Predicting Phenotypic Variation through Human Brain Decoding and Connectivity: Applications for Capturing Individual Differences in Face Recognition Ability, Aging, and Brain-Computer Interfaces

These three dissertation chapters are a collection of studies aimed to better understand individual differences in various cognitive and biological processes. The first chapter focuses on face recognition ability and the cortical structures/ representations that give rise to neurotypical heterogeneity in this behavior. The second chapter focuses on the aging process at the epigenetic and cognitive level, linking these two levels and explaining these relationships through functional connectivity of the human brain. Finally, the third chapter focuses on algorithmic modifications that incorporate individual difference information when training machine learning systems suitable for brain-computer interface systems. The substantive domains across the three chapters differ drastically, but the general analytical framework for approaching these diverse problems is shared across the three chapters. The common thread linking these interdisciplinary efforts is 1) the committed focus to understanding why people’s minds and brains operate differently from one another, and 2) the successful application of modern machine learning tools to function as a mechanism for uncovering new insights about the brain and behavior. In summary, this collection of work leverages the application of modern measurement and analytic tools to design models of behavior that incorporate critical idiosyncrasies between people. These papers add novel research contributions to three different domains relevant to modern psychology and neuroscience: 1) face recognition ability, 2) aging, and 3) brain-computer interfaces.