Abstract
Many tasks we perform daily require multitasking. For example, we frequently perform an action while simultaneously monitoring a visual scene. Here we investigated eye-hand coordination while participants repeatedly performed a ball-drop task that involved inserting a ball in one of three slots using different grasp modes (fingertips, tweezers). We compared gaze allocation when the task was performed in isolation (single task) or while simultaneously monitoring a display of letters that randomly changed every 1.5 to 6 seconds and could only be detected with central vision (dual task). In the single task, participants typically fixated the ball during reach and shifted gaze to the slot once the ball was grasped with either fingertips or tweezers. In the dual task, gaze patterns differed between grasp modes. In many fingertip trials (48.2%), participants exclusively fixated the display (display-only). In most remaining trials (39%), participants shifted gaze from the display to the slot and back (display-slot-display). In most tweezer trials (60.7%), participants shifted gaze from the display to the ball, returning briefly to the display before shifting gaze to the slot and back to the display (display-ball-display-slot-display). Occasionally (19%), participants shifted gaze from the ball directly to the slot without returning to the display (display-ball-slot-display). Irrespective of task conditions, the onsets of ball and slot fixations were spatiotemporally coupled to action sub-goals: the time of ball grasp and slot entry. Interestingly, the chosen gaze pattern depended on the timing of the letter change. Participants were more likely to shift gaze away from the display (display-slot-display in fingertips and display-ball-slot-display in tweezer trials) when the action sub-goals occurred during a 1.5 s time window in which no second letter change could occur. These results indicate that participants exploited the statistics of the environment and adaptively allocated gaze to simultaneously support competing perceptual and motor tasks.