Slide Slam I14
Structure-function relationships and individual differences in reading
Meaghan Perdue1,2, Kenneth Pugh1,2, Nicole Landi1,2; 1University of Connecticut, Dept. of Psychological Sciences, 2Haskins Laboratories
Reading ability has been associated with brain activation and cortical structure, but few studies have examined associations between cortical structure and reading-related brain activation with respect to reading. This leaves a gap in our understanding of neurobiological mechanisms that link cortical structure to reading ability. Here, we examine relationships between cortical thickness (CT), reading-related activation, and individual differences in reading ability to better understand these mechanisms. 73 participants (ages 5-9) completed behavioral assessments of reading and anatomical and functional MRI. Reading ability was measured as a composite of the Letter-Word Identification and Word Attack sub-tests of the Woodcock Johnson Tests of Achievement-III. CT was measured from T1-weighted anatomical MRI scans using Freesurfer. A passive print and speech localizer fMRI task was used to elicit reading-related activation. First, we tested the correlation between cortical thickness, reading-related brain activation, and reading ability in a region of the left superior temporal gyrus (STG) in which we previously identified a significant positive correlation between CT and reading ability (Perdue et al., 2020, Cerebral Cortex). Next, we conducted an exploratory analysis to identify regions of the brain that showed print-specific activation (print > symbol strings contrast). We then tested correlations among cortical thickness, print-related activation, and reading ability in the resulting activation-based regions-of-interest. Age and sex were included as covariates in all correlation models. Results for the analysis of the left STG ROI revealed a non-significant association between CT and reading ability (r=.228, p=.053); no significant correlations with print-related activation were observed in this region. Regions of significant print-specific activation were found in the left occipito-temporal (OT) cortex, left superior temporal sulcus (STS), left precentral gyrus (PreG), and left supplementary motor area (SMA) (p<.05, cluster level FWE-corrected). CT was negatively correlated with reading-related activation in the STS (r=-.246, p=.036), but neither CT nor print activation in the STS were related to reading ability. CT in the left PreG was positively correlated with reading ability (r=.250, p=.033), but no associations with print-related activation in this region were found. No significant correlations were found in the left OT or SMA clusters. Overall, we found only one significant association between CT and reading-related activation in the left STS. The lack of structure-function relationships in the other ROIs shows that CT does not have a clear, direct relationship to reading activation. Notably, no significant correlations between reading-related activation and reading ability were observed, suggesting that although print-related activation could be localized using our task, the degree of activation elicited was not sensitive to individual differences in reading ability in our sample of low- to above- average readers. Thus, links between CT, reading activation, and reading ability may be task-dependent. Other functional measures such as connectivity or response variability may better account for relationships between CT and activation. Alternatively, reading-ability-related effects in CT may reflect anatomical properties of reading network development that are indirectly related to functional processing of print.