Changes in Brain Functional Connectivity Associated with the Emergence of Reaching and Grasping in Infants at Risk for Autism

Background: Early reaching experiences are associated with outcomes across multiple domains implicated in autism spectrum disorder (ASD), including socially oriented attention and language. While the consequences of the transition to independent reaching have been studied, little is known about the underlying neural systems. Characterizing brain changes involved in this motor milestone may inform our understanding of both typical and atypical developmental trajectories and potentially aid in ASD early identification.

Objectives: To identify brain functional connectivity (fc) changes associated with the emergence of reaching and grasping. We hypothesized that children showing the largest changes in manual motor behavior between 6 and 12 months would show the largest concurrent changes in fc between a subset of brain regions.

Methods:

Participants: The Infant Brain Imaging Study collected behavioral and resting-state fc magnetic resonance imaging (MRI) data from high risk (HR; with a sibling with ASD) and low risk (LR; with a typically developing sibling but no familial history of ASD) participants. Analyses included 71 IBIS participants (24 LR) with fcMRI and behavioral assessments at both 6 and 12 months. This mixed-risk sample enabled investigation of brain-behavior relationships across a broad continuum relevant to both typical and atypical outcomes.

Imaging: Each participant completed 2-3 fcMRI scans (3T; 130 frames/scan, TR=2.5 s). A framewise displacement (FD) threshold of .2 mm was used to eliminate motion-contaminated data, and 150 clean frames were used for each participant. fcMRI time traces for 230 functionally defined regions of interest (ROIs) were correlated on a pairwise basis to generate whole-brain fc matrices at each age for each participant (Fig.1b). ROIs were sorted into twelve putative functional networks using the Infomap community detection algorithm run on the average fc matrix across all subjects at both ages (Fig.1a). Six-month fc values were subtracted from 12-month fc values to generate an fc change matrix for each subject (Fig.1b).

Measures: Raw fine motor scores and a “reaching and grasping (RG) composite” derived from the Mullen Scales of Early Learning were used to index manual motor ability. Six-month scores were subtracted from 12-month scores to evaluate behavior change (Fig.1c).

Brain-behavior enrichment analysis: Pearson’s correlation was calculated between ROI-ROI fc change and motor score change (Fig.2a) and thresholded at p<.05 to identify “strong” fc-behavior relationships (Fig.2b). Χ² and hypergeometric tests identified greater-than-chance densities of strong fc-behavior relationships within network pairs for each motor measure, and empirical significance levels were determined using permutation.

Results: Our brain-wide search revealed maturation of functional connections clustered in motor networks and between motor and subcortical/cerebellar regions were important for both measures of manual motor development (Fig.2c,d). For RG, important fc changes also extended to more of the motor system as well as to visual and attention networks (Fig.2e).

Conclusions: This marks the earliest known description of changes in functional brain systems that underlie the emergence of visually guided reaching. As hand-eye coordination may play a role in socially oriented attention during infancy, future work will investigate whether these fc changes are also important for the emergence of early social behaviors.