Zephyr Markley¹, Gabriel Conn¹, Katherine S. Moore¹; ¹Arcadia University

Decades of research on visual search in basic and applied settings have informed theoretical models of how we carry out such tasks in a variety of settings and which external and internal factors affect search accuracy and speed. Among these factors include target visibility, target-distractor similarity, and subsequent miss errors (e.g., early quitting upon finding one of several targets in a display). While these and other factors have been identified in laboratory and applied settings, such as TSA safety searches and radiology scans, little attention has been given to dermatological searches. In the present study, novice searchers (university students) searched for cancerous melanoma and carcinoma moles among benign moles, simulating a skin search. First, participants were trained to identify the three different mole types. Once they reached criterion, they performed a speeded skin search for the cancerous mole targets. We varied (1) target visibility with skin tone, (2) target-distractor similarity (melanoma is more visually similar to benign moles than is carcinoma), (3) the number of targets on a trial, to measure subsequent misses, and (4) time pressure, including the presence of a ticking clock on ½ of all trials, to simulate the time pressure that dermatologists often find themselves under during an office visit. Consistent with visual search literature, skin tone (target visibility) and mole type (target-distractor similarity) affected performance, but there was no significant subsequent miss error effect. Participants were faster and less accurate in the presence of a clock than when no clock was present, suggesting the time pressure included a speed-accuracy trade-off. Our findings may inform how novices can be trained to conduct self-screens. Reliable and valid screenings can not only save lives but also save on expensive and invasive procedures by avoiding false positives.