Abstract
When searching for a target in complex visual scenes, reaction time is faster for targets appearing in the repeated contexts than for targets appearing in different contexts (Chun & Jiang, 1998). One example of these contexts is the association between target and distractor shapes (Chun & Jiang, 1999). The current study investigated whether this contextual cueing effect could occur even when some of the stimuli were invisible. We used the same shapes as in Chun & Jiang (1999). One target and eight distractors were always presented in the dominant eye. We made two of the eight distractors invisible using inter-ocular suppression. Specifically, we presented two pattern masks to the corresponding locations of the two suppressed distractors in the opposite eye (invisible condition). This invisible condition was further divided into the two conditions: the two suppressed distractors could be either physically present (masked) or absent (yoked). In addition to this visibility modulation, we had two types of context (consistent or variable mapping between the target and distractor shapes) and six epochs to learn the association between the target and distractor shapes. Participants' task was to search for the single target that was symmetric about the vertical axis. Distractors were symmetric about the differently oriented axes. In the visible condition, we found that improvement in search performance was significantly more pronounced in the consistent contexts than in the variable contexts, as the epoch increased. These results are consistent with Chun & Jiang (1999). However, we did not find this contextual cueing effect both in the masked and yoked conditions. Nevertheless, overall search performance was faster in the masked condition than in the yoked condition. These results suggest that invisible stimuli cannot induce contextual cueing effect. They can only facilitate procedural aspects of visual search.
Meeting abstract presented at VSS 2012