Predicting multiple-target search performance using eye movements and individual differences

Birch-Hurst, Kayley; Jones, Alice M; Ong, Angela; Stoker, Kayleigh M; Smith, Melvyn L; Zhang, Wenhao; Clark, Kait

Abstract

Accuracy in visual search – the process of detecting a target amongst distractors – is critical for life-saving career searches such as radiology and airport security. These searches often contain multiple targets (e.g., a tumour and a fracture) and are prone to cognitive pitfalls such as “subsequent search misses”, where second targets are less likely to be identified in the presence of a previously detected target. Prior work has assessed individual differences in search performance, and some research has also identified eye movements associated with search errors; however, the complex relationship between observer characteristics and search strategies that may predict performance on multiple-target search tasks remains an open question. To assess the relationship between individual differences, underlying cognitive mechanisms, and search performance, we tested novice (undergraduate) participants on a multiple-target visual search task while monitoring their eye movements. Participants were required to conduct a serial search of displays containing 0, 1, or 2 target ‘T’ shapes (amongst distractor ‘L’ shapes) that varied in visual salience and confirm identification of each target via a mouse-click. After the multiple-target search experiment, participants completed a series of questionnaires on their personality traits, clinical characteristics, and hobbies/experiences related to visual attention (e.g., video gaming and sports). We present comprehensive analyses on the impacts of individual differences and search strategies on the speed and accuracy of target detection for single- and multiple-target search. We found that both observer-related traits and specific patterns of eye movements predict search speed and accuracy across a variety of target types. Our results have implications for which search behaviours are most effective for optimal accuracy and how the strategies adopted may interact with personal characteristics. These findings can directly inform the assessment and training of people with search-related careers and allow for individualised approaches to improving search accuracy.

• Presentation Conference Type: Poster
• Conference Name: European Conference on Visual Perception
• Start Date: Aug 27, 2023
• End Date: Dec 31, 2023
• Publication Date: Dec 12, 2023
• Deposit Date: Jan 16, 2024
• Keywords: visual search, individual differences, eye-tracking
• Public URL: https://uwe-repository.worktribe.com/output/11513252