Computational Neuroscience

Computational neuroscience is the science that asks, for every known neural system, what it computes and how. Its founding conviction—that the brain can be understood as an information-processing system, subject to the same mathematical analysis as an engineered network—did not exist as an organized discipline before a generation of researchers in the 1980s made it one. Terrence Sejnowski was central to that making: he moved to the Salk Institute, founded the journal Neural Computation in 1989, and wrote The Computational Brain with the philosopher Patricia Churchland, a book that helped define what the new science was for. The field's power is its two-sidedness. A neuroscientist who finds a pattern of connectivity in cortex can ask what computation that pattern would perform; an engineer who finds an algorithm that works in an artificial network can ask whether the brain might use something similar. Sejnowski has spent his career moving back and forth across this bridge, treating a good idea as something that should illuminate both