Reduced responsiveness to joint attention (RJA) is an early predictor of autism (Rozga et al., 2011; Landa et al., 2007, Sullivan et al., 2007; Yoder et al., 2009) that is associated with long-term outcomes (Sigman & Ruskin, 1999; Gillespie-Lynch et al., 2011). While retrospective analyses suggest that children who are later diagnosed with autism exhibit reduced social responsiveness by 6 months of age (Maestro et al., 2002), few behavioral differences between autistic and non-autistic infants have been demonstrated prospectively prior to 12 months (Yirmiya & Charman, 2010). Infants later diagnosed with autism exhibited reduced RJA during an in-person assessment at 12, but not 6, months of age (Rozga et al., 2011). Perhaps an eye-tracking measure of RJA might detect symptoms earlier than in-person measures have?
Objectives:
- Determine if an eye-tracking measure of RJA is associated with autism symptoms at 24 months.
- Assess relations between eye-tracking RJA and language development.
Methods:
Infant siblings of children with autism participate in a longitudinal study at 6, 12, 18, 24, and 36 months of age. An eye-tracking measure of RJA is administered each visit. During eye-tracking, infants watch a video of a model fixating 1 of 2 objects while eye movements are recorded with a Tobii 1750 eye tracker. Each trial consists of a baseline phase, a social greeting, and the model turning toward an object for 5 seconds. Social attention is indexed by the overall duration of time the infant attends to the model during the social greeting. RJA is calculated by dividing the number of trials wherein the infant first looks toward the object that the model is attending to by the number of usable trials. Language is assessed at each time point with the Mullen Scales of Early Learning. Social symptoms and restricted, repetitive behaviors are calculated from the Autism Diagnostic Observation Schedule (ADOS: Gotham, Risi, Pickles, & Lord, 2007) administered at 24 months. Diagnostic outcomes will become available this year.
Results:
Six month (r (11) =-.638, p= .044) and twelve month (r (21) =-.517, p= .016) social attention was associated with 24 month social symptoms. Eye-tracking RJA at 6 and 12 months was not associated with 24 month symptoms (p>.05). Six month eye-tracking RJA was associated with expressive (r (19) =.511, p= .025) and receptive (r (19) =.582, p= .009) language at 12 months of age. No relations between eye-tracking RJA at 12 months and language were observed (p>.05).
Conclusions:
The observed link between social responsiveness and subsequent autism symptoms is consistent with retrospective analyses (Maestro et al., 2002). Social attention may be an earlier predictor of social symptoms than RJA. Indeed, reduced social interest in autism may contribute to atypical RJA (e.g. Dawson et al., 1998). Eye-tracking RJA at 6 but not 12 months predicted language development. While eye-tracking RJA measures a similar construct as in-person measures of RJA, eye-tracking may be less effective at eliciting RJA (Navab et al., 2011). Interactive RJA assessment may mirror the contingencies of language learning situations better than pre-recorded stimuli.