Dmitri Chklovskii outlines how single neurons may act as their own optimal feedback controllers
From logical gates to grandmother cells, neuroscientists have employed many metaphors to explain single neuron function. Chklovskii makes the case that neurons are actually trying to control how their outputs affect the rest of the brain.
Dmitri Chklovskii, group leader of the Neural Circuits and Algorithms group at the Flatiron Institute Center for Computational Neuroscience, believes every single neuron in your brain may act as an optimal feedback controller. According to this idea, every time a neuron sends a spike, it then listens to its own inputs to determine whether that spike had the desired effect—akin to listening for a distorted echo in a deep canyon. If the neuron doesn’t like what it hears, its job is to adjust when to send another spike. That implies that every neuron is trying to control the whole system to match what it wants to hear. In this “Brain Inspired” episode with Paul Middlebrooks, Chklovskii discusses how neurons might pull this off, outlines some evidence supporting his theory and explains what it means in the larger context of understanding brain function. (The Flatiron Institute is funded by the Simons Foundation, The Transmitter’s parent organization.)
Read the transcript.