Orthographic Depth May Influence the Degree of Severity of Maze Learning Performance in Children at Risk for Reading Disorder

Abstract

Reading disability (RD), which affects between 5 and 17% of the population worldwide, is the most prevalent form of learning disability, and is associated with underactivation of a universal reading network in children. However, recent research suggests there are differences in learning rates on cognitive predictors of reading performance, as well as differences in activation patterns within the reading neural network, based on orthographic depth (i.e., transparent/shallow vs. deep/opaque orthographies) in children with RD. Recently, we showed that native English-speaking children with RD exhibit impaired performance on a maze learning task that taps into the same neural networks that are activated during reading. In addition, we demonstrated that genetic risk for RD strengthens the relationship between reading impairment and maze learning performance. However, it is unclear whether the results from these studies can be broadly applied to children from other language orthographies. In this study, we examined whether low reading skill was associated with poor maze learning performance in native English-speaking and native German-speaking children, and the influence of genetic risk for RD on cognition and behavior. In addition, we investigated the link between genetic risk and performance on this task in an orthographically diverse sample of children attending an English-speaking international school in Germany. The results from our data suggest that children with low reading skill, or with a genetic risk for reading impairment, exhibit impaired performance on the maze learning task, regardless of orthographic depth. However, these data also suggest that orthographic depth influences the degree of impairment on this task. The maze learning task requires the involvement of various cognitive processes and neural networks that underlie reading, but is not influenced by potential differences in reading experience due to lack of text or oral reporting. As a fully automated tool, it does not require specialized training to administer, and current results suggest it may be a practicable screening tool for early identification of reading impairment across orthographies.