Gross Motor Function and Brain Functional Connectivity in Infants and Toddlers at Risk for ASD

Background:  Gross motor deficits in infancy are one of the earliest findings linked to ASD, highlighting the potential interrelationship between motor delay and ASD-specific symptom emergence. Characterizing brain-behavior relationships for specific aspects of motor function may, therefore, elucidate neurobiology underlying the development of ASD. 

Objectives: To test whether correlations between gross motor behavior and ROI-ROI functional connectivity (fc) 1) are enriched for specific brain networks, and 2) change between 12 and 24 months.

Methods:

Participants:The Infant Brain Imaging Study includes high-risk (HR: has sibling with ASD) and low-risk (LR: has sibling(s) without ASD) infants.  A clinical best estimate ASD diagnosis was assigned at 24 months. Included participants had fcMRI and behavioral data at 12 (n=129; HR+/HR-/LR-=12/76/38) and/or 24 months (n=108; HR+/HR-/LR-=19/63/25).

Imaging: Resting state fcMRI data were acquired on identical 3T Tim Trio scanners at 4 sites with up to 3 BOLD runs (130 frames each run with TR=2.5 seconds). Data processing included motion scrubbing at a FD level of 0.2 mm. One-hundred-fifty frames of clean data were used per subject. Time traces were correlated between 230 functionally-defined regions of interest (ROI) to yield fc values (Fig.1b). ROIs were sorted into 17 putative functional networks using the Infomap community detection algorithm run on the mean connectivity matrix for longitudinal fcMRI data from 37 subjects (Fig.1a).

Measures: Gross motor function was indexed by raw gross motor scores on the Mullen Scales of Early Learning (Fig.1c). Five items were summed to create a “walking scale” at 12 months.

Brain-behavior analysis (Fig.1d): We identified network-network pairs significantly enriched for ROI-ROI fc values that strongly correlated with behavior. fc values for all ROI pairs were correlated against behavioral scores and thresholded at p<.05 (uncorrected). 2x1 Χ² tests and hypergeometric tests established whether enrichment within network pairs exceeded that expected by chance. A 5% false-positive rejection rate was determined by permutation. McNemar tests assessed whether enrichment differed between 12 and 24 months. 

Results:  

Brain-behavior correlations frequently involved the motor network, especially at 12 months (Fig.1d-f). Largely interhemispheric ROI pairs within the motor network markedly overlapped for gross motor and walking scores, mapped predominantly to the presumed lower limb region of motor cortex, and negatively correlated with behavior at 12 months (Fig.1g-h). Conversely, several motor-frontoparietal and motor-dorsal attention network pairs positively correlated with behavior. There were greater involvements of frontoparietal and dorsal attention network pairs at 24 months. Significant age-dependent differences in brain-behavior relationships were observed (Fig.2a,b).

Conclusions:  

Motor network involvement at 12 and 24 months supports the face validity of this analytic approach. Brain-behavior correlations enriched within the motor network suggest that decreased fc between interhemispheric (potentially lower limb) motor ROIs correlates with greater walking ability in early development. At 24 months, enriched positive brain-behavior correlations involving task-control and dorsal attention networks implicate these networks in motor skill development. Future directions include comparisons of brain-behavior relationships for motor functioning in children with and without ASD, and investigating whether similar observed brain-behavior relationships correlate with the differential development of social communication in ASD.