Parent-mediated intervention in infants with an elevated likelihood for autism reduces dwell time during a gaze-following task

Cognitive markers may in theory be more sensitive to the effects of intervention than overt behavioral measures. The current study tests the impact of the Intervention with the British Autism Study of Infant Siblings-Video Interaction for Promoting Positive Parenting (iBASIS-VIPP) on an eye-tracking measure of social attention: dwell time to the referred object in a gaze following task. The original two-site, two-arm, assessor-blinded randomized controlled trial (RCT) of this intervention to increase parental awareness, and responsiveness to their infant, was run with infants who have an elevated familial likelihood for autism (EL). Fifty-four EL infants (28 iBASIS-VIPP intervention, 26 no intervention) were enrolled, and the intervention took place between 9 months (baseline) and 15 months (endpoint), with gaze following behavior measured at 15 months. Secondary intention to treat (ITT) analysis showed that the intervention was associated with significantly reduced dwell time to the referent of another person's gaze (beta = -0.32, SE = 0.14, p = 0.03) at 15-month treatment endpoint. Given the established link between gaze following and language, the results are considered in the context of a previously reported, non-significant and transient trend toward lower language scores at the treatment endpoint (Green et al. (2015) The Lancet Psychiatry, 2(2), 133-140). Future intervention trials should aim to include experimental cognitive measures, alongside behavioral measures, to investigate mechanisms associated with intervention effects. The current study tests the impact of the Intervention with the British Autism Study of Infant Siblings-Video Interaction for Promoting Positive Parenting (iBASIS-VIPP) on an eye-tracking measure of social attention: looking time to the object of another person's gaze. Infants who received the iBASIS-VIPP intervention spent less time looking to the object of another person's gaze. This may be linked to the previously reported trend of slightly lower language scores immediately after the intervention. These findings emphasize the need for future trials to include more fine-grained, experimental measures of social interaction, alongside broad assessment measures, and to better understand how the intervention might change behavior.