Abstract
Purpose: The repetition of spatial configurations of search displays implicitly facilitates visual search (contextual cueing effect; Chun & Jiang, 1998). A similar effect was also observed when the object identities were repeated (Chun & Jiang, 1999). However, it is unclear whether the repetition of the object identities facilitates form processing of search items or attentional guidance to a target. In this study, we measured eye movements in order to clarify the mechanism of the object-identities-based contextual cueing effect. Method: Participants were asked to a search target among 9 distractors. A target was a novel object whose shape was symmetric around the vertical axis, and a distractor was symmetric around an axis apart from the vertical axis. 12 target objects were paired with 9 distractor objects. These 12 sets were repeatedly presented in the experiment (repeated sets). The other target objects were presented with different distractor objects for each trial (non-repeated set). The locations of the target and the distractors were randomized in each trial. In the half of trials, repeated sets were presented. In the other half of trials, non-repeated sets were presented. Eye movements were recorded with Eyelink2. Result: Reaction times were shorter in repeated sets than in non-repeated sets, showing a typical contextual cueing effect. More importantly, fewer saccades were required to find targets in repeated sets than in non-repeated sets, and the difference was greater in later trials than in early trials of the experiment. On the other hand, durations of fixations did not differ between object-set conditions and were constant throughout the experiment. Although latencies of the first saccades became shorter as the experiment progressed, there was no difference between repeated sets and non-repeated sets. These results indicate that object-identities-based contextual cueing modulates attentional guidance to a learned target, as well as configuration-based contextual cueing.