Andrew Clement, PhD

Assessing the Effects of Statistical Learning on Multiple-Target Search

Visual search is the process of locating a target among distractors. Many theories assume that people search for objects by maintaining an internal representation, or template, of the target in working memory. However, the capacity of working memory is severely limited, making it difficult to maintain multiple target templates. Indeed, when participants search for multiple targets, they are slower, less accurate, and more likely to fixate distractors. To reduce these multiple-target search costs, I recently developed a novel statistical learning procedure that can facilitate multiple-target search. Preliminary evidence confirms the effectiveness of this procedure in reducing multiple-target search costs. The primary goal of the proposed research will be to clarify the mechanisms of these effects and assess whether this procedure can reduce other multiple-target search costs. A secondary goal will be to provide new opportunities for involving undergraduate students in neuroscience research at Millsaps College. To accomplish these goals, I propose three specific aims: to identify the attentional and memory mechanisms of these effects (Aim 1), to identify whether statistical learning can reduce switch costs (Aim 2), and to identify whether statistical learning can reduce subsequent search misses (Aim 3). By addressing these aims, the proposed research will provide novel insights into the mechanisms of multiple-target search and will aid in the development of novel training procedures that can reduce multiple-target search costs. The proposed research will also provide new opportunities for involving undergraduate students in neuroscience research, which are currently lacking at Millsaps College.