A new framework assesses the costs of control in the brain

The brain performs various cognitive and behavioral functions in daily life, flexibly switching to various states to perform these functions. Scientists view the brain as a system that performs these many functions by controlling its states. To better understand the properties of this control in the brain, scientists are looking for ways to estimate the difficulty of control, or the cost of control, when the brain switches from one state to another. A team of researchers therefore undertook a study to quantify these costs of control in the brain and succeeded in building a framework that evaluates these costs.

Controlling transitions to some states entails greater “costs” than controlling transitions to others. With the development of a framework to quantify transition costs, scientists will have a way to assess the difficulty of shifts between various brain states. They may also have a quantifiable measure to explain cognitive loads, sleep-wake differences, habituation to cognitive tasks, and psychiatric disorders.

The book is published in the Journal of Neuroscience.

The team worked to build a new framework for quantifying the cost of control that takes into account the stochasticity, or randomness, of neural activity. This stochasticity has been ignored in previous studies. The current control paradigm in neuroscience uses a deterministic framework unable to account for stochasticity. But it is well known that neural dynamics are stochastic and that noise is pervasive throughout the brain. “In this work, we addressed the issue of stochasticity and first proposed a new theoretical framework that quantifies the cost of control by taking into account stochastic fluctuations in neural dynamics,” said Shunsuke Kamiya, PhD student at the School. college of arts and sciences. at the University of Tokyo.

In their study, the researchers established the analytical expression for stochastic control cost, which allowed them to calculate the cost in high-dimensional neural data. Through analytic expression, they found that the optimal control cost can be decomposed into mean and covariance control costs. “This decomposition allows us to study how different brain areas contribute differently to controlling transitions from one brain state to another,” Kamiya said.

The researchers also identified brain regions important for optimal control of cognitive tasks in human brain imaging data. They examined the brain regions important in the optimal control of transitions from the resting state to seven cognitive task states, using human brain imaging data from 352 healthy adults. They found that with these different transitions, the lower visual areas generally played an important role in controlling the means, while the posterior cingulate cortex generally played an important role in controlling the covariances. The posterior cingulate cortex is the upper part of the limbic lobe, the region of the brain that plays an important role in memory and emotional behaviors.

In this study, the team considered only the optimal control cost where brain state transitions are optimally controlled, with a minimization of the stochastic control cost. However, in real neural systems, state transitions are unlikely to be optimally controlled.

An intriguing future direction will be to compare optimally controlled dynamics and actual dynamics using neural data during tasks. »

Masafumi Oizumi, Associate Professor, Graduate School of Arts and Sciences, University of Tokyo

Looking ahead to future research, Oizumi says the ultimate goal of his lab is to understand the connection between brain dynamics and human behaviors, cognitions and consciousness. “For example, we suspect that decreased controllability of brain dynamics may be related to mental fatigue or loss of consciousness. We expect the theoretical perspective of control to provide new insight into this goal,” Oizumi said.