Big neuroimaging data-informed mind-brain association studies: Methodology and applications

Abstract

Difficulty in linking psychological concepts with neural mechanisms prevents the field of psychology from being considered a natural science. Previous mind-brain association studies were usually driven by psychological concepts and used to infer brain activity characteristics. Because psychological concepts are formed by observer-dependent experience and consensus, they may not correspond to the functional organization of neural networks in the brain. Even after two decades of mind-brain association research, we are still unable to establish clear and objective associations between psychological concepts and brain function. In this article we first review the challenges in linking psychological concepts with neural mechanisms and then analyze these difficulties’ causes. Based on these analyses, we propose a novel mind-brain association discovery methodology driven by big neuroimaging data. We discuss the advantage of this neuroimaging data-driven discovery methodology and introduce an algorithm, Graicar, to implement it. Finally, we demonstrate the usefulness of the methodology and the gRAICAR algorithm with two detailed applications