Brain structural and functional networks: from connections to cognition. A study applied to Cerebral Small Vessel Disease

Abstract

Modular and hierarchical brain structural networks are especially well adapted to the functional integration of local neuronal processes that underpin cognition. Altered anatomical or functional connectivity patterns were evidenced in many cognitive and affective diseases, defined as "disconnectivity" syndromes. The central hypothesis in the present study is that structural connectivity constrains but does not determine functional connectivity, and this relationship is the basis of cognitive performance. Furthermore, disruptions in structural and functional networks are related to cognitive impairments. Therefore, by studying the brain structural and functional networks, we can better understand how the brain works and its relationship to cognition. We then sought to answer these questions by assessing how structural connectivity is related to functional connectivity and cognition in healthy participants and how changes in brain connectivity relate to cognitive performance in clinical participants.

We evaluated the location and functional integration in response to executing a cognitive task, the Symbol Digit Modalities Test (SDMT), which evaluates the information processing speed (IPS), adapted to the MRI environment. More specifically, we aimed to investigate how regions within the IPS brain network interact functionally, studying effective connectivity, defined as how one region influences or causes activity in another during a given neural process and its related structural connectivity. For clinical application purposes, cerebral small vessel disease (cSVD) patients are of great interest given their impact on global public health, their classification as Disconnectivity Syndrome, and the open questions about the structural and functional disruptions of connectivity that the disease causes and the high incidence of IPS deficit, here evaluated with a variation of the SDMT: the Letter-Digit Substitution Test (LDST).

It was possible to verify that the structural network restricts but does not determine the functional network of IPS in young, healthy participants. In addition, when evaluating the IPS functional network during SDMT execution, activation and functional connectivity of nodes in the default mode network (DMN), essential for cognitive health, was suppressed during the performance of the IPS test. In contrast, the left precuneus (BA 7) node showed a modulatory behavior dependent on the experimental context. Assessing the associations between structural and functional connectivity with LDST scores in patients with cSVD, the findings suggested that cSVD should be considered a global rather than a localized syndrome.

The mediation effects of the functional connectivity of the sensorimotor network between the lesion and the LDST scores point to the role of this network as another critical functional network affected in this clinical group. Using a lesion-symptom mapping approach called "Brain disconnectome mapping,” the associations between the probability of disconnection and LDST scores involved large brain areas in non-demented cSVD patients. Lesion load and its topographical distribution affected the cSVD patients differently. Both studies with cSVD patients, although with different methodologies, point to the general finding that cSVD is a syndrome of disconnection with a broad impact on the brain.

Thus, with the development of the present work, we investigated the dynamics of functional networks, their relationship with structural networks, and cognitive performance in IPS in healthy individuals. The findings suggest that the structural network constrains but does not determine the functional network. We also evaluated connectivity alterations related to the performance of IPS tasks in cSVD, whose results point to the importance of assessing the effects of focal lesions on brain connectivity to understand the heterogeneity of clinical outcomes.