The Human Connectome Project
In the 1970s, a Cambridge University study completed a wiring diagram of the nervous system of a tiny worm, Caenorhabditis elegans, mapping the connections between its 302 neurons. Such wiring diagrams, called “connectomes”, represent the web of nerve-fibres that criss-cross the brain. The connectome of the worm took over a dozen years of tedious manual labour to complete, but now the Human Connectome Project is tackling a much more ambitious project: constructing maps of the millions of kilometres of neural connections in the human brain. This would be a painstaking manual process, requiring up to 10 researchers to analyse endless electron microscopic images of brain slices and trace the connections. It could literally take thousands of years to complete—and so the project is wisely accelerating the process by teaching vastly powerful computers to analyse the brain slices using a technique called “automated machine learning”, which lets them modify their behaviour in response to new data. Humans will still need to proofread the computers’ work, but it’ll speed up the process exponentially. “No one thinks this is going to produce a wiring diagram like you might have for the electricity in your house,” says Thomas Insel, director of the National Institute of Mental Health—but even a crude map could vastly advance our understanding of the brain, and help us to understand how neural connections give rise to intelligence, personality and memory.