Dopamine research is more interesting to watch.

Dopamine research is more interesting to watch. Dopamine is more influenced by the degree of prediction than the actual reward. What happens here is "Reward Prediction Error (RPE)," which has a huge impact on the satisfaction we feel.

There is a saying that happiness comes from the process, not from the result. As such, dopamine is not involved in the happiness we feel in rewards, but in the pursuit of those rewards.
____________________________________
Chicken, when delivery starts vs. when eating...The secret of "Dopamine" [Deep THINKING]
[Ed. note] The era of artificial intelligence, represented by ChatGPT, is coming. Now, the depth of thinking is more important than knowing a lot. Let's think about deep thinking in the AI era.

Dopamine is the talk of the town. Dopamine, a neurotransmitter, is a substance that makes people feel happy and happy and has a huge impact on our compensation system.

Dopamine, however, is more influenced by the degree of prediction than the actual compensation. What happens here is the "Reward Prediction Error (RPE). Dopamine is actively produced when unexpected or larger-than-expected compensation is received, but if there is no expected or less compensation than expected, dopamine production is lower than usual. As such, the working system of dopamine is difficult.

1. Wolfram Schultz's Monkey Experiment and Compensation Prediction Error

The world's most renowned dopamine expert is Wolfram Schultz, a professor of neuroscience at Cambridge University in the U.K. In the 1990s, Schultz discovered that if the monkey was rewarded with juice by inserting electrodes into the brain of macaque monkeys, the activity of dopamine neurons in the basal nucleus would increase.

In humans, the midbrain in the middle of the brain has about 1 million dopamine neurons, and monkeys have about 200,000 dopamine neurons.

Professor Schultz discovered new facts about the response of dopamine neurons through monkey experiments. In other words, when the reward is predicted, the response to the reward itself disappears (when it is predicted and the reward is also generated in the figure). However, when more rewards are generated than predicted, dopamine neurons show a stronger response (① when the reward is generated without prediction).

Conversely, when there was no reward or less than expected, the activity of dopamine neurons decreased (when it was predicted but no reward).

Through such experiments, Schultz found that the dopamine response represents a "compensation prediction error," which can be explained by the "difference between the obtained and expected compensation." When receiving more compensation than expected (①), a positive dopamine reaction (excitement, activation) was induced, but there was no response to the expected compensation (②). In addition, a smaller-than-expected compensation (③) led to a negative reaction (a contraction in activity).

The same phenomenon was found not only in monkeys but also in humans and rodents. In other words, dopamine neurons do not just respond to existing rewards, but only to "unpredictable rewards."

These mechanisms are clearly created by evolution, Professor Schultz said, explaining why we always want more rewards and are not satisfied with what we have now.

Dopamine activation occurs when we receive more rewards than predicted, and the problem is that the rewards we receive automatically update the forecasts. In other words, the new standard is the previously received, more rewards than predictions, so the next time you receive the same reward, fewer prediction errors occur or no more.

After all, in order to obtain the same prediction error (dopamine stimulation), we need to receive more compensation every time. Professor Schultz called the dopamine system a "little devil." Professor Schultz explained that the reason why we want to buy a new car is not just because our neighbors bought a new car, but because we are used to the current car.

Another interesting phenomenon is that dopamine is proportional to the relative value, not the absolute value.

Professor Wolfram Schultz also demonstrated this fact well in an experiment conducted by Professor Wolfram Schultz. Professor Schultz trained monkeys to expect two or 20 units of compensation depending on the situation. At this time, if each situation unexpectedly received 4 or 40 units of compensation, the monkeys' dopamine secretion soared to the same extent. On the other hand, if 1 or 10 units of compensation were given, dopamine secretion decreased to the same extent. The compensation for the two situations differs by a factor of 10, but the important thing was the relative size, not the absolute size of the surprise.

2. 3 Graphs for Dopamine

Professor Robert Sapolsky of Stanford University, a master of neurophysiology, explained dopamine using three graphs in his book "Behave."

Let's take a look at the monkey experiment. When the monkey's room lights up, it signals that the reward test begins. The monkey presses the lever 10 times and receives the raisin. The important thing here is that when the light is turned on to announce the start of the experiment (when it receives the signal) before the monkey enjoys the lever, it releases a lot of dopamine, as shown in graph ①.

In other words, if the fact that rewards are involved has been learned, dopamine is involved in the expectation of rewards rather than in rewards. As soon as you saw the signal in graph ①, dopamine secretion began to increase, and when you were actually receiving rewards after work, dopamine secretion was less than when you saw the signal. In other words, pleasure comes from the expectation of a reward, and the reward itself is almost a reflection.

Professor Sapolsky said that pleasure comes from appetite rather than satiety once we know that appetite will be satisfied. This seems to be the same reason that I become happy just thinking about ordering chicken on Friday evening.

Graph ② is also unique. Originally, the rule of the monkey experiment was to press the lever when the light turned on, and then the reward came out. But I changed the situation. When the light turns on and the lever is pressed, there is a reward, but it comes out with a 50% chance.

The funny thing is that monkeys who learned new situations had more dopamine released than before. Maybe it'll come out