Response to joint attention in toddlers at risk for autism spectrum disorder: a prospective study.

Sullivan et al. (2007) watched toddlers who had an older sibling with autism. These kids are at high risk. The team scored how well each child responded to joint attention (RJA) at 14 months and again at 24 months. Later they checked which children received an autism diagnosis.

The study had no control group. It simply tracked the same children over time.

Most toddlers who were later diagnosed with autism showed almost no growth in RJA. Their scores stayed low and flat. A low RJA score at 14 months was a red flag for later ASD.

Bottema-Beutel (2016) pooled many studies and found the same link: weak RJA predicts language delays in ASD. The 2007 toddlers fit that bigger picture.

Piatti et al. (2024) added brain data. They showed ASD toddlers have less right-temporal activity during RJA tasks. The flat behavior seen in 2007 now has a neural signature.

Neuringer et al. (2007) seems to disagree. They found infant siblings could reach typical RJA if an extra point cue was added. The difference is age and task. G studied younger babies in a lab cue game; Michelle watched natural behavior over a full year. Both show the same kids start behind, but G proves the gap can close with the right prompt.

Screen RJA at the 12-15 month visit. A low score tells you to start intervention fast, because growth will be slow. Pair the screen with a simple cue package: name, point, and eye-shift together. Track progress monthly; flat lines need dosage boosts or added play-based JA programs like those in Lawton et al. (2012).