Low but accurate detection rates for small degrees of Audiovisual Asynchrony
Durk Talsma
Poster
Time: 2009-07-02 09:00 AM – 10:30 AM
Last modified: 2009-06-04
Abstract
It is a well known phenomenon that near simultaneously presented auditory and visual stimuli are being perceived as one multisensory object. Multisensory integration can take place, even when there is a considerable stimulus onset asynchrony (SOA) between the visual and auditory inputs (up to 250 ms), suggesting that auditory and visual stimuli are internally realighened. In the present study we sought to address the question of not only at which SOAs participants began to detect these offsets, but also whether they would be able to tell whether the auditory stimulus preceded or succeeded the visual stimulus. Auditory and visual stimuli were presented in pairs, with a randomly varying SOA of -450 (auditory stimulus preceding the visual one) to +450 ms (auditory stimulus succeeding the visual one). Participants were instructed to respond only when they believed there was an actual offset between the stimuli, and if so, make a two-choice response to indicate whether the visual stimulus was earlier or later. All stimuli were grouped in 7 bins, depending on SOA. Results show that detection rate increased with SOA, generally being higher when the auditory stimulus preceded the visual one. Interestingly, even when detection rate was low, for the smaller SOAs, detection accuracy remained relatively constant across all SOA conditions, suggesting that observers still have access to the underlying temporal structure, even at small SOAs.