Thread by @martripodi

New work from our lab @MRC_LMB led by Laura Masullo and Letizia Mariotti reveals the existence of a discrete and modular representation of the motor space in the superior colliculus, in @CurrentBiology @CellPressNews . Funded by @ERC_Research @The_MRC
bit.ly/2NGc5ea

The superior colliculus (SC) is a midbrain region canonically divided in sensory and motor domains. Neurons in the motor domain of the SC (mSC) direct head movements in mice as shown by @jonjwilson from our lab last year: bit.ly/2ZBjKNk

With this new work we show that neurons in the mSC be further classified in several functional types based on their electrophysiological properties and that the expression of the transcription factor Pitx2 pinpoints a single functional type of mSC neurons.

Functionally, optogenetic activation of collicular Pitx2 expressing neurons leads to rapid head movements along the three principal axis of rotation (yaw: left-right; pitch: up-down; roll: clockwise and anticlockwise).

Increasing stimulation duration produces concatenated movements of fixed amplitude. In essence, there is a quantal packet of motion that a given set of Pitx2 neurons dispatch. If the SC were a piano and Pitx2 neurons the keys, pressing a given key always produce the same note.

Pitx2 neurons within a module are spontaneously active when the animal is about to produce a head movement of a given amplitude. Interestingly, the amplitude of motion predicted by their activity is approximately the same as the one elicited by their optogenetic stimulation.

Intriguingly, Pitx2 neurons are arranged in an orderly array of discrete modules, shaped like hollow barrels, which tile the entire motor SC. So it seems that, after all, Pitx2 neurons do form a sort of piano keyboard.

What happens if we play different keys? It turns out that the activation of distinct Pitx2 modules produces movement of different amplitudes. The amplitude of the produced movement increases in an orderly and predictable manner across the surface of the SC

So what’s modularity good for? Pitx2 modules are a site of integration of multisensory signals. A bit of speculation here, but, Pitx2 modules might define a site of convergence for spatially coherent signals so that perceptual space can be meaningfully mapped on action space.

While still speculative, there are interesting implications of this interpretation. One implication in particular strides with deeply rooted conception of the continuity of the perceptual space.

From a motor perspective, space might actually be discretely represented and the resolution of such a spatial-motor representation would then depend on the density of these modules.

Network mechanisms leading to a weighted concomitant recruitment of distinct modules would potentially smooth such a granular representation so to finally approximate the perceived spatial-motor continuity.

Here the graphical summary and I’ll leave you to the actual paper