Selbst die kleinste Wahl kann das gesamte Gehirn beleuchten.

A new mouse brain dataset created by an international collaborative team shows that the decision-making process mobilizes neuronal activities throughout the brain.

“Even relatively simple decisions recruit a large number of brain regions,” said Anne Churchland, a professor of neurobiology at the University of California, Los Angeles, and one of the founding members of the International Brain Laboratory (IBL).

Churchland pointed out that the traditional classic model suggests that the pathway for vision-based decision-making activities is: from the visual thalamus to the primary visual cortex and its association areas, and then possibly to the prefrontal cortex. However, the new research results show that “there may be more parallel processing in the brain than the linear, layer-by-layer circuit structure we originally thought.”

Churchland and other scientists established the IBL in 2017, stemming from their frustration with previous small-scale decision-making studies, which often analyzed only one or two brain regions at a time. The goal of the IBL is to study on a large scale how the brain integrates information and makes decisions. “We formed a large team because everyone realized that only collective efforts could make a breakthrough in this research problem that has long troubled us,” Churchland said.

After years of effort, the IBL team unified research methods and equipment among 12 laboratories and finally mapped the whole-brain neural activity of mice performing decision-making tasks. This map was recently published in the journal Nature [1]. The results show that neural activities related to choice and motor behavior are almost ubiquitous throughout the brain. Another paper published by the same team in the same issue also found that the decision-making process based on prior knowledge also involves a wide range of brain regions [2].

The IBL laboratories generated this map by integrating Neuropixels neural recording data from 139 mice that completed visual discrimination tasks. Each laboratory recorded different brain regions, but together they covered the entire brain. The final dataset spanned nearly 300 brain regions and included the activities of more than 620,000 neurons, making it one of the most comprehensive datasets on the decision-making process to date. Floris de Lange, a professor of brain, cognition, and behavior at Radboud University who was not involved in the study, commented.

The IBL has made the dataset publicly available [3]. De Lange said that this will allow researchers to “test new scientific hypotheses in a very simple and powerful way.”

Image source: International Brain Laboratory

Challenging Traditional Views

The IBL chose a visual behavior task — it is complex enough to reveal different aspects of the decision-making process and simple enough to be replicated in different laboratories.

In the experiment, mice were presented with a circular pattern with black and white stripes on the left or right side of the screen. They then had to turn a tiny wheel to move the circle to the center of the screen. If the mice successfully completed the task, they received a sip of water; if they failed, they heard white noise as feedback. During this process, electrodes recorded the mice's brain activities from the moment they saw the visual cue, made a choice, turned the wheel, to the final receipt of water or noise feedback.

Once the animals saw the visual cue, activities appeared in some predictable regions of the brain, including the visual cortex, thalamus, and prefrontal cortex.

Neural activities related to choice not only occurred in cortical regions (which is consistent with previous studies) but also in subcortical regions such as the hypothalamus, hindbrain, and cerebellum. This finding challenges the traditional view that “only a few specific brain regions encode decision-making information” and supports the new concept that “decision-related information is widely distributed in the brain.”

“Surprisingly, neural activities related to choice can be seen almost everywhere,” said Long Ding, an associate research professor of neuroscience at the University of Pennsylvania who was not involved in the study.

Signals related to movement and feedback are also widespread throughout the brain: Among the recorded brain regions, 81% contained information that could predict the animal's wheel-turning speed; almost all the recorded brain regions — including those not directly associated with rewards — could accurately predict whether the mice received a reward, with the most intense activities in the thalamus, midbrain, and hindbrain.

When mice saw the circle appear on one side of the screen more frequently during the experiment, they gradually incorporated this “prior information” into their next decision. The research team reported in a second study that this information is widely distributed in 20% to 30% of the brain regions, including the sensory processing regions in the early visual pathway, such as the dorsal lateral geniculate.

These findings overturn the long-held view that “prior information” is only integrated into the process of the higher cortex or decision-making areas in the final step. Churchland said: “The new results suggest that prior information plays a role throughout the decision-making process.”

Overall, these studies suggest that the current models of decision-making mechanisms and their controlling brain regions may be too limited, and other unexplored brain regions may also be equally important.

Although the analysis shows that the distributed network of the brain contains decision-related information even in the early stages of sensory processing, this does not imply a causal relationship. Ding pointed out that future research needs to determine how this information is actually used for the next decision or learning.

De Lange pointed out that this detailed brain map lays the foundation for subsequent experiments and may become a “repository” to help neuroscientists validate results in their own laboratories. Ultimately, these studies highlight the importance of large-scale, multi-laboratory collaborations, especially when studying brain activities.

This international collaborative organization has now expanded to 21 experimental and theoretical neuroscience laboratories and established a new team called IBL 2.0, planning to share their accumulated tools and experience with new partners. Churchland said: “I hope our research can show that when more people come together, they can accomplish tasks beyond the capabilities of a single laboratory and bring crucial insights to the entire field.”

References

[1] International Brain Laboratory., Angelaki, D., Benson, B. et al. A brain-wide map of neural activity during complex behaviour. Nature 645, 177–191 (2025).

[2] Findling, C., Hubert, F., International Brain Laboratory. et al. Brain-wide representations of prior information in mouse decision-making. Nature 645, 192–200 (2025).

[3] https://viz.internationalbrainlab.org/

Original Article

https://www.thetransmitter.org/decision-making/everything-everywhere-all-at-once-decision-making-signals-engage-entire-brain/

This article is from the WeChat official account “Neural Reality” (ID: neureality). Author: Claudia López Lloreda, Translator: EY. Republished by 36Kr with permission.