226 — A gate-and-switch model for head orientation behaviors in C. elegans

1. "C. elegans is simple," they said. "We've got the connectome. You can study the whole brain."
I fell for it, and have been trying to figure out how the same neuron works for 7 years. Our latest incremental step. https://t.co/DhGtyfz3qs

— Michael Hendricks 🌨️🌨️🌨️ (@MHendr1cks) March 29, 2018

I love academic paper threads on Twitter because it’s like the Unplugged version of the science. “This was what we found out! And doing this particular part was a lot of fun!” Eye-hearts emoji.

This paper first discusses the importance of differentiating between external stimuli and “reafferent sensory stimulation” — stimulation that we induce ourselves through an indirect mechanism (for example, turning your head changes visual input even though the scene is unchanged). The circuitry underpinning these responses is poorly understood in general (for example, yesterday I read about how the visual system adapts eye position to counter head orientation changes).

C. elegans is a small wormfriend whose 302(ish) neurons are stereotypic (conserved between individuals), named by scientists, and well-characterized at a systems level. The worm moves by sampling stimuli (e.g. concentration gradients, temperature gradients, etc) in its vicinity and moving toward the most desirable region in a biased random-walk process.

One pair of the 302 neurons are dubbed RIA; there are two RIAs (left and right). Each RIA interneuron has a dorsal and ventral compartment, and RIA has the potential to receive sensory input from all C. elegans sensory modalities, and it also receives motor feedback from head orientation neurons. The researchers determined that RIA must transmit outputs based upon what they dub a “gate-and-switch” model, in which some stimulus based upon incoming sensory information acts as a “gate,” and the motor feedback acts as “switch” in order to alter the response based upon the current posture of the worm.

In contrast with classical models of the neuron as a passive arithmetic operator, this sort of work demonstrates that each individual neuron has its own complex computational machinery. Perhaps in order to understand the brain at a higher level we may need to spend more time focusing on the computational properties and capabilities of individual neurons.