Low Focused Attention Is Associated with Elevated ASD-Related Behaviours at 15 Months

Background: Focused Attention (FA) is a filter through which environmental input is selected for learning, through orientation toward and sustained engagement with the external world. Higher FA spans in infancy are associated with enhanced language skills and general developmental ability, more complex play behaviour and better executive function. Genetically-influenced difficulties with attentional control have been identified as a likely developmental pathway to ASD (and co-occurring ADHD) and risk for poor control of attention appears to be distributed across individuals with a first-degree relative with ASD – yet little is currently known about early development of control of attention in ASD. Using FA as an index for early development of attentional control in infants with an older sibling with ASD (‘infant-siblings’) may yield insights into this potential causal mechanism.

Objectives: The first aim of this longitudinal study was to evaluate whether poor development of FA appears to be a characteristic of infant-siblings generally. The second aim was to evaluate whether poor development of FA appears to be specifically associated with levels of early-emerging ASD-related behaviours.

Methods: 112 infants were observed during free play with a set of blocks, at ages 10.5 (controls = 22, infant-siblings = 85) and 15 months (controls = 18, infant-siblings = 72). Coders rated infants’ continuous looking and touching behaviour, from which the peak epoch of FA – longest epoch of simultaneous looking to the blocks whilst manipulating the blocks – was identified (ICC =.924). At 15 months, early behavioural markers of ASD were measured using the Autism Observation Scale for Infants (AOSI).

Results: A repeated measures ANOVA on peak FA, with group (controls, infant-siblings) as a between groups factor showed no main effect of time (F(1,75) = 0.546 p = .462, = .007) and no main effect of group (F(1,75) = 0.224 p = .638, partial-eta sq= .003). The interaction effect between time and group was marginal (F(1,75) = 3.145 p = .081, partial-eta sq = .040) but the generalised eta square value was small (.017). Across all infants, lower Peak FA scores at 15 months were associated with higher concurrent total AOSI scores, indicative of elevated ASD-related behaviours (r(61) = -.222, p = .043). No significant association was found from Peak FA at 10.5 months to AOSI scores at 15 months (r(65) = -.057, p = .327).

Conclusions: As predicted, poor FA at 15 months is concurrently associated with elevated ASD-related behaviours. An equivalent predicted association from FA at 10.5 months to ASD-related behaviours at 15 months was not observed thus it is not clear from these data whether ASD-related behaviours are primary or secondary to difficulties with FA in infancy. Contrary to expectations, no evidence was found to suggest that having an older sibling with ASD is linked with impairment in FA in this sample of infant-siblings. Further research is required to establish whether differences in FA are apparent amongst the minority of infant-siblings later diagnosed with ASD.