Socioeconomic Neurophysiology: Exploring the relationship between a child’s environment and their neurocognitive development

Abstract

This thesis investigates the impact of socioeconomic status (SES) on the neurophysiology of children’s brains. Human beings develop within a variety of social and economic environments. This variability is inevitably reflected in cognition and associated brain activity. This includes stark negative impacts on the lower end of this spectrum. Research is often limited to one measure of SES, such as income or housing. There is also a tendency to limit analysis to one outcome, like a cognitive test or a particular form of brain scan. However, both SES and neurocognitive development are multi-faceted. If we are to understand their interactions fully, this complexity must be considered. I endeavour to address this complexity in the current thesis, by considering multiple measures of SES across structural, functional and task-based neuroimaging. I apply data-driven techniques including general linear modelling, auto-regressive models, and graph matching. Chapter 1 reviews the current literature on SES and development. I put forward a multi-level approach to build upon previous work. In Chapter 2 I compare different methods of modelling brain networks and contrast their suitability for capturing SES related variance. I find a distributed network of connections which relate to different elements of SES. I also show that functional neurophysiological methods are superior in capturing this variance. In Chapter 3, I investigate how neurophysiological activity during a passive phonological task predicts SES. I find that later processing is specifically related to subjective parental ratings of SES. In Chapter 4 I extend this approach to an actively involved visual working memory task and find differential associations with objective and subjective measures. In Chapter 5 I integrate these findings with existing theories and models. I find my results support theories connecting SES, inhibition, and language processing. I also reflect on the many distributed associations that do not fit these parsimonious models. Most importantly, this thesis showcases a new approach to SES research in cognitive neuroscience, and the importance of considering SES as multi-factorial.