Mathematical Model of Vision %

Ever wondered how do we actually see the world around us? It’s much more complicated than just the retina transferring huge amount of visual information into our brain and biology alone cannot explain the actual process – this is where mathematics bumps in.

Three scientists from New York University, Young, Shapley and Chariker, are on the quest to use the combined concepts of mathematics and biology to create a single mathematical model to explain this seemingly simple phenomenon. The retina links the outside world to the brain, but this connectivity is minimal. The LGN nerve cells, which provide the only pathway through which visual information travels into the brain, send an impulse to the visual cortex whenever light intensity changes. The rest of the work is left for the visual cortex of the brain which scrutinizes each and every bit of that small amount of information it receives.

Transfer of signals by the LGP cells mark the onset of complex neural activity – hundreds of neurons connecting to a single neuron and each of those neurons receiving signal from hundreds of other neurons. Now it has been discovered that the visual cortex is a complicated system of feedback loops while previously it was thought that there is only one way forward for the flow of information.

“Feedback loops are really hard to deal with because the information keeps coming back and changes you, it keeps coming back and affects you,” Young said. As the signals flow through these loops they get amplified and this model with feedback loop was successfully able to reproduce orientation of edges in an object. This was the first step taken on the long road of explaining vision.

Moving forward, in 2018 they explained that the same edge-detection model can be used to reproduce overall pulse activity in visual cortex. Currently, they are working on directional sensitivity to describe how visual cortex detects the direction in which the object is moving. So far their research revolves around only the first of the five layers of visual cortex and you could imagine the amount of work required to completely explain the process of vision.

Although there is a long way to go, this model, the first of its kind will provide a complete picture, explaining vision.