Newly published research by a team led by BBRF grantees finds distinct age-related patterns in children and adolescents with autism spectrum disorder (ASD) that may help explain challenges in social communication that characterize the disorder and how they change over time.

The new research sheds light on the neurobiological signatures in autistic individuals of what developmental experts call social reorientation—the process that in typically developing (TD) young people is marked by a shift in social orientation from parents to friends and peers who are not members of one’s own family.

This transition, the researchers note, “is thought to reflect an evolutionarily adaptive process that prepares children for independence from parents and caregivers, and is supported by heightened neural sensitivity to social stimuli.” Understanding how social reorientation differs in autism has important implications for intervention and support of those with an ASD diagnosis.

The team was led by Daniel A. Abrams, Ph.D., a 2018 BBRF Young Investigator at Stanford University, whose grant project pertained directly to questions explored in the new research. The team’s senior member was Vinod Menon, Ph.D., a 1998 BBRF Young Investigator who is also at Stanford.

They point out that the behavioral manifestations of autism evolve throughout the lifespan. This applies to social communication challenges, which in certain aspects present differently in children compared with adolescents—something that the new findings help explain.

A prominent signature of social orienting differences in children with ASD is reduced engagement with the voices in their environment. One of the earliest clinical signs of ASD is not consistently orienting to parents and caregivers when their name is called. But little is known about age-related differences in voice processing in children compared with adolescents with ASD.  Does sensitivity to social vs. nonsocial signals that one sees in ASD differ between childhood and adolescence? What about the typical age-related shift from mother’s voice toward nonfamilial voices—a neural pattern theorized to accompany increasing orientation toward peers and other social partners?

To better understand these questions, the team turned to functional neuroimaging of parts of the brain involved in human voice processing. A recent study of typically developing young people ages 7-17 revealed increased age-related neural activity in response to both mother’s voice and nonfamilial voices across multiple social brain systems. This increased activity was consistent with a general increase in the young people’s sensitivity to social signals.

One objective of the new study was to look at these patterns in both children and adolescents with autism vs. matched controls. The study cohort consisted of 79 young people ages 7-17 (average age about 12), 39 of whom were diagnosed with ASD. The ASD group was 85% male; the control group was 73% male. IQ was comparable in the two groups (mean IQ in ASD group, 112; controls, 116). The team presented each participant with brief samples of nonfamilial voices, their mother’s voice, and nonsocial control sounds while a functional MRI (fMRI) scan was being made.

Participants with ASD showed atypical age-related patterns of brain activity and connectivity in the voice processing network in response to both mother’s voice and nonfamilial voices, compared to nonsocial stimuli. These differences—age-related decreases in neural engagement in ASD participants compared with increases in typically developing control participants—were manifest in parts of the brain that process salience (the relevance of a sound or other stimulus) and reward; the default mode network (a key network for social cognition and valuation); and frontoparietal regions.

A second important finding concerned the relative balance of functional brain connectivity in response to nonfamilial voices compared with mother’s voice. Whereas older typically developing participants showed greater functional connectivity for nonfamilial voices relative to mother’s voice, older autistic participants showed the opposite age-related pattern: lower functional connectivity for nonfamilial voices relative to mother’s voice. This pattern was most pronounced among autistic participants with more severe social-communication challenges.

“Together, findings suggest that aberrant organization of voice-selective connectivity in the pSTS [a part of the brain’s superior temporal sulcus, selective for voices vs. other kinds of sounds], and broader disruptions in reward, salience, default mode, and frontoparietal circuitry, represent a key feature associated with atypical social neurodevelopment in adolescents with ASD,” the team reported in Proceedings of the National Academy of Sciences (PNAS).

The researchers theorized that the age-related decreases in vocal engagement in ASD that they observed “may be associated with differences in experience-dependent plasticity and the development of higher-order social cognitive skills, including understanding feelings and perspectives of others.” These are typically strengthened, not weakened, as adolescence proceeds.

The observed patterns in ASD participants are consistent with an interruption and reversal of the typical transition from parents toward nonfamilial social partners, the researchers said. To explain the lower functional connectivity in response to nonfamilial voices relative to mother’s voice among older autistic participants, the team proposed two possible factors. One is experiential differences: the pattern may reflect reduced exposure to and interaction with nonfamily social partners. Another possibility is neurobiological differences, “including atypical social brain circuitry.”

Since the current study did not directly measure social interaction, peer engagement, or the language environment of the participants, it could not distinguish between these possible explanations. The team said that in future studies, it will be vital to incorporate detailed measures of participants’ social experiences. They also cautioned that they are unable right now to determine if the observed age-related divergence in vocal engagement reflects voice-specific processing differences or broader alterations in social orienting and reward-related systems in youths with ASD.

One hopeful perspective on the new findings, the researchers noted, is the fact that many of the brain systems that showed divergent age-related patterns in ASD are involved in voice processing, and “are among the last regions in the human brain to mature,” with a developmental trajectory that extends through adolescence into early adulthood. This extended maturational window can be seen as a vulnerability for adolescent psychopathology, but also “an opportunity for continued refinement.”

The current findings also suggest directions for future intervention research. The researchers proposed that future interventions designed to strengthen pSTS-centered voice processing brain circuitry “may help buffer social brain networks from the combined social and hormonal challenges of adolescent development.” Adolescents also might benefit from interventions that target social motivations and reward, which potentially can modify atypical neurodevelopment patterns.