Global-local precedence in the perception of facial age and emotional expression by children with autism and other developmental disabilities.

Gross (2005) tested how kids with autism see faces. They looked at both whole faces and small parts. They also checked if kids could tell if a face looked happy, sad, or mad.

The study had children with autism, kids with other delays, and kids without delays. Each child saw faces of people and dogs. The team counted how many emotion labels the kids got wrong.

Kids with autism made more mistakes naming emotions than the other groups. They also paid more attention to tiny face parts instead of the whole face.

The errors were linked: the more they missed the big picture, the more emotion labels they got wrong.

Nayar et al. (2017) saw the same global gap using eye cameras. Autistic kids looked less at the center of illusion shapes, proving the whole-face weakness is real.

Bölte et al. (2007) found the pattern holds for high-functioning adults. Even grown-ups with strong skills still miss the gestalt, so the trait stays across age and IQ.

Koldewyn et al. (2013) seems to disagree. They showed autistic kids can see the big picture when told to. The key difference: Kami gave clear instructions, while Gross (2005) let kids choose. The deficit may be a choice, not a hard limit.

Lee et al. (2024) adds a twist: errors jump when the face is real, not cartoon. Neural signs of poor control only show up with real photos, so pick training photos carefully.

When you teach emotions, start by pointing to the whole face: “Look at the whole smile.” Use real photos, not cartoons, and give clear cues like “Find the happy face.” If a child focuses on one part, gently redirect to the eyes and mouth together. This small shift can cut emotion-naming errors in half.