February 26, 2019
Written by: Greer Prettyman
Imagine that you just got back from a fun vacation and you’re excited to post a cute pic on Instagram. You get some likes on the photo and you’re feeling pretty good. What if you then see that a friend posted a picture from a more extravagant trip and hers got way more likes than yours did? The social approval you got for your post might start to feel less rewarding in comparison. The amount of value we place on a reward isn’t stable – it can change based on external factors, including social comparison. In order to make this kind of judgement about reward value, the brain has to keep track of what others have and then use that information to make a calculation about how much to subjectively value a reward.
To figure out how the brain is making these computations, scientists turned to primates. Just like humans, primates are sensitive to social context when thinking about rewards. This was shown in a study by Sarah Brosnan and Frans de Waal1 and is demonstrated quite clearly in this cute video. In this experiment, a capuchin monkey is perfectly content to perform a simple task in exchange for a cucumber reward. That is, until she notices that the monkey in the cage next door is getting a much tastier reward (grapes!) for doing the same amount of work. In comparison, the cucumbers start to feel a lot less valuable, so much so that she begins throwing them back at the experimenter in protest. This kind of behavior indicates that primates are able to compare what others are receiving to what they are receiving and that they recognize when their own compensation isn’t fair.
How does the brain make this type of calculation about reward value in a social context? To study this process, scientists created a paradigm where primates made social comparisons in a setting where their neural activity could be recorded2.
In this experiment, two macaque monkeys sat facing each other across a display screen (Figure 1). They both learned that they might get a water reward depending on which picture they saw on the screen (they are kept thirsty, so this water is very rewarding). One monkey was considered the “self” and the other was considered the “partner”. When the self monkey got the reward, a sound was played over a speaker and when the partner got a reward, a different sound was played. This way, the monkeys could tell who was getting a reward based on the sound they heard.
To introduce social comparison, two different reward structures were used at different times. In the Varied Self Condition, the self monkey got rewards with differing probabilities: a 25%, 50%, or 75% chance of getting a water reward depending on which picture they saw on the screen during each trial. At this time, the partner always had a 20% chance of getting a reward no matter what picture they saw. In the reverse condition, the Varied Partner Condition, the roles flipped so that the self monkey always got a 20% chance of reward while the partner had varied probabilities of 25%, 50%, or 75%. Either one monkey or the other could get a reward on each trial but never both at the same time.
When a monkey is anticipating a big reward, they demonstrate licking behavior, similar to how we might lick our lips when thinking about eating a special meal. By watching how much the monkeys performed this anticipatory licking behavior, the scientists could measure how much the monkey is anticipating and valuing a reward. When the self monkey saw a picture indicating that they had a high probability of getting a reward, they showed increased licking behavior compared to when they had a smaller chance, indicative of higher reward value. Crucially, during the Varied Partner Condition, as the probability that the partner would get a reward increased, the self monkey showed decreased licking in anticipation of their own stable 20% chance at reward – basically as the partner got greater rewards, the self’s own reward started to feel less valuable, just like the cucumber in contrast to a grape.
To make sure that this effect was driven by social comparison, a non-social control condition was added. In this version of the experiment, the partner monkey was replaced by an inanimate object- a water jug that could collect the water reward (Figure 1B). In this case, the self monkey did not change their own licking behavior as the jug got a higher chance of a water reward because it was no longer a matter of social competition. The social aspect of the comparison matters, just like the way you probably wouldn’t get as jealous of an advertisement post getting a lot of likes rather than another person who you feel like you’re competing against.
Now that the experimenters had a paradigm to create this type of social comparison in the lab, they could record from neurons in the monkey’s brain to learn more about the circuits that are responsible for making this type of judgement. The brain relies on signaling of a neurotransmitter called dopamine to represent subjective value: the value you place on something, which takes into account benefits, costs, and contextual information. However, prior to this experiment it wasn’t known whether this brain circuitry also keeps track of the rewards others are receiving and incorporates that information into subjective value.
Using a technique called single unit recording, the experimenters recorded from individual dopamine neurons in the self monkey’s brain while he saw the pictures indicating the reward probabilities to the self and the partner. Some of these neurons in a part of the brain called the dopaminergic midbrain increased their activity as the self reward increased, and decreased activity when the partner’s reward increased. These neurons changed activity in the same pattern as the licking behavior, indicating that they track overall subjective value of the reward including social comparison.
Then, the scientists recorded from a different brain region called the medial prefrontal cortex (mPFC), which interacts with the dopaminergic midbrain and is also involved in encoding value. Here, they found that some neurons increased activity with the self’s reward and other neurons increased activity with the partner’s reward. This pattern of activity suggests that neurons in this region are tracking each monkey’s reward separately.
Lastly, the experimenters investigated the direction of information flow through this circuit by recording from both the mPFC neurons and the dopaminergic neurons at the same time. They found that when making judgements about a partner’s rewards, the neurons in the mPFC activated before those in the dopaminergic midbrain. This suggests that the neurons in the mPFC first encode each monkey’s reward separately and then this information gets integrated into subjective value in the dopaminergic midbrain.
The ability to incorporate information from a social context into reward valuation in this way is crucial for many of our daily experiences. It can help us to know when to negotiate a higher salary to get fair compensation relative to what coworkers are earning, or to stand up against inequality in our communities. It can also lead to feelings of envy or inadequacy, especially with social media making social comparison easier than ever. Social media use has been associated with increased risk of depression. With access to a constant stream of friend’s pictures of vacations, homes, and relationships on Instagram and Facebook, it can be hard at times not to feel like a monkey getting a cucumber in a world of grapes.
- Brosnan, S.F. & de Waal, F. B. M. (2003) Monkeys reject unequal pay. Nature; 425 (6955): 297
- Noritake, A., Ninomiya, T., Isoda, M. (2018). Social reward monitoring and valuation in the macaque brain. Nature Neuroscience 21(1452-1462).
Figure 1 created with BioRender