Neural dynamics driving audio-visual integration in autism

Abstract

Audio-visual (AV) integration plays a crucial role in supporting social functions and communication in autism spectrum disorder (ASD). However, behavioral findings remain mixed and, importantly, little is known about the underlying neurophysiological bases. Studies in neurotypical adults indicate that oscillatory brain activity in different frequencies subserves AV integration, pointing to a central role of (i) individual alpha frequency (IAF), which would determine the width of the cross-modal binding window; (ii) pre-/peri-stimulus theta oscillations, which would reflect the expectation of AV co-occurrence; (iii) post-stimulus oscillatory phase reset, which would temporally align the different unisensory signals. Here, we investigate the neural correlates of AV integration in children with ASD and typically developing (TD) peers, measuring electroencephalography during resting state and in an AV integration paradigm. As for neurotypical adults, AV integration dynamics in TD children could be predicted by the IAF measured at rest and by a modulation of anticipatory theta oscillations at single-trial level. Conversely, in ASD participants, AV integration/segregation was driven exclusively by the neural processing of the auditory stimulus and the consequent auditory-induced phase reset in visual regions, suggesting that a disproportionate elaboration of the auditory input could be the main factor characterizing atypical AV integration in autism.

Rhythmic TMS as a Feasible Tool to Uncover the Oscillatory Signatures of Audiovisual Integration

“…That is, rhTMS-driven entrainment could be deployed not only to up- or down-shift the alpha peak (i.e., IAF ± 1 or 2 Hz [ 38 , 42 ]) but also to recruit neuronal pools oscillating, for instance, within the theta (i.e., around 5 Hz) or beta (i.e., around 20 Hz) range. Akin to what aforementioned about alpha activity, this would allow us to causally probe how pivotal the contribution of these rhythms is to cross-modal operations, trailing empirical findings recently gathered (from occipital sensors) by means of both audio-visual and visuo-tactile paradigms [ 71 , 72 , 96 , 97 ].…”

Alpha-band frequency and temporal windows in perception: A review and living meta-analysis of 27 experiments (and counting)

“…This often appears as two dots “streaming” past each other, but if a “bounce-like” sound occurs around the time of dot coincidence, the dots tend to appear as though they bounced off of one another. Ronconi et al (2023) found a strong correlation between the probability of bounce percepts and IAF in 17 typically developing (TD) children (r = -0.85), indicating that bounce percepts decreased as IAF increased, suggesting smaller windows the higher one’s IAF. Interestingly, a second experiment with 17 children with autism spectrum disorder (ASD) found a much weaker relationship (r = -0.35).…”

Prolonged neural encoding of visual information in autism

“…Relatedly, another potential explanation may rely on the choice of including a smaller portion of trials with shorter SOA, with the aim to reduce potential prediction‐related issues (e.g., Ronconi et al, 2023), and more generally to induce a less strong expectation about the timing of target onset, avoiding a strong prediction about the timing of target appearance. At the theoretical level, there is the possibility that some of the differences between ASD and TD groups highlighted in the present study also arises due to the putative anomalies in forming predictions in individuals with ASD (e.g., Ronconi et al, 2023), although we could not directly probe this hypothesis in the current study, since the analyses were limited selectively to the 500 ms SOA condition, where enough trials were available to compute EEG data analysis.…”