Speaker

Bing Brunton

Bing Brunton

Bing Brunton is a computational neuroscientist and Professor of Biology at the University of Washington, known for reading large-scale neural connectivity as a computation — turning wiring maps into working simulations of how brains control bodies.

Trained at the intersection of physics, applied mathematics, and biology, Brunton describes herself as a computationalist: by the time neural data reaches her it is a giant connectivity matrix, sparse and structured, to be simulated rather than measured cell by cell. Her work sits at the frontier opened by the newly completed connectomes of small animals — the 300-neuron worm C. elegans and, more recently, the fruit fly Drosophila — and asks the harder question that follows the mapping: what can a wiring diagram actually explain? In a long collaboration with the fly experimentalist John Tuthill, she has used brute-force simulation to locate functional circuits inside the fly’s ventral nerve cord.

Core positions

Brunton argues that a connectome is necessary but far from sufficient: the matrix omits cell identity, biophysics, and the context in which a signal is received, and having a map is not the same as understanding it — the worm’s wiring has been known for decades and is still opaque, because much of its computation is chemical and mechanical. Her signature stance is that the brain does not live in a jar: nervous systems evolved to control specific bodies moving through the world, so simulated connectomes must be embodied in physically realistic models to be understood. She is a vocal sceptic of overclaimed ‘brain uploads’ — her demonstration that a worm connectome can drive a fly body via deep learning shows that behaviour fidelity is no evidence of biological fidelity, and that honest interfaces, not lifelike animation, are what make a model meaningful.

In the wiki