Abstract:

Both audiovisual integration and inhibition of return (IOR) can facilitate the processing of sensory information, such as enhancing the perceptual processing. Previous studies found that IOR decreased the audiovisual integration at previously attended locations. Several hypotheses have been put forward to explain the effect of IOR on audiovisual integration: perceptual sensitivity, spatial uncertainty, and differences in unimodal signal strength. In present study, we used cue-target paradigm and manipulated the SOA conditions (400~600 ms vs. 1000~1200 ms) to investigate how audiovisual integration would be modulated by IOR induced by visual exogenous spatial cues.
The current study was a 2 (SOA conditions: 400~600 ms, 1000~1200 ms) × 3 (target modalities: visual, auditory, audiovisual) × 2 (cue validities: cued, uncued) factorial design. Twenty-seven undergraduate students were recruited as paid volunteers from a university. The visual (V) target was a red and white block (1°×1°). The auditory (A) target (duration of 100 ms) was a 1000 Hz sinusoidal tone presented by speakers. The audiovisual (AV) target was composed by the simultaneous presentation of both the visual and the auditory stimuli. At the beginning of each trial, the fixation stimulus was presented for 800~1000 ms in the center of the display. Following the fixation stimulus, a visual white square served as a exogenous cue was presented for 50 ms at the left or right location randomly. Then, the fixation stimulus was randomly presented for 150~250/450~550 ms, which was followed by a central cue with a delay of 50 ms. Before the target (100 ms) occurrence, the fixation stimulus randomly appeared again for 150~250/450~550 ms. Thus, the SOA between the peripheral cue and the target was completed in 400~600/1000~1200 ms. The target (A, V, or AV) randomly appeared (6/7) for 100 ms in the left or right locations, or no stimulus appeared (1/7). During the experiment, participants were instructed to respond to the target stimulus at any possible locations by pressing a response button as quickly and accurately as possible.
The results showed that the responses to AV targets were faster than V or A targets, indicating the appearance of the bimodal advancement effect. A smaller magnitude of audiovisual IOR as compared to visual IOR was found whether it’s in short or long SOA conditions. In addition, visual IOR effect was significantly reduced under the long SOA condition compared with the short SOA condition while the audiovisual integration effect increased by SOA. The results of the relative multisensory response enhancement (rMRE), race model (probability difference) and positive area under the curve (pAUC) showed that audiovisual integration decreased at cued compared to uncued locations under the short SOA condition but not the long SOA condition.
Based on the aforementioned findings, it is assumed that visual IOR decreased the audiovisual integration in the short SOA, and audiovisual integration would be modulated by different SOA conditions. The current result supported the hypothesis of differences in unimodal signal strength.

Key words: exogenous spatial attention, audiovisual integration, inhibition of return, race model, cue-target paradigm