Abstract
Real world action tasks often involve operating in a cluttered environment, in which we search for a target object among distractors. In many cases, the structure of these environments dictate large movement costs associated with reaching the located target (e.g., reaching a cup on a high shelf). While it is well established that people are sensitive to movement costs when selecting between potential movements, it is unclear whether movement costs likewise influence visual search behaviour. Here we tested whether visual search behaviour, as measured by gaze, is biased by the movement costs associated with acting on a target object. In each trial, an obstacle was briefly displayed and then a set of 36 objects, including 4 targets and 32 distractors, were displayed. The length, location, and orientation of the obstacle and the locations of the target objects were randomly varied. The task was to locate a target and then reach for it using a cursor controlled by the handle of a robotic manipulandum. The handle could apply forces simulating any contact between the cursor and the unseen obstacle. The cursor start position was in the center of the display and the objects in a given trial were either on the ‘near’ or ‘far’ side of the obstacle. We found that search, and hence target selection, was biased towards the near side of the obstacle, thus reducing the time and energy costs associated with reaching. This result suggests that humans can readily incorporate the movement costs of an environment when forming visual search strategies.