Brain’s functional network clustering coefficient changes in response to instruction (RTI) in students with and without reading disabilities: Multi-leveled reading brain’s RTI

Abstract

In students in grades 4 to 9 (22 males, 20 females), two reading disability groups—dyslexia (n = 20) or oral and written language learning disability (OWL LD) (n = 6)—were compared to each other and two kinds of control groups—typical readers (n = 6) or dysgraphia (n = 10) on word reading/spelling skills and fMRI imaging before and after completing 18 computerized reading lessons. Mixed ANOVAs showed significant time effects on repeated measures within participants and between groups effects on three behavioral markers of reading disabilities—word reading/spelling: All groups improved on the three behavioral measures, but those without disabilities remained higher than those with reading disabilities . On fMRI reading tasks, analyzed for graph theory derived clustering coefficients within a neural network involved in cognitive control functions, on a word level task the time x group interaction was significant in right medial cingulate; on a syntax level task the time x group interaction was significant in left superior frontal and left inferior frontal gyri; and on a multi-sentence text level task the time x group interaction was significant in right middle frontal gyrus. Three white matter-gray matter correlations became significant only after reading instruction: axial diffusivity in left superior frontal region with right inferior frontal gyrus during word reading judgments; mean diffusivity in left superior corona radiata with left middle frontal gyrus during sentence reading judgments; and mean diffusivity in left anterior corona radiata with right middle frontal gyrus during multi-sentence reading judgments. Significance of results for behavioral and brain response to reading instruction (RTI) is discussed.

Keywords:

• reading disabilities
• behavioral response to reading intervention
• brain response to reading intervention
• graph theory
• correlations between white matter and gray matter clustering coefficients

Public Interest Statement

Some individuals struggle more than others in learning to read and not all reading disabilities are the same. In this study, an interdisciplinary team investigated how students who did and who did not have persisting reading disabilities during middle childhood or early adolescence responded to instruction. Both those who did and did not improved significantly on word reading achievement measures even though those with reading disabilities did not catch up to those without reading disabilities after just 18 lessons. They also showed two kinds of significant brain changes at different levels of language: words, sentences, and text. First, both those with and without reading disabilities changed in clustering coefficients (networks) in specific regions from before to after instruction. Second, correlations between white matter and gray matter emerged only after instruction and instruction may have enabled the white matter to connect with the gray matter clustering. Teaching changes the reading brain!

Acknowledgement

The authors thank William Nagy, Seattle Pacific University, who helped design the multi-leveled language fMRI reading tasks and multi-leveled language learning activities in the instruction. For help with the complex fMRI and DTI analysis, the authors thank Peter Boord and Paul Robinson.

Additional information

Notes on contributors

Todd L. Richards

Todd L. Richards, PhD, Professor of Radiology, is the director of Neuroimaging at the University of Washington Learning Disability Center. He has worked with Dr Berninger and others on the interdisciplinary team for over 20 years to study children with dyslexia and dysgraphia. Recent objectives of the UW learning disability center include (1) Use learning profiles of writing, reading, and oral language skills to diagnose dysgraphia, dyslexia, and oral and written language learning disability (OWL LD); (2) Compare these specific learning disabilities to typical language learning on brain's white matter integrity and functional connectivity; (3) Evaluate which variables predict behavioral and brain response to instruction in reading and writing; and (4) Disseminate findings to researchers, educational professionals, parents, and affected individuals.