Joint attention difficulties in autistic adults: An interactive eye-tracking study.

Caruana et al. (2018) watched autistic and non-autistic adults play a live computer game. The screen showed a face that looked at one of four objects. Players had to follow the face’s gaze and click the correct object.

Eye-tracking cameras recorded where each person looked. The team counted how many trials each group finished and how fast they shifted their eyes.

Autistic adults finished fewer trials and took longer to look at the right object on the first try. After a short practice block, their speed caught up, but they still missed more trials overall.

The gaps were small—no one failed every trial—but they were steady enough to show up in the data.

Fletcher-Watson et al. (2008) seems to disagree. They found autistic adults spotted eye-gaze changes just as fast as non-autistic adults. The key difference is task: Sue used quick picture swaps; Nathan used live joint-attention trials that required timing and turn-taking. Static gaze cues look fine; interactive gaze sharing still lags.

Geurts et al. (2008) and Neef et al. (1986) set the stage. They showed joint-attention problems start in babyhood and are autism-specific, not just a side effect of language delay. Nathan’s adult data prove the subtle gap never fully disappears.

Pitetti et al. (2007) gives hope. Toddlers who received parent coaching improved joint attention. Early practice helps, but Nathan reminds us to keep checking the skill across the lifespan.

If you run social-skills groups for adults, add extra warm-up trials that use shared gaze. Keep instructions clear and allow a few practice rounds—speed picks up quickly. When you write goals, track accuracy and latency separately; both tell you something. Finally, don’t assume good performance on simple eye-gaze tests means joint attention is solid—test it in real back-and-forth situations.