Young autistic children may process social cues more than previously believed, indicating potential for early intervention

New studies challenge the traditional view of autism as a lack of social attention, revealing that very young autistic children are sensitive to others’ viewpoints and may benefit from targeted training to improve social understanding.

A new paper in Autism Research suggests that very young autistic children do register what other people can see, and may even be especially affected when another viewpoint clashes with their own, a finding that complicates the long-running idea that autism is marked by a simple lack of social attention. The result arrives only weeks after another Autism Research trial reported that structured perspective-taking sessions could lift theory-of-mind performance in autistic children, hinting that these skills may be both present early and open to training.

The new study examined spontaneous visual perspective-taking, the quick tendency to take account of another person’s line of sight even when it is not explicitly required. To do that, the researchers used a simplified version of the classic dot-perspective task, arguing that older versions ask too much of preschool children. They first checked whether the adapted task worked in a preschool sample before comparing 34 autistic children with 29 non-autistic children matched for age and IQ. The public versions are not fully aligned on one detail: Wiley’s article page gives the preschool validation sample as 59 children, while the University of Bristol repository entry for the accepted manuscript lists 61. That effort to make the task more manageable for young children echoes earlier developmental work that used block-based, visuospatial activities as an alternative to language-heavy social-cognition exercises.

What matters most is that both groups showed interference from another person’s perspective as well as from their own. In everyday terms, the children were slowed not only by what they themselves could see, but also by what a second observer could see, suggesting that the other viewpoint was being processed even when it was not the target of the task. The authors said this points to “preserved or even heightened sensitivity to social information”, while also suggesting that autistic children may have greater difficulty suppressing their own perspective once two viewpoints compete. That is a notable conclusion because a July paper led by Mi Tian pointed in the same broad direction: if perspective coordination can be strengthened, theory of mind may improve with it.

According to the PubMed record for Tian’s randomised controlled trial, 66 autistic children were initially recruited and 48 completed the study, with a mean age of 5.75 years. The final analysis compared 16 children given visual perspective-taking training, 16 given thought-bubble training and 16 in a control group. Each active programme consisted of 12 sessions lasting 30 minutes, and both training groups improved on visual perspective-taking and theory-of-mind measures compared with control. The abstract describes the study as the first direct test of table-block visual perspective-taking training as a route to better theory-of-mind performance in autistic children.

Trial records suggest that paper sits within a broader research programme than the journal abstract alone makes clear. ClinicalTrials.gov says Nanjing Normal University planned a four-arm, randomised, parallel-group study with an estimated 100 participants aged three to eight, covering visual perspective-taking training, block building, thought-bubble training and treatment as usual. The registry says investigators wanted not only to test whether theory-of-mind gains endured over time, but also to use functional near-infrared spectroscopy alongside an oddball face-periodic visual stimulation paradigm to examine whether any behavioural changes were accompanied by shifts in facial-emotion recognition.

Those public records are not entirely in step on the current status of the wider project. ClinicalTrials.gov says the study was last verified in March 2024 and describes its latest known status as recruiting, while the CTV Veeva listing currently labels it “Enrolling”. TrialScreen carries the same NCT06196060 identifier on a participant-facing trial page, reinforcing that the intervention programme was registered beyond the journal literature even though that listing adds little extra methodological detail.

A companion paper from the same line of work suggests the researchers have also been testing a more spatial route into social understanding. An abstract page for a 2026 International Journal of Developmental Disabilities article says 40 autistic children were recruited for block-building training and 34 completed the study, split evenly between intervention and control groups. After 12 sessions, the trained children reportedly outperformed controls on visual perspective-taking and also showed gains in mental rotation and spatial visualisation, after tasks that encouraged them to inspect structures from different viewpoints.

That rationale has support outside autism-specific research too. A Developmental Psychology paper available through the University of Canterbury repository followed 106 children aged about four and a half in Hong Kong and found that six weeks of block-building work, especially when visual and verbal cues were combined, improved visuospatial perspective-taking and broader social-cognitive measures. The visual-verbal block-building group outperformed a verbal-only version and matched a dialogic-reading approach on emotion understanding and theory of mind, lending weight to the embodied-cognition framework cited in the autism trial registry. Taken together, the new Autism Research paper and the surrounding intervention studies point to a more nuanced picture than a straightforward deficit model: young autistic children appear to process other people’s viewpoints spontaneously, and researchers are now trying to work out how far that sensitivity can be converted into durable gains in understanding minds.