The development of facial gender categorization in individuals with and without autism: the impact of typicality.

The team tracked how well autistic and typical people sort faces by gender. They tested kids, teens, and adults. Each person saw typical and odd-looking faces on a screen and picked "boy" or "girl" as fast as they could.

The study ran in a quiet lab room. No extra teaching or rewards were given. The goal was to see if autistic learners ever catch up to peers.

Autistic people were slower and less accurate at every age. The gap stayed wide for faces that looked unusual. Even high-functioning adults with autism never reached typical speed or accuracy.

Gastgeb et al. (2011) saw the same lag one year earlier. Their autistic adults could not build a mental "average" face. The two studies line up: weak prototypes lead to shaky gender calls.

Van der Donck et al. (2023) seems to disagree. They found normal brain waves when autistic adults viewed fast face changes. The trick is in the task. Quick flashes tap low-level vision; slow gender choices need stored face rules. Different measures, different story—no real clash.

Cook et al. (2014) adds another twist. Their autistic adults showed normal "after-effects" when faces were stretched. Adaptation and gender sorting use separate brain paths. One can work while the other falters.

If your client needs to read faces for job or social groups, do not assume they will pick it up on their own. Add clear visuals: color-coded photos, rule cards, or video models. Drill with odd-looking faces too—long hair on men, short hair on women—so the skill holds in real crowds.