How does the brain prepare for something good? The University of Tokyo identifies a region linking reward expectancy and action

People generally predict what will happen next from the situation at hand and act based on those predictions. When the brain determines that something good is about to happen, a state known as reward expectancy, through what neural circuit is this information translated into action?

A research group including Professor Masanori Matsuzaki (concurrently Team Director at RIKEN Center for Brain Science), Project Assistant Professor Keisuke Sehara, and Assistant Professor Masashi Kondo of the Graduate School of Medicine at the University of Tokyo; Research Advisor Tetsuo Yamamori at the Central Institute for Experimental Animals (at the time of research; currently Team Leader at RIKEN Center for Brain Science); and Professor Masahiko Takada at the Primate Research Institute at Kyoto University, discovered a hub region in the brain that coordinates action according to reward expectancy. The results were published in Communications Biology.

Using classical conditioning, as famously demonstrated by Pavlov's dogs, the researchers trained marmosets on a task in which a reward (sugar water) was delivered shortly after a sound cue. During training, two different sound cues were used: one cue (Sound A) was followed by a reward 70% of the time, whereas the other cue (Sound B) was followed by a reward only 30% of the time. This allowed the researchers to detect brain activity associated with different levels of reward expectancy in the animals

Previous studies of reward expectancy in primates relied primarily on electrode recordings that measure neural activity at discrete locations. By contrast, using wide-field calcium imaging, the research team became the first to identify across the marmoset brain which regions show neural activity associated with reward expectancy.

In this imaging method, calcium imaging was performed over the widest range of brain regions (136.9 mm²) in marmoset research to date, covering not only regions governing audition, somatosensation, and movement, but also regions said to possess more complex functions, such as those related to attention-aligned eye movements.

When investigating brain activity related to reward expectancy, because "the animal moving its body" leads to brain activity over a wide area, there is a problem that it is difficult to extract signals involved in "reward expectancy" itself. Therefore, by recording physical activity in detail and using mathematical models to remove its influence from brain activity, they clarified brain activity that had been difficult to see until now.

As a result, they discovered that in a part of the region previously thought to exclusively govern motor planning (premotor cortex), information about the presented sound is held until the reward timing. From more detailed analysis, they found the possibility that this region acts as a "hub" managing information traffic among various brain regions only between the sound stimulus and the reward timing.

Compared to previous studies using mice, the marmoset brain, which is a primate like humans, showed more noticeable division of functions than the rodent mouse brain. This is expected to lead to an understanding of the basic principle of "how humans judge situations and act" in the future.

The image data analyzed in this study is open to the public on the Brain/MINDS Data Portal (https://doi.org/10.24475/bminds.ci.12650) and can be freely downloaded and analyzed under the Creative Commons Attribution 4.0 International Public License.

Journal Information
Publication: Communications Biology
Title: Medial premotor cortex as a potential reservoir of reward expectancy signals influencing the primate frontoparietal network
DOI: 10.1038/s42003-026-10672-4