endpoint or Trajectory is extinguished at the moment of movement onset.
We found that subjects use visual feedback to correct movement errors online in both direct and tool mediated movements. Most importantly, we found that the availability of visual feedback reduces errors significantly more in the Endpoint than in the Trajectory task. The general pattern of results holds even when subjects fixate a peripheral target and when the visual information that specifies the desired Endpoint or Trajectory is extinguished at movement onset.
We conclude that visual information about the moving hand is used primarily to guide the hand towards a specific endpoint rather than to steer it along a trajectory. Moreover, this is true whether or not participants move their eyes, see the target during the movement, or use a mouse cursor rather than their hand.