The time will come: Evidence for an eye-audiation span in silent music reading

Abstract

Musical literacy allows one to “hear music from the page”. What can we say about this internal music if we follow the reader’s eyes? Do readers hear a given fragment while they are looking at it? Or do they hear it later, when they are already gazing at the following fragment? We hypothesized that the second possibility is more likely, since it allows the reader to start processing one fragment while the previous one is being heard, and thus to keep the musical rhythm going. We refer to this as the eye-audiation span hypothesis, which we tested with an innovative eye-tracking paradigm. We found convergent evidence of an eye-audiation span: first, temporal representations (the internal rhythms) are not concurrent with gaze; second, they emerge later than gaze (gaze-lagged representations). Evidence of lagged temporal representations was stronger in non-experts compared to experts, suggesting either that experts are more efficient in parallel processing, or that their representations are more amodal. Our approach to the relation between gaze and internal rhythm paves the way to mind-reading silent music readers, and provides cues for understanding mechanisms in extra-musical domains, such as implicit prosody in text reading.