Aim of this study was to investigate the neural correlates of audio and tactile integration by means of magnetoenceohalography (MEG). Participants were presented with a one-second sound of a mosquito approaching the head from the left side. Furthermore a spatially correlated tactile stimulus was presented at different stimulus onset asynchronies (SOAs), yielding ambiguous perceptions as to whether the touch co-occured simultaneously with the end of the sound. Participants were asked to evaluate whether the approaching mosquito and the touch belonged to the same event (i.e. “integrated”: the impression that it was the approaching mosquito that touched them) or not (i.e. “segregated”).
In a first run, we calculated the temporal thresholds (difference of offset sound to onset touch) at which the response’s rate integration/segregation was at 50% by means of a staircase procedure. Interestingly participants showed a larger temporal window of integration for tactile stimuli presented before the end of the sound.
In the subsequent MEG experiment, data were recorded for conditions with SOAs individually set in order to match the 50% rate of integration. In this way, bimodal physically identical stimuli that elicit different multisensory perceptions, have been directly compared for avoiding possible confounds introduced by linear additive model.
A comparison between trials in which participants reported an integrated percept versus a segregated percept, showed a decrease in the alpha power in the 350 ms time window preceding the tactile stimulus, for the condition were integration was reported. This pattern supports the influence of ongoing oscillatory brain activity on perception of integration between different senses.

audiotactile; looming; MEG; oscillations