Authors: Laura Wright, Matteo De Marco, Annalena Venneri
Published: 2020-12-07
DOI: 10.1002/alz.038955
Source: Full article
AbstractBackgroundIn accordance with the physiological networks which underlie it, human cognition is characterised by both segregation and interdependence of a number of cognitive domains. Cognition itself, therefore, constitutes a network, the organisation of which may be quantifiable using methods of graph theory. The present study aimed to exploit graph theory methods to assess changes in cognitive profiles throughout healthy ageing and further interrogate how such profiles may differ in the presence of pathology.MethodA comprehensive neuropsychological test battery was performed in 6 participant groups; 3 groups of healthy adults split into the age ranges 18‐39 (N 75), 40‐64 (N 75) and 65 and over (N 70) and 3 patient groups including adults with amnestic mild cognitive impairment (MCI, N 75), non‐amnestic MCI (N 31) and Alzheimer’s type dementia (N 60). Partial correlations were performed on standardised scores between 16 cognitive tests. A binarised adjacency matrix was then created for each group in which significant correlations (p <.05) were retained as the connections or edges and each cognitive test represented a node. Network analysis measures were then applied to each matrix using the Brain Connectivity Toolbox.ResultNetwork wide clustering and local efficiency of nodes showed an increase between the stages of normal ageing and this was considerably exacerbated by the presence of disease. Conversely, average betweenness centrality first showed an increase between the young and middle‐aged groups but was reduced compared to middle‐aged in the healthy older group. This reduction appeared to be again exacerbated within patients. These findings were particularly apparent between patients and healthy adults in tests of memory function.ConclusionWhile whole network approaches have become a staple of functional neuroimaging, such techniques have rarely been applied to the structure of cognition itself. Here, it has been demonstrated that the organisation of our cognitive networks is influenced not only by the ageing process but is further significantly altered by the presence of disease. Further investigation in this area may provide a novel approach for quantifying the effects of disease on cognitive function and could contribute to the development of sophisticated diagnostic tools in the future.