Accuracy of initial diagnostic impressions of autism in toddlers and behaviors that inform these impressions.
Four quick behaviors seen in five minutes flag most toddler autism cases.
01Research in Context
What this study did
Doctors watched toddlers for just five minutes. They wrote their first guess: autism or not. Later they checked this guess against the final diagnosis. The team wanted to know which quick behaviors made doctors suspect autism.
What they found
The doctors’ first guess matched the final autism label about two-thirds of the time. They were perfect when they ruled autism out. Four signs guided them: back-and-forth social play, gestures and facial expressions, eye contact, and where the child put their attention.
How this fits with other research
Wieckowski et al. (2025) took the same idea online. Remote doctors also made fast calls with high accuracy. This shows the five-minute rule works even through a screen.
Gabriels et al. (2001) gave parents a twenty-three-item paper checklist. That tool and this quick look both speed toddler screening. One uses parent answers; the other uses clinician eyes.
van der Miesen et al. (2024) found that kids with spotty nonverbal skills often pass parent screens yet still have autism. The five-minute clinician look catches these children because it spots uneven eye contact and gestures in real time.
Why it matters
You can adopt a two-step plan. First, run a short parent screen like the M-CHAT. Second, if any doubt remains, watch for four behaviors during the first five minutes of play. Note social back-and-forth, gestures, eye contact, and shared focus. If all four feel off, fast-track for full evaluation. If they seem typical, still schedule a longer visit; some kids will need a second look.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →During intake play, set a five-minute timer and tally social reciprocity, nonverbal communication, eye contact, and shared attention—fast-track when three or more look atypical.
02At a glance
03Original abstract
Clinicians form initial impressions about a child's diagnosis based on behavioral features, but research has not yet identified specific behaviors to guide initial diagnostic impressions. Participants were toddlers (N = 55, mean age 22.9 months) from a multi-site early detection study, referred for concern for ASD due to screening or parent/provider concern. Within 5 min of meeting a child, clinicians noted ASD or non-ASD impression, confidence in impression, and behaviors that informed their impression. These clinicians also determined final diagnoses for each child. When a child's final diagnosis was ASD (n = 35), senior clinicians formed an initial impression of ASD in 22 cases (63%) but missed 13 cases (37%). When final diagnosis was non-ASD (n = 20), senior clinicians made an initial impression of non-ASD in all cases (100%). Results were similar among junior clinicians. Senior and junior clinicians used the same behaviors to form accurate impressions of ASD and non-ASD: social reciprocity, nonverbal communication, and eye contact. Senior clinicians additionally used focus of attention when forming accurate impressions of ASD and non-ASD; junior clinicians used this behavior only when forming accurate non-ASD impressions. Clinicians' initial impressions of ASD are very likely to be consistent with final diagnoses, but initial impressions of non-ASD need follow-up. Toddlers who show all four atypical behaviors (social reciprocity, nonverbal communication, eye contact, and focus of attention) might receive expedited ASD diagnoses. However, presence of apparently typical behaviors should not rule out ASD; for some children a longer evaluation is necessary to allow for more opportunities to observe subtle social behavior.
Autism research : official journal of the International Society for Autism Research, 2024 · doi:10.1001/jamanetworkopen.2022.9503