| dc.description.abstract | The executive control network (ECN) is critical for effective, organized behavior. But what mechanisms allow it to play this core role in higher cognition? An important property of the ECN is thought to be its connectivity to the rest of the brain, which makes it a global workspace by bringing diverse kinds of information together, and allows it to control processing by modulating different brain systems. This thesis uses neuroimaging to unpack the relationship between connectivity and function of the ECN in adults, and then explore its origins, by measuring the maturity of connectivity prior to the emergence of many functions in infants. First, I looked at the way that the ECN adapts in time. It has been previously claimed that it is a flexible hub, changing its connectivity from task-to-task. I investigated its ability to reconfigure in two open large access datasets, by examining dynamic functional connectivity during naturalistic cognition, while participants watched a movie. I found that the ECN does not have the trait of being a flexible hub, and that it only adopts this state for particular task sets. I next examined the covariation of task selectivity and connectivity within the ECN. Using fMRI data from the human connectome project, I examined the degree to which different tasks evoked distinct activation patterns in one of the key nodes of the ECN, the dorsolateral prefrontal cortex (DLPFC). I identified five distinct clusters of task selectivity and then examined how this task selectivity was related to structural connectivity. I used machine learning to test if the structural connectivity of a vertex to the rest of the brain was predictive of its functional selectivity, and found that it was. Lastly, I examined the developmental origins of task selectivity. Specifically, I investigated if structural connectivity might provide a ?proto-organization? from which functional selectivity could
7
develop, by assessing the maturity of structural connectivity within the DLPFC in infants, using data from the developing human connectome project. These studies shed light on the complex pattern of interrelations between connectivity and function in the ECN. Although the ECN was not found to be a flexible hub during naturalistic cognition, I did find that in a diverse set of tasks, local functional specialization was predicted by structural connectivity. Furthermore, this connectivity was investigated in neonates, to understand if it may precede functional specialization. | en |
