This site may earn affiliate commissions from the links on this folio. Terms of use.

What a mouse looks like with an optogenetics system plugged in

Your brain has a built-in GPS organization, with bespoke cells for speed, location, management and more. It's found in the hippocampus, buried deep within the brain. Neurons — called grid cells, identify cells, head direction cells, and boundary cells — all coordinate with one another to orient you in space. Accordingly enough, as far as our neurons are concerned we're all navigating an axonometric (axonometric, get it? Considering neurons have axons? …I'll run across myself out) grid of tiny equilateral triangles, that tessellates the whole surface of the Earth. Since the Nobel prize in medicine for 2014 went to the scientists who figured out this organization of cells in the first identify, neuroscientists have leapt to learn more than about the way it works. Now a group of scientists from Stanford have farther advanced our understanding of the brain'due south spatial orientation organisation — past shaking up our models in a big style.

It started with tracking the manner mouse neurons behaved equally the mice walked around a square box. The Stanford grouping establish that boundary cells help "reset" wayward grid cells, which may underlie the mode stumbling on a familiar spot tin adjust your mental map of a place. But their discoveries were noisy in a style that defied explanations. So they kept on running experiments and mapping neuronal networks.

What they found both shakes upward and sharpens our model of how the brain performs navigation. Most of the neurons the Stanford squad came across encoded more than just i navigational variable; as an example, the group establish grid cells and head-direction cells that also tracked speed. Meanwhile, the speed cells were tuned in weird means. Different the smoothen response curve we might have expected, what the researchers saw was that a speed prison cell might fire when a mouse moved either quickly or slowly, but not at intermediate speeds. And every navigational neuron appeared to respond a piffling differently from every other.

"Nosotros didn't run across grid cells or speed cells or caput-direction cells," said coauthor Surya Ganguli. "We saw this big continuum."

Firmware of the brain

Coauthor Lisa Giocomo said we don't yet have a practiced mathematical model for the brain's navigation system. Existing models make assumptions that simply are not uniform with their results. "We need to rethink basically what the mechanism is," she said.

You may already have heard of the brain-every bit-estimator analogy, but information technology has its shortfalls. There are still plenty of parallels, though, but because of the fundamental structures of their networks. The brain is not similar a PC. The brain is more like an assortment of FPGAs and DSCs, strung together between complex logical clusters. But information technology does have a language common to all neurons. That language can be characterized in terms of timing, chip depth, content, and channels. Furthermore, neurons go on to communicate with many other cells, each in its own linguistic communication. If thoughts are the organized electrochemical patterns that move across the brain, then they're closer to software, and the language of neurons is a level of abstraction deeper: similar firmware.

Graduate students, from left, Kiah Hardcastle and Niru Maheswaranathan worked with professors Surya Ganguli and Lisa Giocomo.

We demand to be cautious in how we pigeonhole the functions of neurons, the authors said. While we humans utilize tools similar compasses and speedometers, a navigational neuron cached in the hippocampus knows of no such thing. It'due south a mistake to blithely presume that neurons call back the same way humans do. The authors found a whole continuum of navigational cells that responded to input we tin can't yet identify. "There were all these cell types that didn't have a name," said coauthor Kiah Hardcastle. "They weren't grid or border, head direction or speed cells, which are the four master types. This started as an extension of previous work, merely then it really took a left turn."

In whatsoever case, there'due south clearly something to the idea that information processing in nature can be discussed in much the same way as we understand other, human-built networks. At that place may exist a need to translate between systems, but that's non stopping anyone. The hippocampus is essential to learning and retentiveness, and its navigational abilities inform the brain'due south physics engine. That means there's a way to translate betwixt those systems. "The variables that the brain cares about may not be the aforementioned as the variables that the mind cares nigh," Ganguli said. "At that place may be a discrepancy between those. And if in that location is, then we somehow accept to suspension free of the prejudices of our mind in order to understand the encephalon."