Social attention during object engagement: toward a cross-species measure of preferential social orienting.
SADOE coding during regular ADOS play spots reduced people-looking in toddlers with autism.
01Research in Context
What this study did
The team watched toddlers play with toys during an ADOS session. They coded how often each child looked at people versus objects.
Kids with autism and typically developing kids were compared. No extra tasks were added; they just scored normal play.
What they found
Toddlers with autism looked at people much less than typical toddlers. The gap was large and easy to see.
The measure is called SADOE. It flagged the social-orienting difference with strong numbers.
How this fits with other research
Anthony et al. (2020) saw the same split using eye-tracking when toys were on screen. The method differs, but the social-orienting gap repeats.
Hase et al. (2023) moved the idea to teens. Economic games also show autistic players orient less to social cues, proving the pattern lasts.
Vernetti et al. (2024) ran a 30-second live face-to-face task. Like C et al., they caught reduced social attention in toddlers, but they used masks and familiar adults. The two studies agree: less people-looking is a stable ASD marker across small changes.
Why it matters
You can score SADOE during any ADOS play. No extra gear or time is needed. If a toddler rarely looks up from toys, social orienting is weak and early intervention can start right there.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →During your next ADOS, tally each glance from toy to adult; low counts signal weak social orienting to target in intervention.
02At a glance
03Original abstract
<h4>Background</h4>A central challenge in preclinical research investigating the biology of autism spectrum disorder (ASD) is the translation of ASD-related social phenotypes across humans and animal models. Social orienting, an observable, evolutionarily conserved behavior, represents a promising cross-species ASD phenotype given that disrupted social orienting is an early-emerging ASD feature with evidence for predicting familial recurrence. Here, we adapt a competing-stimulus social orienting task from domesticated dogs to naturalistic play behavior in human toddlers and test whether this approach indexes decreased social orienting in ASD.<h4>Methods</h4>Play behavior was coded from the Autism Diagnostic Observation Schedule (ADOS) in two samples of toddlers, each with and without ASD. Sample 1 (n = 16) consisted of community-ascertained research participants, while Sample 2 involved a prospective study of infants at a high or low familial liability for ASD (n = 67). Coding quantified the child's looks towards the experimenter and caregiver, a social stimulus, while playing with high-interest toys, a non-social stimulus. A competing-stimulus measure of "Social Attention During Object Engagement" (SADOE) was calculated by dividing the number of social looks by total time spent playing with toys. SADOE was compared based on ASD diagnosis and differing familial liability for ASD.<h4>Results</h4>In both samples, toddlers with ASD exhibited significantly lower SADOE compared to toddlers without ASD, with large effect sizes (Hedges' g ≥ 0.92) driven by a lower frequency of child-initiated spontaneous looks. Among toddlers at high familial likelihood of ASD, toddlers with ASD showed lower SADOE than toddlers without ASD, while SADOE did not differ based on presence or absence of familial ASD risk alone. SADOE correlated negatively with ADOS social affect calibrated severity scores and positively with the Communication and Symbolic Behavior Scales social subscale. In a binary logistic regression model, SADOE alone correctly classified 74.1% of cases, which rose to 85.2% when combined with cognitive development.<h4>Conclusions</h4>This work suggests that a brief behavioral measure pitting a high-interest nonsocial stimulus against the innate draw of social partners can serve as a feasible cross-species measure of social orienting, with implications for genetically informative behavioral phenotyping of social deficits in ASD and other neurodevelopmental disorders.
, 2022 · doi:10.1186/s11689-022-09467-5