Identifying the signature of prospective motor control in children with autism

Luca Romeo,Caterina Ansuini,Cristina Becchio,Andrea Cavallo,Francesca Battaglia,Massimiliano Pontil,Stefano Panzeri,Lino Nobili

doi:10.1038/s41598-021-82374-2

Abstract

Failure to develop prospective motor control has been proposed to be a core phenotypic marker of autism spectrum disorders (ASD). However, whether genuine differences in prospective motor control permit discriminating between ASD and non-ASD profiles over and above individual differences in motor output remains unclear. Here, we combined high precision measures of hand movement kinematics and rigorous machine learning analyses to determine the true power of prospective movement data to differentiate children with autism and typically developing children. Our results show that while movement is unique to each individual, variations in the kinematic patterning of sequential grasping movements genuinely differentiate children with autism from typically developing children. These findings provide quantitative evidence for a prospective motor control impairment in autism and indicate the potential to draw inferences about autism on the basis of movement kinematics.

Highlights

Failure to develop prospective motor control has been proposed to be a core phenotypic marker of autism spectrum disorders (ASD)
Data for this study consisted of 1600 movements recorded from 20 ASD and 20 typically developing (TD) children reaching towards and grasping a bottle with one of three prospective intentions: place the bottle into a box, pour some water or pass the bottle to another person
We assessed the potential of multivariate prospective motor control data to discriminate movements performed by ASD children from TD children in a sample of cases matched on age, gender, full-scale IQ, and stature

Summary

Introduction

Failure to develop prospective motor control has been proposed to be a core phenotypic marker of autism spectrum disorders (ASD). Recent years have seen an increasing interest in the motor side of autism spectrum disorders (ASD)[1], with several researchers going as far as proposing that impairments in the prospective control of movements[2,3] may be predictive of ASD and may even underlie some of the core features of A SD4 Despite this enthusiasm, relatively little quantitative information is available about prospective movement alterations in ASD and questions remain regarding whether (and to what extent) kinematic patterning permit differentiating ASD and non-ASD profiles over and above individual differences in motor output[5]. If the classification accuracy lies significantly above the level expected by chance (e.g., 0.50), it can be concluded that a difference between classes exists Studies applying this approach to prospective movement data suggest that patterns related to ASD can be classified with near-perfect accuracy[6,7]. When repeated measures from the same individual are randomly assigned to training and testing as in common machine learning

Methods

Results

Discussion

Conclusion