Our cerebral cortex is responsible for our high-level sensory processing, including what decisions we make and how we enact them. So far, our understanding of the cortex is shaped by its distinct parts that are responsible for different functions and senses, such as the auditory cortex being responsible for hearing and the visual cortex looking after what we see. Numerous different cell types respond to various stimuli in an enormous amount of activity within the cortices, yet the top layer of cells are conversely inactive, seemingly unresponsive. A persistent theory has been that these ‘quiet’ neurons are extremely selective and respond only to specific complicated set of stimuli. A number of studies have searched for a perfect combination to activate them, only to find the cells remain unresponsive in all cases. However, a new paper from the Bruno Lab has found that regardless of the type of stimuli, it is in fact unexpected or surprising events that drives these cells.
Research team Dr Rebecca Rabinovich, Dr Daniel Kato and Professor Randy Bruno focused on investigating the ‘quiet’ cells of the somatosensory cortex, primarily associated with touch, by conducting a series of object detection tasks with mice. They observed that mouse neurons remained unresponsive during repeated exposure to stimuli without reward, and became more responsive when they were rewarded. However, crucially, they found that the responses were stronger when there were unexpected deviations in the timings of the trials.
Professor Bruno said: “We finally hit on something that amped up the activity of these cells really effectively, while not actually changing anything about what the cells are exposed to. We would give the animals the same stimuli but, for instance, delay moving it by a split second, which the animal doesn’t expect. The animal is surprised, and the cells switch on.
“This suggests that in the case of the somatosensory cortex, touch might not be what activates these cells the most. Perhaps it’s actually what you could call ‘behavioural time’. What’s true of everything we do is that we have a sort of model where we follow actions in a certain order, and we have certain expectations of how and when these will play out. So where is that represented in the brain? It’s probably in a very frontal area of the brain but also in these mysteriously ‘quiet’ cells that actually react to surprising circumstances.
“The interesting thing here is that people have always thought that the different areas of the cortex are just dedicated to their specific function, for example, the somatosensory cortex just handles touch; the visual cortex just handles vision. But we show that information about behavioural time, what the animal is expecting, what the animal is planning to do, is also encoded in all these primary cortices, so they’re clearly not just dedicated to their expected functions. The somatosensory cortex has access to information all over the brain about behavioural time. Therefore, it is not touch that is most effective in activating these cells, it is unexpected touches or unexpected events – the notion of surprise. This is probably a very general property of all different parts of the cortex.”
The next step in this line of research for Professor Bruno and his team is to test the extent to which these findings are relevant in the context of attention and learning. Professor Bruno said: “When our brain is dealing with the unexpected, circuits might be more plastic and allow learning to take place. It’s likely, therefore, that they are involved in both learning and attention, so it would be interesting to manipulate the cells during different attention and learning tasks to assess the degree to which behaviour can be altered.”
The full paper “Learning enhances encoding of time and temporal surprise in mouse primary sensory cortex” is available to read in Nature Communications.