Abstract
We conducted a virtual reality experiment to observe the effect of head and eye movements during visual search. 33 participants searched for a salient (O) or non-salient (T) target in an environment with 7 distractors (L), fixated on the target and pressed a button as fast as possible. Each trial started with a fixation cross at 0°. The objects were arranged in four virtual flat panels positioned around the observer, each containing two objects in fixed positions. Inner clusters were visible peripherally (-35° and 35°) at the beginning of each trial. A head movement was required to inspect either outer cluster (-70° and 70°). Objects in the outer clusters were ‘visible’ throughout the trial, or became visible (‘hidden condition’) after gaze passed -50° / 50°. Search time was defined as the time from trial start to target fixation onset before button press. Inner targets were found significantly faster (M = 0.71s) than outer targets (M = 1.82s). Salient inner targets were found significantly faster (M = 0.62s) than non-salient inner targets (M = 0.8s). There was no significant difference between salient (M = 1.85s) and non-salient (M = 1.8s) targets in the outer clusters. Search times in the hidden condition (M = 1.85s) were not significantly different from already visible outer targets (M = 1.81s). When searching for inner targets, participants made the first gaze in the correct direction in 72.6% of salient and 62.7% of non-salient trials. Our results suggest saliency facilitates search when starting from a stationary gaze, making it more likely searches are initiated in the correct direction. For outer targets, however, search times were the same no matter if the potential target was visible or hidden before starting the gaze movement. In a non-stationary gaze situation during an ongoing search, the saliency effect is absent.