Scientists at UCL’s Sainsbury Wellcome Centre have found a way for the neocortex and thalamus to work together to find differences between what animals think will happen in their surroundings and what actually does. These mistakes in predictions are caused by selectively boosting sensory information that isn’t predicted.
These results help us learn more about how the brain makes predictions and might help us figure out how brain systems are changed in people with autism spectrum disorders (ASDs) and schizophrenia spectrum disorders (SSDs).
A study that was just released in Nature talks about how scientists studied mice in a virtual reality world to learn more about prediction error signals in the brain and how they develop.
Uncovering Neural Mechanisms
“Our brains are always making guesses about what will happen in the world and how our actions will turn out.” Different parts of the brain are strongly activated when these predictions turn out to be wrong. These prediction error signs are important for helping us learn from our mistakes and make better predictions in the future.
Despite how important they are, surprisingly little is known about the neural circuit processes that make them work in the brain, according to the paper’s corresponding author and Group Leader at SWC Professor Sonja Hofer.
Researchers put mice in a virtual reality world where they could find their way along a familiar hallway to get to a prize. This was done to see how the brain handles expected and unexpected events. The team was able to exactly control visual input and put up strange images on the walls thanks to the virtual environment.
Also see :- Scientists Discover Peculiar New Species of Dragonfish in Antarctica
Researchers used a method called two-photon calcium imaging to record the activity of many neurons in the primary visual cortex. This is the first part of the neocortex that receives information from the eyes about what we see.
Enhanced Understanding Through Experiments
“Earlier theories said that prediction error signals show how the actual visual input is different from what was expected, but to our surprise, we couldn’t find any experimental evidence for this.” Instead, we found that the brain strengthens the reactions of neurons that like seeing things that aren’t expected the most.
Because of this selected amplification of visual information, we see the error signal. “This means that our brain looks for differences between what we thought would happen and what actually happened to make unexpected events stand out,” said Dr. Shohei Furutachi, senior research fellow in the Hofer and Mrsic-Flogel labs at SWC and study’s first author.
The team used a method called optogenetics to turn off or on different groups of neurons in order to figure out how the brain amplifies the unexpected sensory information in the visual cortex.
They discovered two groups of neurons in the visual cortex that were key to the prediction error signal: inhibitory interneurons that express vasoactive intestinal polypeptide (VIP) in V1 and the pulvinar, a part of the thalamus that brings together data from many neocortical and subcortical areas and is closely linked to V1. The researchers were surprised to find that these two groups of neurons work together in a strange way.
Collaborative Neural Dynamics
“Neuroscience tends to study one part of the brain or pathway at a time.” But because I studied molecular biology, I was interested in how different molecular paths work together to make regulation flexible and dependent on the situation. Dr. Furutachi stated, “I chose to test the idea that VIP neurons and the pulvinar might be working together at the level of neural circuits.”
In fact, Dr. Furutachi’s research showed that VIP neurons and pulvinar work together in a way that is very helpful. When VIP neurons are off, the pulvinar slows down activity in the neocortex.
But when VIP neurons are on, the pulvinar can strongly and specifically boost sensory responses in the neocortex. So, the sensory prediction error messages in the visual cortex are controlled by how these two pathways work together.
Future Research and Implications
The next thing the team needs to do is figure out how and where in the brain the animals’ predictions are compared with the real sensory information to find errors in the predictions and how these errors affect learning. They are also thinking about how their results could help people understand ASDs and SSDs better.
This idea says that both ASDs and SSDs might be caused by a mismatch in the prediction error system. “Now we are trying to use our discovery on model animals with ASDs and SSDs to study the mechanistic neural circuits that cause these disorders,” Dr. Furutachi said.
