In a new study published today in Nature, neuroscientists at UCL’s Sainsbury Wellcome Center investigated the reciprocal interactions between two brain regions that explain visual working memory in mice. The team found that the connectivity between these two working memory sites, the parietal cortex and the premotor cortex, jointly depended on the instantaneous time scale.
“There are many different types of working memory, and for the last 40 years, scientists have been trying to figure out how they are represented in the brain. Sensory working memory, in particular, has been a challenge to study, as during a standard lab. task,” said Dr. Ivan Voitov, “Processes occur simultaneously, such as timing, machine setting, and reward anticipation.”
To address this challenge, SWC researchers have compared working memory-dependent tasks with simpler tasks independent of working memory. In a working memory task, mice were given a sensory stimulus followed by a delay and then had to match the next stimulus to the stimulus they saw before the delay. This means that during the delay period, the mice needed a representation in their working memory of the first stimulus to succeed in the task and receive the reward. In contrast, in a task independent of working memory, the decisions made by mice regarding the secondary stimulus were unrelated to the first stimulus.
By comparing these two tasks, the researchers were able to record the portion of neural activity that relied on working memory rather than normal activity related only to the task environment. They found that most neural activity was unrelated to working memory, and instead representations of working memory were included in a “high-dimensional” activity pattern, meaning that only small fluctuations around the average firing of individual cells carried working memory information together.
To understand how these representations are stored in the brain, neuroscientists used a technique called optogenetics to selectively silence parts of the brain during periods of delay and observe distractions from what the mice remember. Interestingly, they found that silencing working memory representations in either parietal or frontal motor cortical areas resulted in similar deficits in the mice’s ability to recall previous stimuli, implying that these representations were instantaneously dependent on one another during delays.
To test this hypothesis, the researchers disabled one region while recording activity communicated by another region. When the parietal cortex is disturbed, the activity communicated by the frontal motor cortex to the parietal cortex is largely unchanged in terms of average activity. However, working memory activity representation is specifically disabled. This was also true in the reverse experiment, when they disturbed the motor cortex and they looked at the parietal cortex and also noted specific impairments of cortical and cortical communication working memory.
“By recording and processing from long-term circuits in the cerebral cortex, we found that working memory resides in a pattern of dependent activity in interconnected cortical areas, thereby preserving working memory through instantaneous mutual communication,” said Professor Tom Myrsek-Flugel. D., director of the Sainsbury Wellcome Center and co-author of the paper.
The researcher’s next step is to look for patterns of activity that are common in this area. They also plan to study more complex working memory tasks that modulate specific information stored in working memory as well as its strength. To this end, neuroscientists will use a disruptive scatterer that contains sensory information to refract what the rat thinks is its next target. This experiment will allow them to develop a more accurate understanding of working memory representations.
–
