Social Cognition in Adults with Autism Spectrum Disorder: Sex Differences and Aging Correlates

Background: In the United States, approximately 50,000 teens with Autism Spectrum Disorder (ASD) age into adulthood every year (Shattuck et al., 2012). A hallmark symptom of ASD includes pronounced difficulties in social interactions and verbal and nonverbal communication (Lai, Lombardo, & Baron-Cohen, 2014). These social cognition difficulties consist of difficulties interpreting social cues, employing appropriate adaptive behavioral responses in various social contexts, as well as the ability to interpret emotions and mental states of others, known as theory of mind (TOM; Premack & Woodruff, 1978). In neurotypical (NT) adults, women perform better on social cognition tasks and difficulties become more prevalent with age, however little is known how sex differences and aging may impact social cognition in adults with ASD (Carstensen, Fung, & Charles 2003).

Objective: This research intended to characterize the influence of sex and age on social cognition in adults with ASD using an adult sample. We hypothesized Reading the Mind in the Eyes (RME) scores would be lower in adults with ASD, with a stronger relationship between decreasing performance aging effects compared to NTs. Additionally, we hypothesized deficits would be more severe in in males with ASD compared to females with ASD.

Methods: The RME task was administered to 181 adults to quantify ToM abilities. The participants consisted of 100 adults with ASD (69 males, 32 females; age range: 18-71, mean=39.45±1.613) and matched 81 NT adults (47 males, 34 females; age range: 18-70, mean=41.51±1.883). Multiple regression analyses examined interactions between diagnosis and age, diagnosis and sex, and diagnosis by age by sex. Exploratory within group analyses assessed 1) sex differences using ANCOVA, and 2) associations with age using Pearson correlation in SPSS.

Results: We found that NT adults performed better on the RME task than adults with ASD. Worse performance on the RME task correlated with greater age for the NT, but not ASD. Additionally, no influence of sex on RME scores was identified.

Discussion: These results are consistent with other studies indicate social cognition deficits in adults with ASD compared to NT adults. Additionally, we replicated findings that suggest ToM performance declines with age in NT adults. Fewer social relationships, smaller social networks, and reduced social engagement have been associated with aging in both NTs and individuals with ASD (Pratt & Norris, 1994). However, our cross-sectional sample suggests ToM abilities may not decline with age in adults with ASD as hypothesized. Longitudinal studies are needed to corroborate these findings. Further developments in this line of research may inform novel interventions tailored toward the growing population of adults with ASD. Ultimately, our research aims to improve quality of life across the lifespan for an already vulnerable population.