Research in mice by scientists at the Okinawa Institute of Science and Technology Graduate University (OIST) Neural Computation Unit has identified specific areas of the brain that individually promote patience, through the action of serotonin. The team said that increasing our understanding of how different, defined areas of the brain are more or less affected by serotonin could have implications for the development of targeted drug therapies.
“Serotonin is one of the most famous neuromodulators of behavior, helping to regulate mood, sleep-wake cycles, and appetite,” said co-lead research Katsuhiko Miyazaki, PhD. “Our research shows that release of this chemical messenger also plays a crucial role in promoting patience, increasing the time that mice are willing to wait for a food reward.”
Miyazaki, together with co-research lead Kayoko Miyazaki, PhD, and colleagues, report on their findings in Science Advances, in a paper titled, “Serotonergic projections to the orbitofrontal and medial prefrontal cortices differentially modulate waiting for future rewards.”
We’ve probably all experienced times when we need to be patient, perhaps when stuck in traffic at the end of a long day, or waiting for the release of a new book or film. In fact, learning to suppress the impulse for instant gratification can be vital for future success. “Waiting appropriately is often critical in dynamic environments to obtain future rewards,” the authors wrote. However, how patience is regulated in the brain has been poorly understood.
The newly reported work was designed to investigate the role of serotonin more closely. “In the present study, we ask where and how serotonergic projections promote waiting for future rewards,” the investigators explained. Their studies drew on previous research, which applied optogenetics—a technique that uses light to stimulate specific neurons in the brain—to establish a causal link between serotonin and patience.
The scientists bred genetically engineered mice, which had serotonin-releasing neurons that expressed a light-sensitive protein. This allowed the researchers to stimulate these neurons to release serotonin at precise times, by shining light, using an optical fiber implanted in the brain. The results of experiments showed that stimulating these neurons while the mice were waiting for food increased their waiting time, with the maximum effect seen when the probability of receiving a reward was high, but when the timing of the reward was uncertain. “In other words, for the serotonin to promote patience, the mice had to be confident that a reward would come but uncertain about when it would arrive,” Miyazaki stated.
For their previous work, the scientists had focused on an area of the brain called the dorsal raphe nucleus (DRN)—the central hub of serotonin-releasing neurons. “Our previous research revealed a causal relationship between dorsal raphe serotonergic neuron activation and patience while waiting for future rewards,” they noted.
Neurons from the dorsal raphe nucleus reach out into other areas of the forebrain, and for their newly reported study, the scientists explored specifically which of these other brain areas contributed to regulating patience. The team focused on three brain areas that had been shown to increase impulsive behaviors when damaged, including the deep brain structure called the nucleus accumbens, and two parts of the frontal lobe called the orbitofrontal cortex (OFC) and the medial prefrontal cortex (mPFC). “Impulse behaviors are intrinsically linked to patience—the more impulsive an individual is, the less patient—so these brain areas were prime candidates,” explained Miyazaki.
The scientists implanted optical fibers into the dorsal raphe nucleus, and also one of either the nucleus accumbens, the orbitofrontal cortex, or the medial prefrontal cortex, in their experimental mice. “In this study, using optogenetic stimulation of serotonergic terminals, we examined which DRN serotonergic projection target areas promote waiting,” they wrote. They trained the mice to perform a waiting task, whereby the animals held their nose inside a hole—a “nose poke”—until a food pellet was delivered. The scientists rewarded the mice in 75% of trials. In some test conditions, the timing of the reward was fixed at six or ten seconds after the mice started the nose poke and in other test conditions, the timing of the reward varied. In the remaining 25% of trials, called the omission trials, the scientists did not provide a food reward to the mice. They measured how long the mice continued performing the nose poke during omission trials—in other words, how patient the animals were—when serotonin-releasing neurons were and were not stimulated.
