Early markers of autism spectrum disorders in infants and toddlers prospectively identified in the Social Attention and Communication Study.

Josephine and her team watched 104 babies who had an older sibling with autism. These babies were already at higher risk.

They filmed each child at 12, 18 and 24 months during normal play with a parent. They counted four simple acts: looking at Mom’s eyes, pointing at toys, showing toys, and pretend play like feeding a doll.

They kept notes until each child turned two. Then they checked who got an autism diagnosis and who did not.

Kids who rarely looked at eyes or pointed by 18 months were almost always diagnosed with ASD by 24 months.

If the child also failed to show toys or play pretend, the forecast got even stronger. These four tiny acts together gave a clear red flag before age two.

Rutherford et al. (2007) already showed that joint attention leads to later pretend-play gains. Josephine’s work moves the clock earlier: lack of both skills in toddlerhood now predicts the diagnosis itself.

Anbar et al. (2024) followed similar kids into grade school. They found that the same early joint-attention scores keep shaping pragmatic language years later. The 2013 markers are not just quick flags—they echo through childhood.

Begeer et al. (2006) seems to disagree: they saw typical face looking when kids were told to care about emotions. But their task used short lab clips, not free play. The contradiction fades once you see the setting difference—natural play reveals the everyday gap.

You can spot autism risk with items you already track: eye contact, pointing, showing, pretend. No extra kit is needed. During intake, watch a 15-minute parent-toddler play. If three of the four acts are missing at 18 months, refer for full evaluation and start joint-attention intervention right away. Acting one year earlier gives the child a head start and fits most insurance early-age requirements.