Fractal Dynamics of Circle Drawing in Children with ASD

Amanda Fleury, MASc Candidate, University of Toronto; Dr. Tom Chau, Bloorview Research

Institute; Dr. Evdokia Anagnostou, Bloorview Research Institute

Abstract:

Objective:To study the fractal dynamics of circle drawing in children with ASD, which could potentially provide a simple, non-invasive test for a quantitative biomarker for ASD.

Target Population:This study will analyze a sample of children aged 4 to 8 years, with a primary diagnosis of ASD.

Introduction and Background:ASD is a neurodevelopmental disorder with a broad range of symptoms associated with many areas of the brain. The pervasive nature of this disorder makes it difficult to pinpoint a single cause or mechanism. Although various studies have been done to identify potential biomarkers for ASD, behavioral tests are still most commonly used.

Fractals are patterns (that can be either spatial or temporal), where the statistical properties of the pieces are proportional to the statistical properties of the whole. Temporal fractals can signify the presence of long-range correlations in a time series. That is, events at one time are statistically related to events at different times in the series. Temporal fractal dynamics are thought to describe a natural rhythm of motion and are found in various physiological processes such as heartbeat and human gait. Fractal dynamics have been shown to degrade with age and certain movement disorders such as Parkinson’s.

Methods:Participants will be asked to draw circles using a computerized tablet and pen, which record spatio-temporal data. Subjects will be asked to draw circles with both hands separately at their preferred pace, both continuously and discontinuously, and then as fast as they can. Using the tablet data, we will estimate the time taken to draw each circle, the strength of self-affinity in the inter-circle time series, as well as several measures indicative of the quality of the circles. We will then compare these estimated quantities to those of typically developing controls.

Results and Conclusions:Data collection is still in progress and only preliminary results will be presented. One of the main challenges in the care of children with autism is accurate and consistent diagnosis. This research has the potential to introduce a surrogate biomarker to aide in quantitative diagnosis to supplement the qualitative observations of clinicians.