Neural processing of social reciprocity in autism

Abstract

Social reciprocity and interpersonal synchrony implicitly mediate social interactions to facilitate natural exchanges. These processes are altered in autism, but it is unclear how such alterations manifest at the neural level during social interaction processing. Using task-based fMRI, we investigated the neural correlates of interpersonal synchrony during basic reciprocal interactions in a preregistered study. Participants communicated with a virtual partner by sending visual signals. Analyses showed comparable activation patterns and experienced synchrony ratings between autistic and non-autistic participants, as well as between interactions with virtual partners who had high or low synchronous responses. An exploratory whole brain analysis for the effect of task revealed significant activation of the inferior frontal gyrus, insular cortex, and anterior inferior parietal lobe; areas associated with cognitive control, rhythmic temporal coordination, and action observation. This activation was independent of the virtual partner's response synchrony and was similar for autistic and non-autistic participants. These results provide an initial look into the neural basis of processing social reciprocity in autism, particularly when individuals are part of an interaction, and hint that the neural processing of social reciprocity may be spared in autism when their partners' behavior is predictable.

Keywords: Autism; Interpersonal synchrony; Reciprocity; Social interactions; fMRI.

Fig. 1

Depiction of the sequence for each block beginning with the communication sequence within each 12-second trial and ending with an example of a rating question in German.

Fig. 2

(A) Mean and standard error of the experienced synchrony ratings for the low and high latency and dispersion conditions for each group. Based on the scale: 1 (‘not at all’) to 7 (‘very much’). (B) Mean and standard error of rapport ratings for the low and high latency and dispersion conditions for each group. Based on the scale: 1 (‘not at all’) to 7 (‘very much’). (C) Mean and standard error of the communication frequency for the low and high latency conditions for each group. (D) Mean and standard error of participants’ average response latencies for the low and high latency conditions for each group. Stars indicate a significant effect of latency.

Fig. 3

Results of the whole brain analysis of the entire sample pooled across conditions. A t-contrast is shown highlighting surviving clusters (blue) from a non-parametric, TFCE approach considering p < 0.05. Participants showed significant activation in the bilateral inferior frontal gyrus, insular cortex, right anterior inferior parietal lobe, left motor cortex (precentral and postcentral gyri) extending into the anterior inferior parietal lobe, and the supplementary motor area (not pictured). The hypothesized ROI mask is overlaid (yellow), highlighting parts of the surviving cluster that fall within the IPL ROI (green, arrow). Surviving clusters for the effect of task overlaid on an uncorrected t statistic map are depicted in Supplementary Information S10. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)