Diagnosis and classification in autism.

The team studied 194 preschoolers who all showed social delays. They asked: do DSM and ICD autism labels match how kids actually cluster in the real world?

Each child got play-based tests and parent interviews. Stats grouped kids into natural social types, then checked if those types lined up with official diagnoses.

Two clear social patterns popped out. The "Aloof" group mostly fit classic autism. The "Active-but-Odd" group mapped to milder PDD labels.

In plain words, the manuals were right. Kids who ignored peers usually met full autism criteria. Kids who talked at peers, but oddly, got PDD-NOS or similar.

Tassé et al. (2013) later showed the same social split exists in typical college students. Aloof versus Active-but-Odd is not just a clinical thing—it is a spectrum that runs through everyone.

Cohen et al. (2018) pushed the idea younger. They found parents spot these social styles in 12-month-olds better than short clinic visits. Early pattern recognition starts at home, not in the office.

Shu et al. (2022) moved it bigger. They used parent surveys plus math to guess cognitive level in 20 000 SPARK kids. The 1996 two-group picture still helped the model sort severe from mild.

You can trust the DSM categories for preschool intake. If a child is aloof and non-verbal, plan for classic autism support. If the child chats but misses cues, think PDD-NOS and target social nuance. Use parent report early; it captures the same taxa the manuals describe.