A Nobel Prize-Winning Scientist Built an Invisible Maze for Rats

The honeycomb maze is better than the alternative, both ethically and scientifically.
January 25, 2018, 4:06pm
Image: O'Keefe et. al./UCL

Nobel Prize-winning researcher John O’Keefe and his colleagues at University College London have built a better mouse trap. Well, technically it’s a maze, but either way it’s cool as hell.

As detailed in Nature, the honeycomb maze consists of 37 hexagonal platforms that that rise and fall to make a path for the rat. If the rat walks to the right edge of the platform, another platform rises so it can move forward. If it walks to the wrong edge of the platform, no platform rises and it has to turn back. The researchers observe the rodent’s choice at each juncture to determine how it makes choices as it navigates the maze.

According to O’Keefe, who won a Nobel Prize in 2014 for discovering the type of neuron necessary for spatial sense, it’s an improved way to study spatial reasoning in rodents.

There are a number of different mazes that researchers use to study the way animals recognize their location and move themselves toward a goal or reward. There’s T-mazes, radial arm mazes, and Barnes mazes, although for decades the golden standard among behavioral neuroscientists has been the Morris water maze.

In experiments using a Morris water maze, a rodent is placed in a pool of water that also includes an invisible platform hidden just below the surface of the water. The rodent’s ability to find the submerged platform during and across trials provides insights into spatial memory and how the animal learns.

“The Honeycomb Maze replicates all the advantages of the justly-famous Morris water maze in that it forces the animal to approach a hidden goal from different directions,” O’Keefe said in a statement. “In addition, by forcing the animal to choose between two alternatives, it provides a simple measure of success at each point in the maze.”

When a rat enters O’Keefe’s new honeycomb maze, it is presented with two possible paths on each new platform which are raised to the same height along two of the hexagon’s six sides. The goal of the rodent is to choose the path that will bring it to the reward (a piece of food placed on another hexagonal platform).

According to O’Keefe, this approach to studying spatial sense in rodents is an improvement over the water maze because it allows researchers to control the choices the rodents make at each step in the maze. This means that researchers have a more granular view of how the rat understands its position on the platform. Importantly, it also gives O’Keefe and his team the ability to study neuronal activity in the rodents as they navigate the honeycomb, something that was prohibitively difficult to do in a water maze.

After testing the maze with three control groups of rats, the researchers noticed that three main variables affected how the rats navigated the honeycomb platforms.

The first had to do with the angle of the two platforms that are presented as a choice to the rats, which were better able to navigate the maze if the two choices were further apart rather than close to one another. The second factor was how far away the reward was from the rats in the maze: the further the reward, the worse the rats’ performance. Finally, the researchers noticed that rats’ performance decreased when the angle of a choice platform relative to the goal was increased.

“These are exciting times in systems neuroscience,” O’Keefe said. “We see the maze as a general-purpose behavioural testing apparatus, which will enable us to study other spatial behaviours such as direction-following as well as non-spatial tasks such as approaching a moving object.”