Researchers have found out how two specific regions of the brain, the neocortex and thalamus, collaborate to notice differences between what animals anticipate and what actually happens around them. These “prediction errors” occur when unexpected sensory information is selectively enhanced. This research deepens our comprehension of how the brain processes predictions and could also shed light on the alterations in brain circuits seen in autism spectrum disorders (ASDs) and schizophrenia spectrum disorders (SSDs).
Researchers have identified the collaborative function of the neocortex and thalamus in recognizing discrepancies between expected and actual experiences in animals. This process involves the selective enhancement of unexpected sensory inputs to handle prediction errors. The study increases awareness of predictive processing in the brain and could provide valuable insights into changes in brain circuits associated with autism spectrum disorders (ASDs) and schizophrenia spectrum disorders (SSDs).
A study published today in Nature details how scientists from the Sainsbury Wellcome Centre at UCL examined mice in a virtual reality setup, progressing our understanding of prediction error signals in the brain as well as the underlying mechanisms that produce them.
“Our brains are in a constant state of predicting what we should expect from our surroundings and the results of our actions. When these predictions are incorrect, various brain regions are activated significantly. These prediction error signals are crucial for our learning from errors and for refining our expectations. Despite their significance, there is a surprising lack of knowledge regarding the neural circuits that facilitate these signals in the brain,” explained Professor Sonja Hofer, the Group Leader at SWC and one of the paper’s authors.
To investigate how the brain interprets expected versus unexpected occurrences, the researchers placed mice in a virtual reality scenario where they could traverse a known corridor to earn a reward. The virtual setup allowed the researchers to meticulously manipulate visual inputs and insert unexpected images into the environment. Utilizing a method known as two-photon calcium imaging, they captured the neural activities of numerous individual neurons in the primary visual cortex, the first part of the neocortex that processes visual data received from the eyes.
“Existing theories suggested that prediction error signals represent the difference between actual visual inputs and predictions, but we surprisingly found no evidence supporting this view. Instead, we discovered that the brain amplifies the responses from neurons that exhibit a strong preference for unexpected visual inputs. The prediction error signals we observed are an outcome of this specific enhancement of visual information. This suggests that our brain identifies differences between predictions and real inputs to emphasize unexpected occurrences,” noted Dr. Shohei Furutachi, a Senior Research Fellow in the Hofer and Mrsic-Flogel laboratories at SWC and the principal author of the study.
To grasp how the brain enhances unexpected sensory information within the visual cortex, the team employed optogenetics to activate or deactivate distinct groups of neurons. They identified two neuron groups significant in generating prediction error signals in the visual cortex: vasoactive intestinal polypeptide (VIP)-expressing inhibitory interneurons in V1 and a thalamic region called the pulvinar, which integrates data from several neocortical and subcortical areas and has strong connections to V1. Interestingly, the interaction between these two neuron groups was unexpected.
“In neuroscience, we often focus on analyzing single brain regions or pathways independently. However, coming from a molecular biology background, I was intrigued by how various molecular pathways work in harmony to ensure flexible and contextual regulation. I wanted to investigate whether there could be cooperation occurring within neural circuits, particularly between VIP neurons and the pulvinar,” Dr. Furutachi explained.
Dr. Furutachi’s research showed that VIP neurons and the pulvinar work together harmoniously. VIP neurons function like a switchboard: when inactive, the pulvinar undermines activity in the neocortex, but when VIP neurons are active, the pulvinar can significantly and selectively enhance sensory responses in the neocortex. Therefore, the collaborative effect of these two pathways mediates sensory prediction error signals within the visual cortex.
The next phase for the researchers involves examining how and where in the brain the predictions made by animals are contrasted with actual sensory input to calculate sensory prediction errors and how these signals promote learning. They are also investigating how their discoveries could assist in understanding ASDs and SSDs.
“It has been suggested that both ASDs and SSDs might stem from an imbalance in the prediction error system. We are now applying our findings to models of ASDs and SSDs in animals to study the mechanistic neural circuit foundations underlying these disorders,” Dr. Furutachi stated.
This research was supported by the Sainsbury Wellcome Centre Core Grant from the Gatsby Charity Foundation and Wellcome (219627/Z/19/Z and 090843/F/09/Z); a Wellcome Investigator Award (219561/Z/19/Z); the Gatsby Charitable Foundation (GAT3212 and GAT3361); the Wellcome Trust (090843/E/09/Z and 217211/Z/19/Z); the European Research Council (HigherVision 337797; NeuroV1sion 616509); the SNSF (31003A 169525); and core funds from Biozentrum (University of Basel).
