Resume: Mental maps in the brain are activated when thinking about sequences of experiences, even without physical movement. In an animal study, they found that the entorhinal cortex houses a cognitive map of experiences, which is activated during mental simulation.
This is the first study to demonstrate the cellular basis of mental simulation in a non-spatial domain. The findings could advance our understanding of brain function and memory formation.
Key Facts:
- Mental maps are created and activated without physical movement.
- The entorhinal cortex contains cognitive maps of experiences.
- This study provides insight into the cellular basis of mental simulation.
Source: MIT
As you travel your usual route to work or the grocery store, your brain uses cognitive maps stored in your hippocampus and entorhinal cortex. These maps store information about the trails you’ve traveled and locations you’ve been to before, so you can navigate when you go there.
New research from MIT has found that such mental maps are also created and activated when you just think about sequences of experiences, without any physical movement or sensory input.
In an animal study, researchers found that the entorhinal cortex houses a cognitive map of what animals experience while using a joystick to scroll through a series of images. These cognitive maps are then activated when thinking about these sequences, even when the images are not visible.
This is the first study to demonstrate the cellular basis of mental simulation and imagination in a non-spatial domain through activation of a cognitive map in the entorhinal cortex.
“These cognitive maps are used to perform mental navigation, without any sensory input or motor output. We can see a signature of this map presenting itself as the animal mentally goes through these experiences,” said Mehrdad Jazayeri, associate professor of brain and cognitive sciences, member of MIT’s McGovern Institute for Brain Research, and senior author of the study.
Research scientist Sujaya Neupane of the McGovern Institute is the lead author of the paper, which will appear in Nature. Ila Fiete, professor of brain and cognitive sciences at MIT, member of MIT’s McGovern Institute for Brain Research, director of the K. Lisa Yang Integrative Computational Neuroscience Center, is also an author of the paper.
Mental maps
Much work in animal models and humans has shown that representations of physical locations are stored in the hippocampus, a small seahorse-shaped structure, and the nearby entorhinal cortex. These representations are activated when an animal moves through a space it has been in before, just before it crosses the space or when it sleeps.
“Most previous studies have focused on how these areas reflect the structures and details of the environment as an animal physically moves through space,” says Jazayeri.
“When an animal moves in a room, its sensory experiences are nicely encoded by the activity of neurons in the hippocampus and entorhinal cortex.”
In the new study, Jazayeri and his colleagues wanted to investigate whether these cognitive maps are also built and then used during purely mental walkthroughs or imagining movement through non-spatial domains.
To investigate that possibility, the researchers trained animals to use a joystick to follow a path through a series of images (“landmarks”) separated by regular time intervals. During training, the animals were shown only a subset of pairs of images, but not all pairs. After the animals learned to navigate through the training pairs, the researchers tested whether the animals could handle the new pairs they had never seen before.
One possibility is that animals do not learn a cognitive map of the sequence, but instead solve the task using a memorization strategy. If so, they are expected to wrestle the new pairs. If the animals had to rely on a cognitive map instead, they should be able to generalize their knowledge to the new pairs.
“The results were unequivocal,” says Jazayeri. “Animals were able to mentally navigate between the new pairs of images from the first time they were tested. This finding provided strong behavioral evidence for the presence of a cognitive map. But how does the brain draw up such a map?”
To answer this question, the researchers recorded from individual neurons in the entorhinal cortex while the animals performed this task.
Neural responses had a striking feature: As the animals used the joystick to navigate between two landmarks, neurons showed noticeable bumps in activity related to the mental representation of the intervening landmarks.
“The brain experiences these bursts of activity at the expected time when the intervening images would have passed by the animal’s eyes, which never happened,” Jazayeri says.
“And the timing between these bumps was, critically, exactly the timing that the animal expected to reach each of these bumps, which in this case was 0.65 seconds.”
The researchers also showed that the speed of the mental simulation was related to the animals’ performance on the task: if they were slightly late or early in completing the task, their brain activity showed a corresponding change in timing.
The researchers also found evidence that the mental representations in the entorhinal cortex do not encode specific visual features of the images, but rather the arrangement of the landmarks.
A model of learning
To further investigate how these cognitive maps might work, the researchers built a computational model to mimic the brain activity found and demonstrate how it could be generated.
They used a type of model known as a continuous attraction model, which was originally developed to model how the entorhinal cortex tracks an animal’s position as it moves, based on sensory input.
The researchers modified the model by adding a component that could learn the activity patterns generated by sensory input. This model could then learn to use these patterns to reconstruct those experiences later, when there was no sensory input.
“The most important element we had to add is that this system has the ability to learn bidirectionally by communicating with sensory input. Through the associative learning that the model goes through, it will actually recreate these sensory experiences,” says Jazayeri.
The researchers now plan to investigate what happens in the brain when the landmarks are not evenly spaced, or when they are arranged in a ring. They also hope to record brain activity in the hippocampus and entorhinal cortex as the animals first learn to perform the navigation task.
“Seeing how the memory of the structure is crystallized in the mind, and how that leads to the neural activity that arises, is a really valuable way to ask how learning happens,” says Jazayeri.
Financing: The research was funded by the Natural Sciences and Engineering Research Council of Canada, the Québec Research Funds, the National Institutes of Health and the Paul and Lilah Newton Brain Science Award.
About this memory research news
Author: Abby Abazorius
Source: MIT
Contact: Abby Abazorius – MIT
Image: The image is credited to Neuroscience News
Original research: Closed access.
“Vector production via mental navigation in the entorhinal cortex” by Mehrdad Jazayeri et al. Nature
Abstract
Vector production via mental navigation in the entorhinal cortex
A cognitive map is an appropriately structured representation that allows new computations based on previous experiences; for example, planning a new route in a known space. Research in mammals has found direct evidence for such representations in the presence of exogenous sensory input in both spatial and non-spatial domains.
Here we tested a fundamental postulate of the original cognitive map theory: that cognitive maps support endogenous computation without external input.
We made recordings from the entorhinal cortex of monkeys during a mental navigation task that required the monkeys to use a joystick to produce one-dimensional vectors between pairs of visual landmarks without seeing the intervening landmarks.
The monkeys’ ability to perform the task and generalize to novel pairs indicated that they relied on a structured representation of the landmarks. Task-modulated neurons showed periodicity and slope that matched the temporal structure of the landmarks and showed signatures of continuous attractor networks.
A continuous attraction network model of path integration, supplemented with a Hebbian-like learning mechanism, provided an explanation for how the system could endogenously recall landmarks.
The model also made an unexpected prediction that endogenous landmarks temporarily slow path integration, resetting dynamics and thereby reducing variability. This prediction was confirmed in a reanalysis of firing rate variability and behavior.
Our findings link the structured activity patterns in the entorhinal cortex to the endogenous recruitment of a cognitive map during mental navigation.