Two children's faces smiling
Autism in motion: Unveiling atypical movement profiles

Many years ago, we both - independently – noticed that many of the autistic people that we were working with seemed to move their body in a way that was slightly different from non-autistic people. It was hard to put a finger on exactly what was different but autistic people’s movements tended to look slightly less smooth. This puzzled us; we had read and learnt extensively about the social characteristics of autism but were unaware of any movement differences.

Since then, a growing literature has demonstrated that certain movement differences are both highly prevalent in autism – occurring very early in development and persisting over time – and specific to autistic individuals. That is, studies have shown that the movements of autistic people differ from those of non-autistic people, people with ADHD, and those with other movement-related conditions.

In Professor Cook’s lab, we have comprehensively characterised and compared the arm and face movements of autistic and non-autistic adults, discovering a variety of considerable differences. For example, using a task in which participants trace different shapes, our lab group observed that autistic adults have a different drawing style in which they more dramatically slow down for very curvy parts of shapes. Autistic adults also moved their arm in a less smooth, more jerky, fashion when tracing particular shapes, confirming our observations from many years ago.

In a separate study, in which we employed motion capture technology to track facial movements, we uncovered further evidence supporting our earlier observations: autistic adults, when prompted to express anger, exhibited much more jerky mouth movements. This study also found that, when producing a happy expression, the autistic adults displayed less exaggerated smiles that tended not to reach the eye region.

In both of these recent studies, the specific movement atypicalities we identified were highly prevalent and pronounced. Thus, raising the possibility that these particular movement features could serve as useful indicators of autism, potentially enhancing current screening tools. To explore this possibility, in some preliminary work, we have trained machine learning classifiers on these movement features, and then assessed how well the classifier can separate autistic from non-autistic adults. At this early stage, our results look promising, with the algorithm distinguishing between the groups with high accuracy. However, so far, our work has focused on the movements of adults. If we want to facilitate early identification of autism, it is important to examine the movements of children.

We are therefore really excited that we will soon be advertising a funded PhD studentship to explore this avenue. Specifically, the student will work with us to adapt our existing tasks to make them child-friendly, and then record and compare, the facial and bodily movements of autistic and non-autistic children. Subsequently, we will (hopefully!) create a machine-learning classifier that can distinguish between the movements of the autistic and non-autistic children that we have recruited. Of course, we’re not suggesting that machines should replace clinicians! However, we hope that this work will enable us to one day build a classifier that can give some indication of autism based on body movements and that this tool, when combined with other assessment tools, might speed up the process of autism diagnosis, thus facilitating earlier support for autistic people and their families. If you are interested in discussing this project, please contact us at or This is just one of the PhD studentships that are part of the new ESRC Centre for National Training and Research Excellence in Understanding Behaviour (CENTRE-UB). Follow @Centre_ub on X to hear about other studentships coming soon.

CENTRE-UB is a new centre for doctoral training hosted by the University of Birmingham and funded by the Economic and Social Research Council as part of a larger investment in a national capability in behavioural research. From October 2024, CENTRE-UB will be training three cohorts of PhD students, as well as hosting postdoctoral research fellows, and offering national continuing professional development in behavioural research. For enquiries about CENTRE-UB, please email