How does the brain evaluate rewards?

Neuroscientists Raymundo Báez-Mendoza and Fabian Grabenhorst have shown how nerve cells in the amygdala not only encode the probability and magnitude of rewards, but also dynamically process this information to predict value and risk. Their findings, which are published in the journal Nature Communications, provide new insights into the neural basis of decision-making and could also be important for understanding certain mental illnesses.

“The amygdala is an important centre for emotions and decision-making,” said Báez-Mendoza, who is Head of the Junior Research Group Social Neurobiology at the German Primate Center (DPZ) – Leibniz Institute for Primate Research. “We wanted to find out how individual neurons in the primate brain extract information about the probability and magnitude of a reward based on visual cues and how they translate these attributes into risk assessments, which ultimately form the basis for decision-making.”

Grabenhorst, an associate professor of experimental psychology at the University of Oxford, added, “Probability and risk govern our lives. Try to estimate how probable it is that something good will happen to you today; someone offering you a treat or a promotion. The amygdala cells we identified may process probability and risk to guide our emotions and social behaviours, functions we know the amygdala specialises in.”

To gain these insights, two rhesus monkeys were trained in an experimental environment that allowed the researchers to measure neural activity during a decision-making process. The monkeys learned to sit in front of a screen and look at visual cues that provided them with information about the probability and magnitude of the associated rewards, which they received in the form of fruit juice from a tube. While the monkeys viewed and processed the visual stimuli, the researchers recorded the activity of individual neurons in the amygdala using hair-thin microelectrodes.

In addition, the monkeys performed behavioural tasks in which they could choose between two reward options presented to them on the screen: a ‘safe’ option with a fixed reward probability and a ‘risky’ option with a variable probability. In this task, the animals had to decide by touching the screen whether they preferred the safe or the risky option. This allowed the researchers to confirm that the animals were able to use the information provided to maximise rewards.

The scientists observed that many neurons in the amygdala represent the probability of a reward in an abstract way that is independent of the visual stimuli. This means that the cells recognise the probability regardless of whether it is represented by images or other visual signals, for example. Some neurons specialise in pure probability, while others also flexibly consider the magnitude of the reward.

The researchers were also able to show that many neurons in the amygdala process information about the probability and magnitude of rewards one after the other. These cells first react to the probability and then to the magnitude of the reward. This dynamic processing enables the brain to flexibly evaluate different aspects of a reward and integrate them into an overall evaluation.

“Our study documents that neurons in the amygdala process not only the probability and magnitude of rewards, but also the uncertainty associated with the rewards,” Báez-Mendoza said. “This is an important step in understanding how the brain makes complex economic decisions that must take into account both the value of a reward and the associated risk.”

The study also showed that some neurons in the amygdala can assess the risk of a reward. Risk is defined as the uncertainty about the magnitude of the reward. The cells dynamically integrate information about probability and magnitude to predict risk. This ability to assess risk is important for making decisions in uncertain situations.

“Our study documents that neurons in the amygdala process not only the probability and magnitude of rewards, but also the uncertainty associated with the rewards,” Báez-Mendoza said. “This is an important step in understanding how the brain makes complex economic decisions that must take into account both the value of a reward and the associated risk.”

“These probability cells and risk cells in the amygdala may well underlie our ability to decide whether a highly valued but improbable goal is worth pursuing,” Grabenhorst added.

In the long term, the results of the study could also be important for understanding mental illnesses such as depression and anxiety. According to Báez-Mendoza, “The amygdala plays an important role in the processing of emotions and the evaluation of rewards. A malfunction in this area could lead to a distorted perception of rewards and risks, which in turn could contribute to mental health problems.”

Image caption: Computer-generated illustration of a rhesus monkey brain showing the shape and location of the amygdala (coloured green): Image credit: Igor Kagan.