Predicting reading performance from motion coherence thresholds in six- and seven-year-old children.

Abstract

Purpose: Individuals with reading difficulties tend to perform more poorly than controls on tasks requiring them to identify the overall direction of motion within a random dot kinematogram. However, because a majority of research in this area examines motion coherence thresholds in the context of a previous diagnosis, the question becomes whether the degree to which one is sensitive to motion predicts the ability to read. Method: In this investigation, a group of school-aged children (N=55) were asked to complete a 4AFC motion discrimination task in addition to standard measures of reading (Woodcock-Johnson), phonological awareness (PA), and nonverbal ability (K-BIT). Results: Performance on the motion task was negatively correlated with the WordID, Word Attack, and Passage Comprehension subtests of the Woodcock-Johnson (i.e. low motion threshold with good reading ability), negatively correlated with PA, and not correlated with the Matrices subtest of the KBIT. Of particular interest, however, was the finding that splitting the sample by their motion coherence thresholds predicted their reading and PA scores; whereas, splitting the sample by their reading or PA scores did not predict their motion coherence thresholds. Dividing the scores by motion coherence thresholds did not predict nonverbal ability. Conclusion: Motion coherence threshold is a statistically significant indicator of one's reading ability. The controversy in the literature regarding the motion processing capabilities of individuals with reading difficulties may partly lie in the asymmetrical relation between the two abilities.

Footnotes

 Supported by NSF (USA), FQRSC, FCAR, & NSERC (Canada).