Human search strategies are informed by complex target distribution statistics

Cain, Matthew S; Vul, Ed; Clark, Kait; Mitroff, Stephen R

Abstract

Many real-world visual searches contain multiple targets in one display and such searches can be especially error prone. We investigated whether searchers adapt their strategies based on complex statistics of target distributions in multiple-target searches. Separate groups searched arrays drawn from three target distributions with the same average number of targets per display, but different target-clustering properties. People searched longer after finding targets when targets tended to cluster in displays, indicating that they adjusted their strategy based on higher order statistics. These findings can help optimize critical multiple-target searches (e.g., radiological, baggage, and military searches).

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: Object Perception, Attention, and Memory
• Start Date: Nov 1, 2010
• End Date: Nov 1, 2010
• Acceptance Date: Aug 1, 2010
• Publication Date: Nov 1, 2010
• Peer Reviewed: Not Peer Reviewed
• Public URL: https://uwe-repository.worktribe.com/output/973689
• Additional Information: Title of Conference or Conference Proceedings : Object Perception, Attention, and Memory