Predicting language processing efficiency in preterm children: Social-environmental and neuro-biological factors

点击翻译按钮获取中文摘要

Each year in the U.S., more than one in 10 children are born preterm (PT). Approximately half of very preterm survivors, born at < 32 weeks’ gestation, develop language-based learning impairments that may be discovered late and put children at substantial risk for poor outcomes throughout their lives. Our previous grant (HD069150) convincingly demonstrated that language processing efficiency, assessed at 18 months in an eye- tracking paradigm, called looking-while-listening (LWL), was more predictive of long-term outcomes than standardized tests and parent reports. PT children who were faster at language processing at 18 months showed advantages in both verbal and non-verbal skills at 54 months. Our next step is to understand early predictors of language processing efficiency in PT children. In this renewal, we enroll PT neonates (n = 140) from two language groups, primarily English- and primarily Spanish families, to increase the diversity of our sample and to improve generalizability. We assess social-environmental predictors at 12 months (infant environment) and 18 months (toddler environment) using day-long audio recordings of the child’s language environment and naturalistic laboratory observations of caregiver-child interactions. We assess neurobiological predictors, focusing on white matter microstructure, in the neonatal period (neonatal scans) and at 12 months (infant scans). We use two complementary types of MRI scans to assess white matter axonal properties and myelin content. At 18 months, the primary outcome measure is language processing speed in the LWL task, the time it takes the child to shift eye gaze to the picture of an object that was just named. Parent reports of vocabulary and scores on a standardized test of language development are secondary measures. Our aims are to: (1) determine if properties of the infant and/or toddler environments predict language processing speed and secondary outcomes in PT children from the two language-groups; (2) determine if properties of white matter pathways, assessed from neonatal and/or infant MRI scans, predict language processing speed and secondary outcomes, after consideration of language group, clinical variables, and other covariates; and (3) investigate the contributions of social-environmental factors and white matter development on language processing speed in this diverse sample of children born PT. Our main hypothesis is that relations between language learning environments and language processing speed are mediated by changes in white matter development, suggesting that supportive learning environments impact language outcomes because learning environments advance the development of white matter microstructure. The demonstration that white matter change mediates the association of social-environmental factors on language outcomes provides a clear example of experience-dependent plasticity in the human brain. This finding would represent a theoretical contribution to models of learning and development in PT children across language groups and would inform clinical practice and early intervention for PT children who are at high risk for poor language outcomes.