Audiovisual relative timings determine sound-induced flash fission versus flash fusion effects

Abstract

When observers are required to make judgments of the number of rapidly presented flashes of light, there is a tendency to either overestimate the count (‘flash fission’) or underestimate the count (‘flash fusion’), depending on the duration of the inter-flash interval (Bowen, 1989). Similarly, pairing the flashes with more or less loud, rapid beeps also results in fission and fusion effects (the sound-induced flash illusion, Andersen, Tiippana, & Sams, 2004; Shams, Kamitani, & Shimojo, 2000, 2002). Our aims were to determine how much these sound induced effects are dependent upon timings between clicks and flashes, and how these critical timings relate to the audiovisual ‘window of integration’ of around 100msec. A high contrast, 2° disc was flashed (11.7 ms, 7° periphery) in the presence of 0, 1, 2, or 3 beeps (7 ms, ∼ 75dbA, 3.5 kHz) with various audiovisual relative timings between 12 and 300 msec. Results from naive observers demonstrate flash fusion when >100 ms separated all stimuli, whereas flash fission was reported for separation <50 ms, with the transition from fission to fusion occurring rapidly. These results were replicated in stepped-on, ramped-off or ramped-on, stepped off discs. Large and consistent fission effects occurred when the sounds were presented: 1. during the ramp and 2. within 100 ms of the transient. A control experiment showed that no consistent illusory rapid change was induced by the beeps in a disc smoothly ramping on and off. We propose that the fission effects are occurring as the result of an intersensory process, whilst the fusion effect are occurring as the result of response bias.