Characterisation of cognitive activity using minimum connected component

Abstract

The concept of functional brain networks offers new and interesting avenues for studying human brain function. One such avenue, as described in the current paper, involves spanning subgraphs called Minimum Connected Compo‐ nents (MCC) that contain only the influential connections of such networks. This paper investigates cognitive load driven changes across different brain regions using these MCC sub-graphs constructed for different states of brain functioning under different degrees of cognitive load using the graph theoretic concept of clique. The presence of cliques signifies cohesive interconnections among the subsets of nodes in MCC that are tightly knit together. To further characterise the cognitive load state from that of the baseline state, the hemisphere wise interac‐ tions among the electrode sites are measured. The empirical analysis presented in this paper demonstrates the efficiency of the MCC based clique analysis in detecting and measuring cognitive activity with the technique presented poten‐ tially having application in the clinical diagnosis of cognitive impairments.