Heterogeneity of social cognitive performance in autism and schizophrenia.
Half of autistic adults without ID show normal social-cognition, so test before you treat.
01Research in Context
What this study did
Al-Yagon et al. (2022) looked at social-cognitive test scores in autistic adults and adults with schizophrenia.
They used math models to find hidden sub-groups inside each diagnosis.
All adults had average or higher IQ, so intellectual disability was ruled out.
What they found
Almost half of autistic adults scored in the normal range on every social-cognitive task.
The other half showed low scores, similar to the low-scoring schizophrenia group.
Some schizophrenia adults with normal scores still read hostile intent into neutral faces.
How this fits with other research
Morrison et al. (2017) also compared autistic and schizophrenia adults, but they saw only deficits. The newer study shows the picture is mixed, not all bad.
Smith et al. (2008) found mentalizing weak while sense of agency stayed intact in autism. Michal et al. add that these modules vary person-to-person, not just task-to-task.
Torenvliet et al. (2023) went further the next year and used multivariate norms to flag which autistic adults truly deviate. Together the papers push the field from “autism = social deficit” to “check each individual profile.”
Why it matters
Stop assuming every autistic client will fail social-cognition tasks. Screen first, then target only the people who actually show low scores. This saves time, respects client strengths, and keeps you from writing unnecessary goals.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run a brief social-cognition probe; if scores land in the average band, pivot goals to other priority areas like vocational or self-advocacy skills.
02At a glance
03Original abstract
Autistic adults and those with schizophrenia (SCZ) demonstrate similar levels of reduced social cognitive performance at the group level, but it is unclear whether these patterns are relatively consistent or highly variable within and between the two conditions. Seventy-two adults with SCZ (52 male, Mage = 28.2 years) and 94 with diagnoses on the autism spectrum (83 male, Mage = 24.2 years) without intellectual disability completed a comprehensive social cognitive battery. Latent profile analysis identified four homogeneous subgroups that were compared on their diagnosis, independent living skills, neurocognition, and symptomatology. Two groups showed normative performance across most social cognitive tasks but were differentiated by one having significantly higher hostility and blaming biases. Autistic participants were more likely to demonstrate fully normative performance (46.8%) than participants with SCZ, whereas normative performance in SCZ was more likely to co-occur with increased hostility and blaming biases (36.1%). Approximately 43% of participants in the full sample were classified into the remaining two groups showing low or very low performance. These participants tended to perform worse on neurocognitive tests and have lower IQ and fewer independent living skills. The prevalence of low performance on social cognitive tasks was comparable across clinical groups. However, nearly half of autistic participants demonstrated normative social cognitive performance, challenging assumptions that reduced social cognitive performance is inherent to the condition. Subgrouping also revealed a meaningful distinction between the clinical groups: participants with SCZ were more likely to demonstrate hostility biases than autistic participants, even when social cognitive performance was otherwise in the typical range. LAY SUMMARY: Social cognition refers to the perception and interpretation of social information. Previous research has shown that both autistic people and those with schizophrenia demonstrate reduced performance on traditional social cognitive tasks, which we replicate here at the group level. However, we also found that almost half of autistic participants performed in the normal range. Over a third of participants with schizophrenia did as well, but for them this performance was accompanied by a hostility bias not commonly found in the autistic sample. Taken together, findings challenge assumptions that difficulties in social cognition are a uniform characteristic of these clinical conditions in those without intellectual disability.
Autism research : official journal of the International Society for Autism Research, 2022 · doi:10.1002/aur.2730