A quarterly publication of the Autism Research Institute

The Autism Research Review International is quarterly publication of the Autism Research Institute

Fall, 2022 | Number 4, Volume 36

Body movement patterns at four months may help to predict ASD risk

Researchers in Japan report that it may be possible to identify infants with autism spectrum disorders (ASD) as early as four months after birth by analyzing their body movements. 

Hirokazu Doi and colleagues enrolled mother-child pairs from the general population in their study. When the children were four months old, the researchers made video recordings of their spontaneous movements. Once the children reached the age of 18 months, the researchers asked their mothers to fill out the Modified Checklist for Autism in Toddlerhood (M-CHAT) questionnaire, which is designed to identify symptoms of ASD. A total of 41 infants were included in the final analysis. 

The researchers report, “Infants at high risk for ASD at 18 months of age exhibited less rhythmic and weaker bodily movement patterns at four months of age than low-risk infants.” When the researchers conducted a machine learning-based analysis of their data, they found that they were able to predict ASD-like behavior at 18 months of age with an accuracy “at the level acceptable for practical use.” 

The researchers caution, however, that their study has several limitations. For example, they say, they relied solely on the M-CHAT—which has been criticized for its low sensitivity and positive predictive value in diagnosing ASD—to categorize children at high or low risk for ASD at 18 months. In addition, they were not able to follow up to see if any of the children actually received a later diagnosis of ASD.


“Prediction of autistic tendencies at 18 months of age via markerless video analysis of spontaneous body movements in 4-month-old infants,” Hirokazu Doi, Naoya Iijima, Akira Furui, Zu Soh, Rikuya Yonei, Kazuyuki Shinohara, Mayuko Iriguchi, Koji Shimatani, and Toshio Tsuji, Nature Scientific Reports, October 27, 2022 (free online). Address: Toshio Tsuji, Graduate School of Advanced Science and Engineering, Hiroshima University, Higashihiroshima, Japan. [email protected]