Presentation

Martina Villa^1,2,3, Nabin Koirala², Meaghan Perdue^2,4, Elena Grigorenko^2,5,6, Nicole Landi^1,2; ¹Department of Psychological Sciences, University of Connecticut, ²Institute for the Brain and Cognitive Sciences (IBACS), University of Connecticut, ³Haskins Laboratories, ⁴Department of Radiology, University of Calgary, ⁵Department of Psychology, University of Houston, ⁶Baylor College of Medicine

Environmental effects explain about 44% of the variability in reading skills (Andreola et al., 2020), however the paths that characterize these influences remain little understood. Specifically, while individual level effects of genes on behavior must be routed through the brain, the involvement of brain structure and activity in the relationship between environmental risk factors and reading (related) abilities has been understudied. Socio-Economic Status (SES), one of the most robust environmental risk factors for reading disabilities, has been previously associated with white matter (WM) in the bilateral Inferior and Superior Longitudinal Fasciculi (ILF and SLF) and this relationship is thought to mediate the effects of SES on reading (e.g. Ozernov-Palchik et al., 2018; Zuk et al., 2020; Turesky et al., 2021). Our group previously linked WM microstructure within several tracts (including bilateral ILF and SLF) to reading and phonological processing (PP) and found that PP mediated the relationship between WM and reading (Koirala et al., 2021). This analysis was performed using the Healthy Brain Network (HBN) biobank (Alexander et al., 2017), a large dataset that includes neurobiological, behavioral, and demographic data of over 4,000 children, ages 5-21. Expanding on our previous work, we aim to characterize the complex pathways between SES and reading to test the possible mediation role of the bilateral ILF and SLF and of PP. Further, we will include several additional environmental risk factors such as parental stress and weight at birth as they may relate to SES, WM and reading. Although reports have considered environmental risk factors for reading (see Mascheretti et al., 2018 for a review), their effects are typically tested separately; structural equation modeling (SEM) offers the opportunity to examine multiple risk factors in one model. Furthermore, thanks to its large sample size and given the large age range of participants, the HBN biobank will also afford examination of the mediation pathways for children of different ages using multiple groups SEM. Finally, in addition to traditional diffusion measures, we will examine measures of neurite orientation density and dispersion, which were shown to be more sensitive and specific to different neurite features than traditional diffusivity measures (Zhang et al., 2012). We will use SEM to test a mediation model, using environmental risk factors as predictors and reading (measured by the test of word reading efficiency) as the predicted variable. The mediators to be evaluated will be WM for the bilateral ILF and SLF (measured by Fractional Anisotropy, Orientation Dispersion Index, and Neurite Density Index) and PP (measured by the comprehensive test of phonological processing), an endophenotype more closely related to brain structure than reading. Our model is informed by existing literature on the relationship between environmental risk factors, WM, PP, and reading, and by data driven approaches such as exploratory factor analysis and model fit evaluation. We hypothesize that WM will mediate the effect of SES on reading, and that this path is further mediated by phonology; additionally, we postulate that environmental effects on WM and reading will be larger for younger (relative to older) children.

Topic Areas: Reading, Development