Scientists Have Connected Rabbit Retinas To Computer Chips To Learn How Our Brains Experience The World

Individually, neurons are highly imprecise, but together they are extremely accurate. nobeastsofierce/Shutterstock
Ben Taub 11 Jul 2016, 17:48

Scientists are one step closer to understanding how the brain converts external stimuli into thoughts, sensations and experiences, after observing how an entire population of neurons in a rabbit’s retina work together to encode the visual world.

In particular, the researchers were interested in observing how the brain compensates for “neuronal noise”. This refers to the phenomenon whereby individual neurons respond differently to the same stimulus on different occasions, as the activity of any given neuron is affected by a range of internal and external factors.

As such, the reliability of a neuron is somewhat compromised, as the precision of its responses to a repeated stimulus is unreliable. Because of this, it has often been assumed that neuronal noise is a bad thing, as it makes it harder for our brains to figure what is actually going on around us.

However, writing in the journal Neuron, the study authors explain that while neuronal noise may tarnish the precision of individual neurons, large populations of neurons may in fact have evolved certain mechanisms that actually allow them to benefit from this response variation.

To test this, they connected rabbit retinas to a multielectrode array containing 11,011 platinum electrodes, enabling them to observe the activity of each neuron in these retinas as they responded to a moving bar of light.

They discovered that the responses of each neuron to the same stimulus varied greatly between repeat trials, meaning that neuronal noise was high. However, the overall output signal given off by this population of neurons was highly precise. Describing this effect, study co-author Felix Franke explained in a statement that “if the stimulus is the number three, then one neuron will perhaps give us a two, and the neuron next to it a four. If we take the average of them both, the answer is correct. Viewed individually, each answer would be incorrect.”

It therefore stands to reason that the higher the number of neurons in a population, the more accurate the average signal will be. Unsurprisingly, therefore, the researchers found neuronal noise to be increasingly beneficial in larger neuron populations than in smaller ones, as it allowed for a more precise average signal to be generated.

Not only does this research suggest that it may in fact be more useful for neuroscientists to examine overall populations of neurons rather than individual neurons when attempting to discover how the brain works, but it could also lead to new forms of treatment and therapy. For instance, Franke claims that “if we can understand how neural networks function, then we can also better understand the diseases that are connected with them.”

Image: Neuroscientists have tended to focus on signals given off by individual neurons, though this new research suggests it may be more useful to examine populations of neurons. Sebastian Kaulitzki/Shutterstock

Comments

If you liked this story, you'll love these

This website uses cookies

This website uses cookies to improve user experience. By continuing to use our website you consent to all cookies in accordance with our cookie policy.