Abstract

The central premise of implementing machines to understand minds is perhaps based on Emerson Pugh’s (in)famous quote: “If the human brain were so simple that we could understand it, we would be so simple that we couldn’t”. This circular paradox has led us to the quest for that quintessential ‘mastermind’ or ‘master machine’ that can unravel the mysteries of our minds. An important stepping stone towards this understanding is to examine how perceptrons—models of neurons in artificial neural networks—can help to decode processes that underlie disorders of the mind. This chapter focuses on the rapidly growing applications of machine learning techniques to model and predict human behaviour in a clinical setting. Mental disorders continue to remain an enigma and most discoveries, therapeutic or neurobiological, stem from serendipity. Although the surge in neuroscience over the last decade has certainly strengthened the foundations of understanding mental illness, we have just started to rummage at the tip of the iceberg. We critically review the applied aspects of artificial intelligence and machine learning in decoding important clinical outcomes in psychiatry. From predicting the onset of psychotic disorders to classifying mental disorders, long-range applications have been proposed and examined. The veridicality and implementation of these results in real-world settings will also be examined. We then highlight the promises, challenges, and potential solutions of implementing these operations to better model mental disorders.