In Japan, the prevalence of autism, a condition that affects communication and social development, is 10 times higher than it was in the past. The number of children receiving special education in public elementary and junior high schools across Japan has surpassed 90,000, corresponding to an increase of 7.4-fold over the past 20 years (since 1993). The authors aimed to establish an objective diagnostic criterion for children with autism spectrum disorder. To this end, they conducted eye-tracking tests on children with autism and neurotypical children. Using a technique developed by one of the authors, which uses afterimages to determine eye blinking (Patent No. 5871290), they obtained the pixel number variation (a numerical value) in gaze direction based on the center of mass of pixels associated with the pupil. The results were then plotted onto a two-dimensional graph, and distributions based on probability density function and receiver operating characteristic curve analysis were ascertained. This analysis yielded a decision boundary clearly demarcating autism and neurotypical distributions and, thus, confirming the reliability of the method. The results revealed that autistic subjects failed to track the moving object. This finding suggests that this technique of measuring abnormality in pixel number is effective for distinguishing individuals with autism from those with typical development and, thus, can serve as an objective criterion for the diagnosis of autism. The assessment method enables the capture of eye movement based on afterimages without the need for any special equipment: only a camera-mounted personal computer is required. In other words, the technology provides a clear and simple method of detection and, as such, can serve as a supplementary assessment tool for psychiatrists in diagnosing autism in children. Moreover, it can also assess learning and treatment efficacy among individuals with autism in educational settings.

PDF (Download Full text)