Abstract
Models of visual search often assume constant revisiting rates and dwell times– irrespective of search difficulty. Previous studies showed, however, that both gaze dwell times and the number of refixations on stimuli covary with target-distractor similarity. Moreover, dwell times and refixations appear to determine search times in target absent trials even more strongly than the number of fixated stimuli. However, these results were obtained in studies including searches for naturalistic faces which had higher complexity than stimuli that are usually used for visual search experiments. In the present study, it was tested whether such findings generalize to simple search stimuli, where participants search for a closed ring among rings with a gap. Results showed that, compared to experiments with complex stimuli, the relative contribution of the number of fixated stimuli increased, whereas the effects of dwelling and revisiting were still substantial. Crucially, all three gaze parameters covaried both with target-distractor similarity and distractor-distractor similarity, which were manipulated in a parametrical and orthogonal fashion. Implications for current visual search models are discussed.
Acknowledgement: The work was supported by the Cluster of Excellence – Cognitive Interaction Technology