Intranasal Oxytocin Enhances Perceptual Mechanisms for Voice-Identity Recognition.

Background: Individuals with autism spectrum disorder (ASD) have difficulties recognizing other people by the voice – a deficit associated with reduced responses in the right posterior superior temporal sulcus/ gyrus (pSTS/STG) (Schelinski et al., 2016, SCAN; Schelinski et al., 2017, Autism Res). The right pSTS/STG has been implicated in the perceptual analysis of voice-identity information (Maguiness et al., 2018, Neuropsychologia). Intranasal oxytocin can promote face recognition and modulate face-sensitive brain responses in ASD (Domes et al., 2013, Biol Psychiatry; Guastella et al., 2010, Biol Psychiatry).

Objectives: To investigate whether intranasal oxytocin can enhance the impaired voice-identity recognition and increase the right pSTS/STG responses in ASD.

Methods: Nineteen adults with ASD (ASD group) and nineteen typically developed (TD) controls (control group) participated in a randomized, double-blind, placebo-controlled, cross-over design study. The groups were pairwise matched on age, gender, handedness and intelligence quotient (IQ). All participants had normal hearing and were free of psychopharmacological medication. Participants with ASD had already received a formal clinical diagnosis of ASD (Asperger Syndrome/ childhood autism (1 male; verbal IQ = 109)) according to the diagnostic criteria of the International Classification of Diseases (ICD 10; World Health Organization, 2004). Within the study, they underwent additional clinical assessment using Autism Diagnostic Observation Schedule (ADOS, Lord et al., 2000, J Autism Dev Disord) conducted by researches with formal training.
Participants completed two sessions of a functional magnetic resonance imaging (fMRI) experiment on voice-identity recognition: after oxytocin and after placebo administration. The experiment included two conditions: voice-identity task and speech task. In both conditions, participants listened to blocks of neutral two-word sentences spoken by four male speakers. In the voice-identity task, participants matched the identity of each speaker to a target speaker. In the speech task, participants matched the content of each sentence to a target sentence. Targets were presented at the beginning of each block.
We calculated a three-way ANOVA: Substance (oxytocin, placebo) x Task (voice-identity, speech) x Group (ASD, control). Region of interest (ROI) was the right pSTS/STG, because it is known to show decreased BOLD response to voice-identity vs. speech recognition in ASD compared to TD controls (Schelinski et al., 2016, SCAN).

Results: Oxytocin did not have any effect on the behavioral performance in the voice-identity recognition experiment. For the ROI, we found a significant three-way interaction Substance x Task x Group (p < .05 FWE corrected). Post hoc t-tests revealed that oxytocin increased right pSTS/STG responses to voice-identity vs. speech in the control group, but not in the ASD group.

Conclusions: Our results suggest that oxytocin can enhance perceptual analysis of voice-identity features in the right pSTS/STG in TD individuals. This was not the case in ASD, at least not to the same extent as in TD population, which supports previous evidence that perception of voice identity is atypical in ASD.