12-Month peak alpha frequency is a correlate but not a longitudinal predictor of non-verbal cognitive abilities in infants at low and high risk for autism spectrum disorder

Abstract

Although studies of PAF in individuals with autism spectrum disorder (ASD) report group differences and associations with non-verbal cognitive ability, it is not known how PAF relates to familial risk for ASD, and whether similar associations with cognition in are present in infancy. Using a large multi-site prospective longitudinal dataset of infants with low and high familial risk for ASD, metrics of PAF at 12 months were extracted and growth curves estimated for cognitive development between 12-36 months. Analyses tested whether PAF 1) differs between low and high risk infants, 2) is associated with concurrent non-verbal/verbal cognitive ability and 3) predicts developmental change in non-verbal/verbal ability. Moderation of associations between PAF and cognitive ability by familial risk status was also tested. No differences in 12-month PAF were found between low and high risk infants. PAF was associated with concurrent non-verbal cognitive ability, but did not predict change in non-verbal cognitive over development. No associations were found between PAF and verbal ability, along with no evidence of moderation. PAF is not related to familial risk for ASD, and is a neural marker of concurrent non-verbal cognitive ability, but not verbal ability, in young infants at low and high risk for ASD.

Keywords: Autism spectrum disorder; Cognitive development; EEG; Infant siblings; Peak alpha frequency.

Fig. 1

Illustration of the parameters extracted by the FOOOF algorithm. The figure on the left demonstrates an example of appropriate alpha peak identification, whereas the figure on the right demonstrates an example of a participant who had no clear alpha peak. Power values are log transformed.

Fig. 2

12-month power spectral density’s (PSD) at 0-40 Hz for all participants at central (C3, C4, Cz) region of interest. Peak alpha frequencies were extracted from the highlighted frequency range (6−12 Hz). Power values are log transformed.

Fig. 3

Growth Curve Model to Test Concurrent and Longitudinal Associations between Peak Alpha Frequency (PAF) and Cognitive Development.

Fig. 4

No Difference in 12-Month Peak Alpha Frequency by Familial Risk Status. Although group comparisons were only made on the basis of risk, both risk and outcome groups are depicted in the figure to aid interpretation

Fig. 5

Estimated Growth in Non-Verbal (top) and Verbal Ability (bottom) Between 12-36 Months in Infants at Low and High Familial Risk for ASD.

Fig. 6

Positive Association Between 12-Month Peak Alpha Frequency and Estimated Non-Verbal Ability Intercept. Although the model was only run with risk, both risk and outcome groups are depicted in the figure to aid interpretation. Lines represents the best fitting regression line for each group, shaded areas represent 95 % confidence intervals.