Can children with autism read emotions from the eyes? The eyes test revisited.
Simplified eye-reading tests reveal hidden strength in kids with autism.
01Research in Context
What this study did
Franco et al. (2014) asked kids with and without autism to read feelings from eye photos. They tried two test styles: one word plus two pictures, or two words plus one picture.
All children did both styles. The team watched which version felt easier for each group.
What they found
Kids with autism scored higher on the simpler test: one word and two pictures. Both groups got better as they grew older.
The hard test hid their skill. The easy test let it show.
How this fits with other research
Evers et al. (2015) saw small deficits when kids watched moving faces. Fabia saw gains when only still eyes were shown and the task was short. The clash fades once you see motion adds load.
Sherwell et al. (2014) moved the same eye test to adults and still found trouble. Fabia’s child-friendly tweak proves the format, not the skill, often blocks success.
Song et al. (2018) later showed ASD kids need stronger eye cues for anger, fear, disgust. Fabia’s simpler layout may give that extra clarity without calling it intensity.
Why it matters
If you test eye-reading, strip the task. One label plus two clear photos beats a crowded array. You will see what the child truly knows, plan sharper social goals, and avoid false deficit labels.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Swap dense emotion arrays for one-word/two-picture eye cards during intake.
02At a glance
03Original abstract
This study aimed to test two new, simplified tasks related to the eye-test, targeting children with autism spectrum disorder (ASD) and typically developing controls (TD). Test-1 assessed the recognition of emotion/mental states with displays using one word and two eye-pictures, whereas Test-2 presented displays using two words and one eye-picture. Black and white photographs of children were used as materials. A cross-cultural study (Caucasian/East-Asian) with adults was initially carried out to verify generalizability across different ethnic groups. Cross-sectional trajectory analyses were used to compare emotion recognition from the eyes in the two tests. Trajectories were constructed linking performance on both tests either to chronological age or to different measures of mental age (receptive vocabulary based on the BPVS, CARS or ASQ for the ASD group). Performance improved with chronological age in both the ASD and TD groups of children. However, performance in Test-1 was significantly superior in children with ASD, who showed delayed onset and slower rate of improvement than TD children in Test-2. In both the ASD and TD groups the lowest error rate was recorded for the item 'anger', suggesting that threat-detection cue mechanisms may be intact in autism. In general, all children showed good performance on our novel tests, thus making them good candidates for assessing younger children and those with lower general abilities.
Research in developmental disabilities, 2014 · doi:10.1016/j.ridd.2014.01.037