Why the brain is not a computer
1 hr 15 min
It's one of the most powerful ideas in the history of science. That a brain is, in essence, a computer. That we can look past its gelatinous flesh and see through to its underlying workings. That these workings are best described in the language of programs and computations, algorithms and representations. There's no question that this "brain as computer" analogy has been—in many ways—enormously generative. But in other ways could it be leading us astray? My guest today is Dr. Mazviita Chirimuuta. Mazviita is a Philosopher of Mind and Neuroscience at the University of Edinburgh, in Scotland. She's the author of the widely acclaimed and widely discussed 2024 book, The Brain Abstracted: Simplification in the History and Philosophy of Neuroscience. Here, Mazviita and I talk about the early days of her career spent trying to build computational models of cells in the visual system—and how this informed her later philosophical work. We discuss how neuroscience—indeed all science—involves key simplifying strategies: analogy, mathematization, and reductionism. We talk about the history of the "brain as computer" analogy and the different ways of understanding its status. We consider how a computational understanding of the brain feeds directly into certain ideas about technology—for example, the idea that machines are on the cusp of consciousness or the idea that we may one day soon upload our minds to the digital realm. Along the way, we touch on Marr's levels of analysis, maps and territories, biological naturalism, medium dependence and independence, analog vs digital computing, neuromorphic AI, Moravec's paradox, the lower bounds of cognition, and whether consciousness is more like the game of chess or more like soccer. I learned a lot from this one, folks—and I think you will, too. On to my conversation with Dr. Mazviita Chirimuuta. Enjoy! Notes 7:00 – Dr. Chirimuuta's book, again, is here. You can also read a précis of the book here, which was published al...