The One Reason Why I Don't Do All the Maths in Machine Learning?

The prediction error minimization hypothesis is the first grand unified empirical theory about how the brain implements the mind. The hypothesis, which is as bold as it is controversial, proposes to explain the mind via one core mechanism: a process of comparing predicted sensory input with actual input, updating our hypotheses in light of the difference, and generating new predictions. In The Predictive Mind (Oxford University Press), Jakob Hohwy introduces this theory to a wider audience, develops the theory’s explanation of perception, and explores its potential for explaining consciousness, attention, representation, and mental illness. In this interview, Hohwy, who is associate professor of philosophy at Monash University, considers how the theory turns the traditional view of perception on its head and addresses its implications for the relation between cognition and perception and the possibility of knowledge of the external world.