精东传媒

Mind

AI can tell where a mouse is by reading its brain activity

An artificial intelligence can interpret a mouse's brain activity to tell scientists where the animal is located and the direction it is looking

By James Woodford

22 February 2024

A micrograph of a section through a mouse’s brain highlighting neural pathways (green)

MARK AND MARY STEVENS NEUROIMAGING AND INFORMATICS INSTITUTE/SCIENCE PHOTO LIBRARY

Analysing a mouse鈥檚 brain activity tells scientists where the animal is located and the exact direction it is looking. With further research, the findings could one day help robots navigate autonomously.

Mammalian brains use two main types of neurons for navigation: 鈥渉ead direction cells鈥 show where an animal is facing and 鈥済rid cells鈥 help provide a two-dimensional brain map of where it is located.

To learn more about the firing of these neurons, at the University of Tennessee, Knoxville, and his colleagues – together with the US Army Research Laboratory – analysed data from a previous study.

In this experiment, probes were inserted into several mice鈥檚 brains. Data on their neural firing patterns were then paired with video footage showing their locations and head positions as they moved around an open environment.

From this, Maroulas and his colleagues developed an artificial intelligence algorithm that can use neural activity to work out where a mouse is looking and where it is located.

Free newsletter

Sign up to The Daily

The latest on what鈥檚 new in science and why it matters each day.

New 精东传媒. Science news and long reads from expert journalists, covering developments in science, technology, health and the environment on the website and the magazine.

In effect, it is like the drop pin and directional arrow on a smartphone鈥檚 map app, except instead of connecting to a GPS satellite, scientists analyse a subject鈥檚 brain signals.

鈥淭his method allows us to not be dependent on a preloaded map or updating the GPS coordinates based on, say, satellite data,鈥 says Maroulas. 鈥淚n some sense, the algorithm 鈥榯hinks鈥 and recognises space as a mammalian brain would.鈥

The AI could eventually enable intelligent systems to navigate autonomously, he says. 鈥淚n other words, we take advantage of how the mammalian brain processes data and incorporate it in the architecture of the algorithm.鈥

at the Queensland University of Technology in Australia says the smartphone app analogy is a helpful one. 鈥淵ou have positional information (the drop pin) aligned with direction (blue arrow), and during navigation the two are constantly updating as you move. Grid cells are like the GPS and heading cells are like a compass.鈥

Journal reference:

Biophysical Journal

Topics:

Sign up to our weekly newsletter

Receive a weekly dose of discovery in your inbox. We'll also keep you up to date with New 精东传媒 events and special offers.

Sign up
Piano Exit Overlay Banner Mobile Piano Exit Overlay Banner Desktop