Social and non-social cueing of visuospatial attention in autism and typical development.

The team watched autistic and typically developing children look at pictures. A face on the screen looked left or right. The kids had to find a star that popped up on that side.

Cameras tracked every eye jump. The goal was to see if autistic kids followed the gaze cue as fast as the other kids.

Both groups found the star just as quickly. Autistic children did make more tiny eye jumps, but the cue still worked.

The extra jumps slowed them down a little, yet the gaze direction itself was not ignored.

Bigham et al. (2013) seems to disagree. They saw weaker reflexive gaze following in preschoolers with autism. The clash fades when you notice they tested younger kids and used a faster, reflex-only task.

Rombough et al. (2013) backs up the new result. They also found intact automatic cueing, but poorer performance when kids had to pick eyes over arrows.

Hou et al. (2024) adds a fresh layer. Their 2024 scan-path data show autistic children make more scattered eye moves, matching the extra saccades seen here.

You can stop teaching gaze following as a broken skill. Instead, cut visual clutter and give an extra beat for the child to settle their eyes. A calmer screen or table layout may speed up responses more than gaze drills.