Electrocardiograms (ECGs) are frequently misdiagnosed with potentially serious consequences, yet evidence for effective teaching strategies remains limited. Eye-tracking offers insight into the perceptual processes behind accurate interpretation. We conducted an eye-tracking study among medical students to assess whether priming with normal ECGs improves diagnostic accuracy. We hypothesized that participants viewing normal ECGs before interpreting unknown ECGs would perform more accurately. This between-subjects priming study presented students with one of three image types prior to an ECG task: normal ECGs with instruction that they were normal, normal ECGs without instruction, or chest X-rays (control). The diagnostic task comprised 30 ECGs across six conditions. The study was built using Experiment Builder, and eye movements recorded using an EyeLink 1000 Plus system. Students performed well above chance, with accuracy comparable to the 54% reported in meta-analysis. Atrial fibrillation was most often identified correctly, whereas left bundle branch block was the most challenging. Across groups, faster responses and finally fixating the primary abnormality (PA) predicted accuracy, although inaccurate participants spent longer overall on the PA area of interest. Students who were informed that the priming ECGs were normal responded faster and more accurately than those primed unknowingly or with chest radiographs. ECG-primed groups also showed longer fixations in non-relevant areas, while explicitly primed students showed no significant related secondary abnormality fixation sequences and more dispersed spatial attention. These early findings suggest that targeted priming with normal ECGs may support learning. Variation in diagnostic difficulty across conditions indicates that tailored training approaches may be required.
Edwards et al. (Thu,) studied this question.