Tom Mitchell - Understanding Neural Processes: Getting Beyond Where and When, to How

Cognitive neuroscience has always sought to understand the computational processes that occur in the brain. Despite this, years of brain imaging studies have shown us only where in the brain we can observe neural activity correlated with particular types of processing, and when. It has taught us remarkably little about the key question of how the brain computes the neural representations we observe.

The good news is that a new paradigm has begun to emerge over the past few years, to directly address the how question. The key idea in this paradigm shift is to create explicit hypotheses concerning how computation is done in the brain, in the form of computer programs that perform the same computation (e.g., visual object recognition, sentence processing, equation solving). Alternative hypotheses can then be tested to see which computer program aligns best with the observed neural activity when humans and the program process the same input stimuli. We will use our work studying language processing as a case study to illustrate this new paradigm, in our case using ELMo and BERT deep neural networks as the computer programs that process the same input sentences as the human. Using this case study, we will examine the potential and the limits of this new paradigm as a route toward understanding how the brain computes.