The plots inside dreams are often disparate pieces of life pieced together to produce bizarre effects, such as dreaming that you are buying milk at the supermarket and when you walk up to it, the bottle turns into a talking fish!
Humans and many animals dream, and scientists have come up with many theories to try to explain where dreams come from. Recently, Erik Hoel, a neuroscientist at Tufts University, has proposed a new theory that dreams have the role of stimulating the brain to learn to cope with new situations and achieve self-improvement.
Hoel has been inspired in recent years by the “learning” patterns of artificial intelligence neural networks. He says a common problem that AI network systems encounter during training is that at a certain point, the system feels so familiar with the input data that it thinks it can handle any situation, and therefore can no longer improve its skills. Scientists refer to this as overfitting.
When this happens, researchers generally need to introduce some unstructured data, such as noise or invalid input data, to stimulate the system to continue “learning.
Hall believes that the human brain will experience similar problems. Especially as we get older, the brain is used to the routine of daily life repetition, which is like the brain can be used to continue to “train” the machine’s information becomes very limited. However, the brain also has to be able to cope with unexpected situations, including physical reactions, thinking reactions or comprehension of new things.
In daily life, people’s brains have to focus on what they want to do, so they can not find the so-called “noise” information into the brain. But the situation is different when you sleep.
Hall believes that sleep is like the brain itself in the exercise, the simple situations in daily life complex, or practice from a more perspective to understand the problems encountered when awake, etc..
Hall presents some evidence to support his theory. For example, if people work very hard during the day to carry out a training, such as training acrobatics, on the simulator repeated training skiing skills, this period of time at night are likely to dream of still doing these training. The theory is that this is the result of hard training to the brain caused by “overfitting” situation. So much so that at night, while sleeping, the brain will experience some training situations that allow the brain to further enhance the skills learned during the day.
People may have similar experiences, after a hard high-density training, a good sleep, often the next day skills will be more proficient. This theory would explain the phenomenon well.
Hall also gave the example that even dreaming about activities that are physically impossible during the day may also have its benefits. For example, dreaming of flying may enhance the body’s balance and stability when actually running.
Dreams are very complex phenomena and there are many types, and Hall’s theory is trying to explain one of them. There are many different theories about what role dreams actually play. Although there is not enough evidence to confirm these theories, Hall believes that dreams must serve some purpose, saying, “There is no evidence so far that dreaming is just a useless byproduct of neural activity.”
This study was published September 24, 2020 on the preprint network arXiv.
Recent Comments