Purchase this article with an account.
Artem Violentyev, Ladan Shams; Effects of auditory grouping on visual percept. Journal of Vision 2004;4(8):698. doi: 10.1167/4.8.698.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Previous studies have shown that perceptual grouping within visual modality affects the degree of modulation of the auditory percept. We examined whether perceptual grouping within auditory modality can affect the level of interaction with the visual percept even when the saliency of the sound remains unchanged. We used sound-induced flash illusion as a testbed. In this illusion, a single flash is perceived as multiple flashes when it is paired with multiple brief sounds. We tested whether the illusion is affected if the perceptual grouping of the beeps is undermined. To this end, we varied the frequency of the sounds, while keeping all other parameters constant. ERP and fMRI studies of this illusion have suggested that it occurs as a result of modulation of activity in early visual areas. Therefore, it may be expected that any sequence of multiple sounds regardless of whether or not they are grouped perceptually would lead to the illusion. We presented 7 observers with 1 or 2 flashes, paired with 0–2 beeps. The double-beep conditions consisted of either 2 uniform sounds (either both low frequency or both high frequency) or non-uniform sounds (high followed by low or vice versa). The task was to report the # of flashes. We found that the non-uniform double-sounds were significantly less effective in inducing the illusion than the uniform double-beeps. Similarly, the non-uniform double-beeps were less effective in improving the detection of physical double flashes as well. These results clearly indicate that perceptual grouping within auditory modality can affect the auditory-visual integration, and structural consistency is an important factor in cross-modal binding. In a different experiment, we examined whether or not auditory grouping always precedes auditory-visual integration by comparing conditions in which the timing of a multiple beep stream varied relative to the flash. We found the least integration when the sounds preceded the flash.
This PDF is available to Subscribers Only