The dual neural effects of oxytocin in autistic youth: results from a randomized trial

Abstract

Recent discoveries have highlighted the effects of oxytocin (OT) on social behavior and perception among autistic individuals. However, a gap persists in the literature regarding the potential effects of OT and the neural temporal dynamics due to OT administration. We explored the effect of OT on autistic individuals using magnetoencephalography (MEG), focusing on M100, M170, and M250, social perception-related components that tend to show atypical patterns in autistic individuals. Twenty-five autistic adolescents participated in this randomized, double-blind MEG study. Autistic individuals arrived at the lab twice and received an acute dose of intranasal OT or placebo in each session. During the scans, participants were asked to identify pictures of social and non-social stimuli. Additionally, 23 typically developing (TD) adolescents performed the same task in the MEG as a benchmark that allowed us to better characterize neural regions of interest and behavioral results for this age group in this task. A source-model beamformer analysis revealed that OT enhanced neural activity for social stimuli in frontal regions during M170. Additionally, in each of the preselected time windows, OT increased activation in the left hemisphere, regardless of the content of the presented stimuli. We suggest that OT increased the processing of social stimuli through two separate mechanisms. First, OT increased neural activity in a nonspecific manner, allowing increased allocation of attention toward the stimuli. Second, OT enhanced M170 activity in frontal regions only in response to social stimuli. These results reveal the temporal dynamics of the effects of OT on the early stages of social and non-social perception in autistic adolescents.Trial registration: This study was a part of a project registered as clinical trial October 27th, 2021. ClinicalTrials.gov Identifier: NCT05096676.

Conflict of interest statement

The authors declare no competing interests.

© 2022. The Author(s).

Figures

Figure 1

Behavioral paradigm—an example of two separate trials. The experimental procedure was adapted from the 'Reading the Mind in the Eyes' Task (RMET). It consisted of eight blocks—half contained pictures of faces (social condition, see example above the line) and the other half contained pictures of vehicles (non-social condition, see example below the line). The task lasted approximately 13 min in total. The order of the blocks and the order of the pictures inside each block were randomized and counterbalanced across participants. The three dots at the end of each row mark the continuity of the block.

Figure 2

Evoked responses locked to social and non-social stimuli. (A) Root mean square (RMS) responses to social and non-social stimuli averaged for all experimental groups. (B) Scalp topographies showing the differences between social and non-social neural responses in each component (M100 – left, M170 -middle, M250 -right). Black dots represent MEG sensors with a significant difference in the average neural response between social and non-social stimuli. (C) RMS responses to social and non-social stimuli for each study group separately: ASD-OT (top panel), ASD-PL (middle panel), and TD (bottom panel).

Figure 3

Source-level interaction between OT and PL sessions in the M170 time window. OT increased neural activation in medial frontal regions in response to social stimuli compared to non-social pictures. (A) Significant differences in neural activity between social and non-social stimuli were identified based on the t-values. (B) The average power of the significant cluster. Lines represent the average activity of the ASD group in each session. Dots represent individual data. (C) Source-level activity from a virtual sensor in the significant cluster. The horizontal black line represents the significant time window (p < .05, corrected for multiple comparisons). Straight lines represent neural activity in social trials, the dashed line represents activity in non-social trials. The color-coding is the same as in panel B.

Figure 4

Main effect of OT is source-level. In general, compared to PL, OT increased neural activation in frontal, temporal, and occipital regions, regardless of the content (social/non-social) of the presented stimuli. Top panel–significant differences in each of the preselected components of interest between OT and PL sessions. Significant differences were observed in the left hemisphere only. Bottom panel–average power in the significant clusters for each group separately. Vertical lines represent mean values ±1 SE.

Figure 5

Neural activity in typical social-related ROIs of the experimental groups. These ROIs were pooled from the M170 time windows of the TD group. In this period, higher neural activity was observed in response to social stimuli than non-social stimuli. The middle section shows posterior and bottom views of social-related ROIs in the TD study. Color coding represents t-values. These regions represent neural regions that were more active in social trials than in non-social trials. The left and right panels show the average neural activity in these clusters (left and right, respectively) in each group separately.

Figure 6

Behavioral results for accuracy rates and reaction times. After OT administration, the accuracy rates of the autistic individuals more closely resembled those of the TD study group. Analyses also revealed that all individuals presented higher ACC and lower RTs in the non-social trials. Accuracy (top panel) and RT (bottom panel) in response to social (left side) and non-social (right side) stimuli in the ASD-OT, ASD-PL, and control groups. Black lines represent the mean values of the groups. Vertical lines represent 1 SEM. Dots represent the individual data in the group.