Visual analytics of brain effective connectivity using convergent cross mapping

Abstract

To elucidate the dynamics of information processing in the brain, it is necessary to identify the direction of neural information transmission in the neuronal network and clarify the effects (i.e., the causal relationship) of neuronal activity in one area on neuronal activity in another area. Convergent cross mapping (CCM) has been employed in the neuroscience field to examine the effective connectivity of brain functions. CCM can detect causality from time series data created from deterministic and nonlinear systems. Because CCM includes complicated processes such as the determination of advance parameters, the confirmation of nonlinearity, and the interpretation of results, which results in a lowering of the usability of CCM, there is a strong need for an effective visual interface. In this paper, we propose a visual analytic system that increases the usability of CCM and contributes to new discoveries in effective connectivity. The usability was evaluated using a domain expert questionnaire. It was confirmed that the usability was improved by comparing the proposed system to the original character user interface from the viewpoint of the results and process comprehensibility. In addition, with the proposed system, new findings in human brain connectivity have been obtained from actual magnetoencephalography data during visual cognitive task and resting-state task.