Brain regions and networks supporting working memory and selective attention processing in typically developing and extremely preterm-born children

Structural and functional development of the brain during the maturation results in an improvement of higher cognitive abilities such as attention, working memory (WM), and executive functions. The most pronounced changes take place in the frontal, parietal and temporal association cortices which orchestrate the goal-directed behaviour in response to the incoming information. During development, a set of brain regions establishes a processing system of functionally coupled large-scale networks which maintains higher cognition. As an example, the frontal and parietal areas constitute the fronto-parietal network which supports WM and selective attention. The protracted maturation of the association brain regions such as the prefrontal cortex (PFC), which also regulates activity in posterior brain regions, is reflected in the age-related gradual improvement of cognitive abilities in children. The neuroanatomical basis of higher cognition starts to shape during the third trimester of pregnancy when the brain undergoes significant structural changes and forms the precursors of neuronal networks that will support a wide range of cognitive functions. This critical stage of brain development is extremely sensitive to endogenous and environmental factors. Extreme preterm birth (<28 gestational week) poses a high risk of developing lifelong neurological and cognitive impairments. This thesis investigated the functional organization of brain regions underlying attention and WM, and the network functional connectivity in young, school-aged children. Study I investigated task-related activation and top-down regulation induced by PFC in the category-specific visual areas in healthy children and young adults. In three studies of the thesis the consequences of extremely preterm birth were investigated in 7.5-year-old children. Study II investigated the distribution of brain activation during visuospatial n-back tasks and Study IV the dynamics of functional connectivity between two brain states, resting state and WM task performance. Moreover, the effects of extremely preterm birth on the white matter microstructure were investigated in Studies II and III. The data for thesis were obtained using functional magnetic resonance imaging and diffusion tensor imaging (DTI) techniques. Study I found that both adults and children activated the retrosplenial complex (RSC) and parahippocampal place area (PPA) during scene processing, although the RSC responded less than the PPA. Adults demonstrated weaker task-related modulation of activity in the RSC compared with the PPA, while this modulation in children was comparable between the two regions. Together, these results suggest that cognitive control over category specific regions is still under development in 7–11-year-old children. Study II investigated brain responsivity during WM n-back tasks and the white mater microstructure in extremely preterm-born (EPB) and term-born (TB) children. The EPB, compared with TB children, showed weaker WM-related brain activation in the PFC and weaker deactivation of the right temporal lobe. Moreover, they failed to recruit additional neuronal resources when the cognitive load was increased. In addition, the EPB children performed the n-back tasks poorer than the TB children during the scanning. The EPB children also showed alteration of the white matter microstructure in multiple white matter tracts. Unlike the EPB children, the TB children showed significant associations between the microstructure of the anatomical connections and performance of the n-back tasks. Overall, these results suggest that even in EPB children without neurological or neurosensory impairments, the microstructure of the white matter neural tracts and recruitment of brain areas related to WM are compromised. Study III presents the protocol that was used in Study II for the preprocessing and analysis of DTI data acquired from young school-aged children. The protocol pointed out several problems related to the DTI from child populations, and presented solutions to the collecting of the data, preprocessing, template generation, data registration and statistical analysis. Study IV explored differences in the functional architecture of the brain networks between EPB and TB children during resting state and task performance. Tasks performance showed stronger functional connectivity (FC) within and between the regions of the default mode and fronto-parietal networks that are involved in WM processing, whereas the resting state connectivity was largely characterized by stronger between-network FC. The EPB, compared with TB controls, exhibited weaker dynamic modulation of FC between resting state and task performance. In both groups, larger modulation of FC between brain states associated with better performance of the tasks. These results underline the importance of flexible network connectivity for cognitive performance and demonstrate that this ability may be compromised in preterm-born children with no obvious cognitive impairments. The thesis demonstrates a protracted functional specialization for the visual category-selective brain regions and their top-down regulation in school-aged children. The thesis also shows that extreme prematurity may be reflected on the white matter microstructure, brain responsiveness and functional connectivity even in school-aged children with normal global cognitive abilities.