Abstract
A hallmark of visual expertise is the ability to efficiently encode domain-related visual patterns. We explored this perceptual component of expertise in the domain of music reading by monitoring the eye movements of expert musicians (with a minimum of 10 years of music reading experience) and novices (who could not read music) while they completed a music-related flicker paradigm task. Specifically, participants searched for a single note in a music score that was rapidly appearing and disappearing (with a blank screen presented in between each screen change). This change detection paradigm allowed us to explore interactions between expertise and visual complexity using stimuli that varied dramatically in complexity. As well, this paradigm allowed us to study a wide span of expertise differences (since both experts and novices could complete this task with high accuracy rates), while also controlling for potentially confounding effects of variables known to vary across music scores (e.g., tempo differences between simple and complex pieces during musical performances). Using this paradigm, we demonstrated strong effects of both expertise and complexity. Relative to visually complex music scores, visually simple scores elicited shorter fixation durations, shorter dwell durations, larger saccade amplitudes, faster times to the first fixation on the changing note (i.e., "time to first fixation"), and faster reaction times. Further, relative to novices, experts displayed shorter dwell durations, larger saccade amplitudes, faster time to first fixation, and faster reaction times. Interestingly, visual complexity interacted with expertise, such that novices displayed larger complexity effects than experts. This pattern of results is consistent with the predictions of chunking and template theories of human visual expertise. The increase in expertise effects for complex music scores may reflect the expert musicians' ability to use their memory for domain-related visual configurations to compensate for increases in complexity.
Meeting abstract presented at VSS 2018