Untangled, context-dependent persistence makes it easier to encode valuable signals in the brain
In 2019, researchers at the University of California at San Diego discovered the area of the brain where “value decisions” are made.
An area in the brain known as the retrosplenial cortex (RSC), they discovered, is the site we use to make valuable choices, such as which restaurant we decide to visit for tonight’s dinner. We then update the RSC with new information based on new impressions of how much we enjoyed the evening’s soup and pasta.
New research led by Division of Biological Sciences postdoctoral researcher Ryoma Hattori and Professor Takaki Komiyama now reveal details of How? ‘Or’ What this dynamic information is processed. The results, published on November 23 in the journal Neuron, show that persistence allows valuable signals to be most efficiently represented, or “encoded”, across different areas of the brain, particularly the RSC.
To study the details of how brain activity represents value-based decision making, a basic animal behavior that is altered in neurological conditions such as schizophrenia, dementia, and drug addiction, the researchers put in instead reinforcement learning experiences in which mice were presented with options and their choices were rewarded with certain probabilities. They recorded the corresponding brain activities during reinforcement learning. The resulting data and network simulations underscored the importance of persistent coding in how mice and their value decisions were represented and RSC as the link for this activity.
“These results suggest that although the encoding of information is highly distributed, not all of the information represented in neuronal activity can be used in every domain,” the authors explain in the article. “These results reveal that the unraveled and context-dependent persistence facilitates reliable coding of the signal and its distribution in the brain.”
Neurons are known to cycle through different patterns of activity, Hattori says, with some neurons increasing their activity and others remaining silent. These brain activity patterns have been shown to correlate with certain task-related information, such as information of value for decision-making. Because the RSC plays a central role in connecting several brain networks and functions, the new findings reinforce ideas about the fundamental importance of the site.
We believe that in the mouse brain, RSC functions as a stable repository of valuable information. RSC appears to distribute valuable information to other areas of the brain that are vital for further processing of valuable signals as mice perform reinforcement learning and decision making. “
Ryoma Hattori, Postdoctoral Fellow, Division of Biological Sciences
To further test their findings, Hattori and Komiyama drew from their treasure chest “big data” of more than 100,000 mouse decisions recorded during the experiments. They programmed artificial intelligence (AI) networks to mimic behavioral strategies in computerized reinforcement trials and found results remarkably similar to real-world experiments.
“When we trained the artificial intelligence network to adopt the same behavior, it adopted the same strategy and the same way of representing information in neural activity,” said Komiyama, professor of neurobiology (Division of Science biological) and neuroscience (Department of Neurosciences, School of Medicine), with affiliations to the Center for Neural Circuits and Behavior at UC San Diego and the Halıcıoğlu Data Science Institute. “This suggests that this is an evolutionary chosen strategy for neural circuits to perform this behavior. This parallel between the biological brain and the AI that Ryoma has formed is really interesting.”
University of California San Diego
Hattori, R., et al. (2021) Context-dependent persistence as an encoding mechanism for robust and widely distributed value encoding. Neuron. doi.org/10.1016/j.neuron.2021.11.001.