Autism & Developmental

Recognising 'social' and 'non-social' emotions in self and others: a study of autism.

Williams et al. (2010) · Autism : the international journal of research and practice 2010
★ The Verdict

Standard emotion-recognition tests can look normal in autistic adults—check process, not just score.

✓ Read this if BCBAs writing adult assessments or social-skills programs
✗ Skip if Clinicians only serving autistic children under ten

01Research in Context

01

What this study did

Williams et al. (2010) asked autistic adults and adults with learning disabilities to name emotions. They used photos, voices, and stories. The team wanted to see if autistic adults were worse at social feelings like guilt or pride.

Each person also rated their own feelings after tasks. The study matched groups by IQ so skill level was even.

02

What they found

Both groups scored the same on every task. They named happy, sad, guilt, and pride cues equally well. Self-ratings also looked the same.

The result clashes with the old idea that autism always brings emotion-recognition deficits.

03

How this fits with other research

Perez et al. (2015) found autistic children scored lower than typical peers on emotion recognition. David used adults, M used kids. The gap may close with age or extra years of practice.

Arwert et al. (2020) extends the story. Autistic adults read faces fine when cues were clear. They only slipped when scenes were fuzzy and they had to weigh many hints at once.

Wehman et al. (1989) saw social-cognition gaps in autistic youth matched for mental age. David found no gap in adults matched for IQ. Better test tools, wider life experience, or different tasks may explain the change.

04

Why it matters

A flat score on a standard emotion test does not rule out autism in adults. Probe how they reach the answer: do they need extra time, rules, or repeated practice? Build goals around flexible cue use in messy, real-life settings rather than drilling basic labels.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Ask clients to explain how they knew the face was 'surprised' and note any rule-based or slow strategies.

02At a glance

Intervention
not applicable
Design
quasi experimental
Sample size
42
Population
autism spectrum disorder, intellectual disability
Finding
null

03Original abstract

Studies of emotion processing in autism have produced mixed results, with fewer studies observing autism-specific deficits than might be imagined. In the current study, 21 individuals with autism and 21 age- and ability-matched, learning disabled comparison participants were tested for their ability to (a) recognise, in others, expressions of 'social' emotions (e.g., embarrassment) and 'non-social' emotions (e.g., happiness) and; (b) report their own previous experiences of each of these emotions. In line with predictions, amongst both groups of participants, social emotions were more difficult to recognise and report than non-social emotions. Also amongst both groups, the ability to report social emotion-experience was significantly associated with the ability to recognise social emotions in others, independent of age and verbal ability. However, contrary to predictions, no between-group differences in levels or patterns of performance on the experimental tasks were observed. In light of previous research, these results suggest either that emotion-processing is not as specifically impaired in autism as is traditionally thought to be the case, or that individuals with autism are implementing compensatory strategies to succeed on experimental tasks in the absence of emotion-processing competence.

Autism : the international journal of research and practice, 2010 · doi:10.1177/1362361309344849