BACKGROUND: In typical infant development, parents and their children jointly contribute to establishing frequent episodes of joint attention that boost language acquisition and shape social cognition. Here we used novel live eyetracking technology to evaluate the degree to which autism spectrum disorder (ASD) is related to reduced responding to others' joint attention bids in infancy (RJA) and to a reduced tendency to initiate joint attention episodes (IJA). Because young infants use their gaze for both RJA and IJA, this approach allowed us to quantify these elusive processes early in life. METHODS: The final sample consisted of 112 infants (54 boys and 58 girls), of whom 81 were at familial risk for ASD and 31 were typically developing low-risk infants. At follow-up (36 months of age), 22 children in the high-risk group were diagnosed with ASD. RESULTS: At 10 months of age, rates of IJA were lower in infants later diagnosed with ASD than in the comparison groups (effect sizes d = 0.78-0.95) and followed an atypical developmental trajectory from 10 to 18 months (p , .002). RJA distinguished infants based on familial ASD risk, albeit not ASD diagnosis. The differences in IJA could not be explained by overall looking time, social preference, eye movement latencies, or number of fixations. CONCLUSIONS: This live eye-tracking study suggests that during an important period for the development of social cognition (10-18 months of age), infants later diagnosed with ASD show marked atypicalities in IJA but not in RJA. The results indicate that IJA is an important target for future prodromal intervention trials.

BackgroundThe ability to follow gaze is an important prerequisite for joint attention, which is often compromised in children with autism spectrum disorder (ASD). The direction of both the head and eyes provides cues to other people’s attention direction, but previous studies have not separated these factors and their relation to ASD susceptibility. Development of gaze following typically occurs before ASD diagnosis is possible, and studies of high-risk populations are therefore important.MethodsEye tracking was used to assess gaze following during interaction in a group of 10-month-old infants at high familial risk for ASD (high-risk group) as well as a group of infants with no family history of ASD (low-risk group). The infants watched an experimenter gaze at objects in the periphery. Performance was compared across two conditions: one in which the experimenter moved both the eyes and head toward the objects (Eyes and Head condition) and one that involved movement of the eyes only (Eyes Only condition).ResultsA group by condition interaction effect was found. Specifically, whereas gaze following accuracy was comparable across the two conditions in the low-risk group, infants in the high-risk group were more likely to follow gaze in the Eyes and Head condition than in the Eyes Only condition.ConclusionsIn an ecologically valid social situation, responses to basic non-verbal orienting cues were found to be altered in infants at risk for ASD. The results indicate that infants at risk for ASD may rely disproportionally on information from the head when following gaze and point to the importance of separating information from the eyes and the head when studying social perception in ASD.Electronic supplementary materialThe online version of this article (doi:10.1186/s13229-016-0069-9) contains supplementary material, which is available to authorized users.

Whether gaze following—a key component of joint attention—is impaired in children with autism spectrum disorder (ASD) is currently debated. Functional gaze following involves saccading towards the attended rather than unattended targets (accuracy) as well as a subsequent processing bias for attended objects. Using non-invasive eye tracking technology, we show that gaze following accuracy is intact in intellectually low-functioning 3-year-olds with ASD. However, analyses of the duration of first fixations at the objects in the scene revealed markedly weaker initial processing bias for attended objects in children with ASD compared to children with typical development and non-autistic children with developmental delays. Limited processing bias for the objects other people attend to may negatively affect learning opportunities in ASD.

During the first year of life, infants develop the capacity to follow the gaze of others. This behavior allows sharing attention and facilitates language acquisition and cognitive development. This article reviews studies that investigated gaze‐following before 12 months of age in typically developing infants and discusses current theoretical perspectives on early GF. Recent research has revealed that early GF is highly dependent on situational constraints and individual characteristics, but theories that describe the underlying mechanisms have partly failed to consider this complexity. We propose a novel framework termed the perceptual narrowing account of GF that may have the potential to integrate existing theoretical accounts.