Poster B18, Thursday, August 16, 3:05 – 4:50 pm, Room 2000AB

How the brain learns to read in environments with high-risk of illiteracy: an fNIRS study of reading development in rural Côte d’Ivoire

Kaja Jasinska1,2, Axel Blahoua3, Fabrice Tanoh4;1University of Delaware, USA, 2Haskins Laboratories, USA, 3Centre d'Action et de Recherche pour la Paix, Abidjan, Côte d'Ivoire, 4Université Félix Houphouët-Boigny, Abidjan, Côte d'Ivoire

Introduction. Learning to read is complex process dependent on multiple linguistic and cognitive systems, and crucially requires quality education to achieve mastery. Millions of children in sub-Saharan Africa fail to achieve target literacy outcomes. The failure to provide quality primary education to all children has profound consequences: 39% of youth aged 15-24 and 41% of adults aged 15+ are illiterate (World Bank, 2012), creating severe economic and social disadvantage within the local community and globally. In Côte d’Ivoire, youth (16-24 years) literacy rates are 53% (UNESCO, 2017). Child labor in cocoa agriculture is a significant problem in rural Côte d’Ivoire, associated with poverty and adversity (e.g. children engaged in hazardous work). These agricultural activities also limit school attendance. Our specific aim is to understand how the brain learns to read when faced with the challenges of inconsistent access to education and low educational quality. We examine language, cognitive, and reading development and measure how the brain’s reading circuit develops in children at high-risk for illiteracy. Methods. 650 primary-school children ages 6-14 (Grade 1, 3, 5) completed language (French, local language) and literacy assessments and were interviewed about cocoa labor. 67 children completed a functional near infrared spectroscopy (fNIRS) neuroimaging study that examined neural activation for reading. fNIRS is highly portable and ideally-suited for studying brain development using field neuroimaging protocols developed for low-resource contexts (Jasinska and Guei, 2018). fNIRS task used a 2 x 3 design: modality (speech, print) by lexicality (word, pseudoword, false-font/vocoded speech). Neuroimaging data were analyzed using a general linear model (NIRS-SPM). Results. Many of the basic components of literacy are not evident even among 5th graders. On average, a 5th grader could correctly read only 27/100 letters and 2-letter combination. Phonological awareness in both French (std. beta = 0.33, p < .01) and a local language (std. beta = 0.34, p < .01) positively and significantly predict reading outcomes. However, cocoa labor has a negative impact on children’s reading scores (std. beta = -0.37, p < .05). A child’s environment (i.e., socioeconomic status, cocoa labor) influenced neural activation while reading: decreased neural activation in the left IFG was observed with increased familial household resources. Among children with minimal reading ability, greater activation for false-fonts was observed in the left IFG, suggesting the reading circuit is not yet fully sensitive to orthography and lexical information, even by 5th grade. Crucially, this pattern of brain activation is different from younger children at comparable reading levels reported elsewhere in the literature (e.g. Turkeltaub et al., 2003; Jasinska and Petitto, 2014). Discussion. Poverty and adversity impact the neural systems for reading in a sample of highly vulnerable children. Understanding the complex relationships between impoverished environment, delayed and/or inconsistent access to education, children’s linguistics and cognitive development, and reading outcomes yields new information about reading development in environments with a high risk of illiteracy. Portable fNIRS informs how the developing brain responds to high-risk environments and can identify neurodevelopmental periods when targeted educational programs can have maximal impacts on learning outcomes.

Topic Area: Perception: Orthographic and Other Visual Processes