Abnormal use of facial information in high-functioning autism.

Kleinert et al. (2007) watched how autistic adults read faces.

They tracked eye movements while people judged happy, sad, angry and fearful photos.

The team asked: do adults with autism look at the same parts of a face as typical adults?

Autistic adults spent less time on the eyes and more time on the mouth.

This odd pattern hurt their ability to name the correct emotion.

Poor eye use explained most of their emotion-matching errors.

Ma et al. (2021) pooled 144 eye-tracking papers and confirmed the same eye-avoidance pattern across ages and cultures.

Matson et al. (2008) repeated the lab task the next year and again linked less eye time to missing genuine smiles.

McLennan et al. (2008) seems to disagree: they saw no eye-look difference. The catch is they used only simple happy and angry faces; when emotions are easy, gaze gaps shrink.

Tell learners to look at the eyes first. A quick prompt like "check the eye region" before each emotion drill can boost accuracy. Pair the prompt with photos that show fear and surprise, where eye cues matter most. Track if the learner’s eye tracker or your own data sheet shows longer eye fixations after the prompt.