|Institution:||Louisiana State University|
|Keywords:||visual search; episodic memory; semantic memory|
|Full text PDF:||http://etd.lsu.edu/docs/available/etd-03312017-151949/;|
Search efficiency is usually improved by presenting observers with highly detailed target cues (e.g., pictures). However, in the absence of accurate target cues, observers must rely only on categorical information to find targets. Models of visual search suggest that guidance in a categorical search results from matching categorically-diagnostic target features in the search display to a top-down attentional set (i.e., the search template), but the mechanisms by which such attentional set is constructed have not been specified. The present investigation examined the influences of both semantic and episodic memory on search template formation. More precisely, the present study tested whether observers incorporated a recent experience with a target-category exemplar into their search template, instead of relying on long-term learned regularities about object categories (Experiment 1) or on the semantic context of the search display (Experiment 2). In both experiments participants completed a categorical search task (75% of trials) in conjunction with a dot-probe response task (25% of trials). The dot-probe response task assessed the contents of the search template by capturing spatial attention if the dot-probe was presented at an inconsistent location relative to objects matching the search template. In Experiment 1 it was shown that observers include recently encoded objects into their search templates, when given the opportunity to do so. Experiment 2, however, showed that observers rely on context semantics to construct categorical search templates, and they continue to do so in the presence of repeated target cues related to different contexts. These results suggest that observers can, and will, rely on episodic representations to construct categorical search templates when such representations are available, but only if no external cues (i.e., scene semantics) are present to identify criterial target feature.Advisors/Committee Members: Papesh, Megan (chair), Beck, Melissa (committee member), Hicks, Jason (committee member), Hout, Michael (committee member).