Pupillary Light Responses in Infants at Low and High Risk for ASD

Background: Research has identified atypical physiological responses in individuals with ASD (e.g., Bal et al., 2008; Joseph et al., 2008), including an abnormal pupillary light reflex (PLR) that discriminated children with ASD from controls with 92.5% accuracy (Fan et al., 2009). The PLR is a well-studied index of the cholinergic system (for discussion, see Fotiou et al., 2009), and recently, Nyström et al. (2015) examined the PLR in infant siblings of children with ASD, a group with as high as a 1 in 5 chance of developing the disorder (Ozonoff et al., 2011). This work found increased sensitivity in the PLR in 10-month-old infants at high risk for ASD (HRA), marked by stronger and faster responses in comparison to low-risk controls (LRC; Nyström et al., 2015).

Objectives: The present study extends the findings of Nyström et al. (2015) to examine sensitivity in the PLR in HRA and LRC infants at 6 and 12 months, in hopes of gaining a richer picture of the trajectory of this response across the first year of life.

Methods: Participants included 32 infants, 12 6-month-olds (HRA: n=5; LRC: n=7) and 20 12-month-olds (HRA: n=5; LRC: n=15). As part of a paradigm examining attention to faces, a Tobii x120 eye-tracker was used to present infants with up to 14 trials of a black fixation screen followed by 32-s videos with white backgrounds showing familiar and unfamiliar faces. On trials where infants were attending to the screen during the shift from dark to light, the maximum absolute acceleration during the PLR and the relative pupil constriction were obtained. Pupil diameter was sampled at 60 Hz and Gaussian-smoothed with a standard deviation of 5 samples (83 ms). Velocity and acceleration were similarly smoothed prior to further processing. Relative pupil constriction was calculated from D₀ (baseline diameter) and D_m (minimum diameter) as (D₀-D_m)²/D₀². Infants with at least three valid trials were included.

Results: Univariate ANOVAs were used to examine the influences of age (6, 12) and group (HRA, LRC) on constriction and acceleration during the PLR. For constriction, a marginal age*group interaction was found (p=.06), with LRC showing decreased constriction across age while HRA show increased constriction (Figure 1). For acceleration, a significant age*group interaction was found (p=.046), with HRA showing marginally increased acceleration in their PLR across age (p=.058) and LRC showing no change (p=.63; Figure 2).

Conclusions: Despite the modest sample size, this study extends the work of Nyström et al. (2015) to illustrate changes in the PLR between 6 and 12 months in HRA and LRC. Their work found stronger and faster responses in HRA as compared to LRC at 10 months, but the present findings suggest this is not yet the case at 6 months. Further analyses are underway with an extended sample, and future work will examine how individual differences in this physiological response might relate to developmental outcomes.