Abstract
Oculomotor inhibition of return (O-IOR) is an increase in saccade latency when returning to a recently fixated location relative to other locations. It has been proposed that this temporal O-IOR has spatial consequences, facilitating foraging by inhibiting return to previously attended regions (Klein & MacInnes, 1998). Models of visual attention incorporate this spatial IOR effect to drive attention to new locations. However, the foraging facilitator hypothesis has been a matter of debate recently in the scene literature, with several studies observing increased rather than decreased return probability (Hooge et al., 2005; Smith & Henderson, 2009, 2011a,b). To test whether spatial O-IOR occurs and can facilitate foraging, we monitored participants’ eye movements while they viewed arrays of objects or words. Arrays were chosen over scenes because arrays offer the advantage that fixation locations can be unambiguously assigned to specific objects. Temporal O-IOR was observed for both objects and words, but there was no evidence for spatial O-IOR. Instead, we observed spatial facilitation of return; participants were more likely than chance to refixate just-visited locations. Further, the likelihood of making a return saccade to an object or word was contingent on time spent previously on that object or word, suggesting that return saccades are triggered when further visual information is needed. The results also revealed strong evidence for saccadic momentum, an oculomotor bias to repeat the most recently executed saccade program. Such repetitions were the most frequently-occurring saccades in our data. Taken together, these results suggest that inhibition of return is not a foraging facilitator. We suggest that models of visual attention could incorporate saccadic momentum as a foraging facilitator instead of O-IOR. We further suggest that such models should incorporate a mechanism that triggers return saccades when more information is needed from a location.
Meeting abstract presented at VSS 2013