When the researchers stimulated serotonin-releasing neural fibers that reached into the nucleus accumbens, they found no increase in waiting time, suggesting that serotonin in this area of the brain doesn’t have a role in regulating patience. But when the scientists stimulated serotonin release in the orbitofrontal cortex and the medial prefrontal cortex while the mice were holding the nose poke, they found the mice waited longer, with a few crucial differences. In the orbitofrontal cortex, release of serotonin promoted patience as effectively as serotonin activation in the dorsal raphe nucleus; both when reward timing was fixed and when reward timing was uncertain, with stronger effects in the latter. But in the medial prefrontal cortex, the scientists only saw an increase in patience when the timing of the reward was varied, with no effect observed when the timing was fixed. “We find that serotonin stimulation in the OFC is most effective at promoting waiting and that serotonin stimulation in the NAc does not promote waiting,” they wrote. “We also find that serotonergic stimulation in the mPFC promotes waiting only when timing of future rewards is highly uncertain.”
“The differences seen in how each area of the brain responded to serotonin suggests that each brain area contributes to the overall waiting behavior of the mice in separate ways,” Miyazaki said. The scientists constructed a computational model to help explain the waiting behavior of the mice. The model assumes that the mice have an internal model of the timing of reward delivery and keep estimating the probability that a reward will be delivered. They can therefore judge over time whether they are in a reward or non-reward trial and decide whether or not to keep waiting. “The Bayesian decision model of waiting assumes that a mouse has an internal model of the timing of reward delivery and keeps estimating the probability for the trial to be rewarded while waiting,” the team explained. “The likelihood for the trial to be rewarded declines as the mouse keeps waiting with the reward yet to come, and the posterior probability for the trial to be rewarded is estimated by multiplication with the prior probability for a rewarded trial.”
The model also assumes that the orbitofrontal cortex and the medial prefrontal cortex use different internal models of reward timing, with the latter being more sensitive to variations in timing, to calculate reward probabilities individually. “Our Bayesian decision model of waiting assumes that serotonin signals the prior probability of reward delivery and that the OFC and the mPFC use different models of reward timing to compute posterior probabilities independently,” they further stated.
The researchers found that the model best fitted the experimental data of waiting time by increasing the expected reward probability from 75% to 94% under serotonin stimulation. Put more simply, serotonin increased the animals’ belief that they were in a reward trial, and so they waited longer. Importantly, the model showed that stimulation of the dorsal raphe nucleus increased the probability from 75% to 94% in both the orbital frontal cortex and the medial prefrontal cortex, whereas stimulation of the brain areas separately only increased the probability in that particular area. “These results suggest that in the mPFC, serotonin affects evaluation of time committed, while serotonin in the OFC is responsible for overall evaluation of delayed rewards.”
“This confirmed the idea that these two brain areas are calculating the probability of a reward independently from each other, and that these independent calculations are then combined to ultimately determine how long the mice will wait,” explained Miyazaki. “This sort of complementary system allows animals to behave more flexibly to changing environments.”
The investigators aim to further investigate further how different areas of the brain are affected by serotonin, with a view to potentially developing targeted drug treatments. For example, selective serotonin reuptake inhibitors (SSRIs) are drugs that boost levels of serotonin in the brain and are used to treat depression. “This is an area we are keen to explore in the future, by using depression models of mice,” Miyazaki commented. “We may find under certain genetic or environmental conditions that some of these identified brain areas have altered functions. By pinning down these regions, this could open avenues to provide more targeted treatments that act on specific areas of the brain, rather than the whole brain.”
As the authors concluded, “Our model may be used to evaluate serotonin function in depression model mice. Depression model mice include both serotonin-selective reuptake inhibitor (SSRI)–responsive model mice and SSRI nonresponsive model mice. The Bayesian decision model of waiting may evaluate which parameters are affected by serotonergic neuron activation in the DRN and serotonin projection areas in depression model mice. These data may reveal which neural circuits are impaired in each depression model.”