Is Poor Eye Contact Really a Reliable Sign of Autism?

SHANGHAI — “Never looked at any person’s face.” That’s how Leo Kanner first described children with autism in 1943, establishing a diagnostic marker that has persisted for eight decades. Ask almost anyone to name a sign of autism, and they’ll likely mention poor eye contact. This behavior has become so central to our understanding of autism that it’s often one of the first red flags parents and pediatricians look for. But new research suggests we might have gotten it wrong.

Scientists at the Shanghai Institute of Artificial Intelligence for Education at East China Normal University recently published findings in the ECNU Review of Education that challenge this foundational belief. Their research shows that during play, children with autism look at toys and adults’ faces in nearly identical patterns as typically developing children. Both groups spend most of their time (60-80%) looking at toys, while only briefly glancing at adults’ faces (6-14%).

This raises a question that might transform how we think about autism diagnosis: What if avoiding eye contact isn’t actually an autism-specific behavior at all?

Natural Play Reveals Surprising Gaze Patterns

The team’s discovery emerged from observing children in more natural settings than those typically used in autism research. Traditional autism assessments often take place in sterile clinical environments where children are directly prompted to engage with an examiner. However, the researchers created a specialized laboratory designed to feel comfortable for children while still allowing precise measurement of their behaviors.

In this more natural play environment, they found something surprising. Whether a child was autistic, typically developing, or had developmental delays, their visual attention patterns were statistically indistinguishable. All children mostly watched the toys they were playing with and only occasionally looked up at the adult’s face.

If limited eye contact during play is normal for all children, then using it as a key diagnostic marker could lead to misidentification or missed diagnoses for autism.

This aligns with another recent study by Yurkovic-Harding and colleagues, published in Current Biology. This study used head-mounted eye trackers to record children’s gaze patterns during play. Their research found both autistic and typically developing children looked at their parents’ faces only about 1% of the time. Instead, children in both groups focused significantly on their parents‘ hands.

Children might establish joint attention, the shared focus between two people on an object or event, through different means than adults expect. While adults often rely on eye contact to establish a connection, children might be tracking hand movements and gestures as their primary way of maintaining social coordination during play. Joint attention has long been considered critical for child development, linked to language acquisition, social growth, and cognitive advancement.

This doesn’t mean there are no differences in how autistic children engage socially. Rather, it indicates that the differences might manifest in more subtle ways than simply avoiding eye contact. The researchers noted that children with autism might use different strategies to establish joint attention, such as monitoring an adult’s hands rather than their face.

The last time you played with a child, did they stare into your eyes the whole time, or were they focused on the toy or activity? Most likely, their attention was primarily on what you were doing together rather than on maintaining eye contact. This normal behavior pattern appears consistent across neurotypical and autistic children alike.

Transforming Autism Diagnosis and Support

For parents concerned about autism, these findings offer reassurance that limited eye contact during play isn’t necessarily cause for alarm. Children naturally focus on objects rather than faces during play activities.

For professionals, relying too heavily on eye contact as a diagnostic marker could lead to misdiagnosis, especially when assessments occur in artificial clinical environments rather than natural play settings.

What other aspects of autism might be misunderstood due to contextual factors or incomplete observations? If such a fundamental characteristic as eye contact has potentially been misinterpreted, what other autism traits deserve reconsideration?

While autism remains a distinctive neurological condition with real differences in social communication, sensory processing, and behavior patterns, this research refines our understanding of how those differences actually manifest in everyday contexts. The findings suggest focusing more on how autistic children engage in joint attention through alternative means rather than simply noting a perceived lack of eye contact.

We may need to update our understanding of autism’s core symptoms and the behavioral cues we use to identify it, particularly during early childhood when accurate diagnosis is crucial for effective support.