Abstract
Research on audiovisual integration has shown the superior temporal sulcus (mSTS) to respond differentially to temporally-synchronous and asynchronous stimuli. Specifically, a sub-region of mSTS responds only to stimuli that are temporally aligned in a binary fashion, with no response to stimuli out-of-sync even 100ms. A second sub-region of mSTS has been shown to produce a graded response: the more asynchronous a stimulus, the higher the BOLD response. Here, we investigate the latter sub-region by testing a hypothesis that this sub-region of mSTS responds differentially to the perception of asynchrony, as opposed to the level of asynchrony of the stimulus.
To test this hypothesis, we identified the asynchrony level at which each participant perceived audiovisual spoken-word stimuli to be synchronous 50% of the time and asynchronous 50% of the time (mean = 167ms). Two types of fMRI runs were presented. First, block-design localizer runs in which audio-only, visual-only, or audiovisual spoken words were presented while participants performed a one-back task. Second, event-related runs in which audiovisual spoken words were presented with either no offset (synchronous), a 400ms offset (asynchronous), or at the individuals 50%-level offset while participants performed a synchrony-judgment task.
A sub-region of mSTS was identified using localizer runs, defined as the conjunctions of areas that responded to audio-only and visual-only presentations. BOLD responses were then extracted from event-related runs. As previously shown, asynchronous stimuli elicited greater BOLD responses than synchronous stimuli. Trials at the 50%-level of asynchrony were divided by perception into perceived-synchronous and perceived-asynchronous trials. Importantly, BOLD responses with perceived-synchronous trials matched those of physically-synchronous trials (offset = 0ms), and BOLD responses with perceived-asynchronous trials matched those of asynchronous trials (offset = 400ms). Thus, activation in this region of mSTS is influenced primarily by the perceived synchrony of the audiovisual stimulus, rather than the actual physical synchrony.