Your brain might just be a Prediction Engine

The best way to inovate is to mimic what is already working but in other fields

As humans, we tend to think of our brains as powerful, highly-evolved machines that can process an enormous amount of information and perform complex tasks. But what if we told you that your brain is nothing but a prediction engine? That's right - according to the latest research in neuroscience, our brains are essentially just making predictions about the world around us based on our past experiences.

The idea of the brain as a prediction engine is not a new one. In fact, it dates back to the work of Karl Friston, a neuroscientist who proposed the "free energy principle" back in 2006. According to Friston's theory, the brain is constantly trying to minimize the amount of "free energy" in the system - in other words, it's trying to make the best possible predictions about what will happen next in order to minimize surprises or uncertainty.

So how does this work in practice? Well, our brains are constantly taking in information from the world around us - sights, sounds, smells, tastes, and touch sensations. Based on this information, our brains create models of the world and use these models to make predictions about what will happen next. For example, if you see a red light, your brain predicts that the car in front of you will stop, so you hit the brakes. If you hear a loud noise, your brain predicts that something has fallen or broken, so you look around to see what's happened.

This process of prediction is known as "predictive coding," and it involves a complex network of neurons working together to create a predictive model of the world. As one paper notes,

the brain forms expectations about the sensory input it will receive in the future, and then updates these expectations based on new sensory information.

This process is constantly ongoing, and it's what allows us to navigate our environment, interact with others, and make decisions. Of course, our brains are not perfect prediction engines. Sometimes our predictions are wrong, and we're surprised by events that we didn't expect. In fact, some researchers argue that surprise is an essential part of learning - it's what helps us update our predictive models and improve our predictions in the future. As one paper notes,

Surprise is a crucial component of the learning process, as it indicates the need to revise and update the predictions the brain is making about the environment.

Another interesting aspect of the brain as a prediction engine is the concept of "Generative Models". A generative model is a type of statistical model that is used to create a simulated dataset that approximates the real-world data. In the context of the brain as a prediction engine, generative models are used to simulate the patterns of sensory input that the brain receives, based on the predictive models that it has built.

Interestingly, this is similar to how a language model like the one used by ChatGPT works. A language model is a type of generative model that is used to generate text that is similar to human-generated text. In the case of ChatGPT, the language model has been trained on a large dataset of human-generated text, and it uses this dataset to generate text that is similar to what a human might say.

Just like ChatGPT, the brain is constantly generating predictions based on the predictive models it has built. These predictive models are based on past experiences, and they are used to create simulations of the world that the brain expects to see. This is why we are able to make quick decisions and react to unexpected events - our brains are constantly simulating different scenarios and predicting what might happen next.

The idea that the brain operates like a generative model has been supported by a number of recent studies. For example, one study used brain imaging to show that the brain's predictive coding mechanism is active when people listen to music. The researchers found that the brain's predictive models were able to predict the next note in the melody, even before it was played.

In another study, researchers found that the brain's predictive models can even be used to simulate the movements of other people. The researchers used brain imaging to show that the brain's predictive models are able to simulate the movements of other people, based on the sensory input that the brain receives.

The idea that the brain operates like a prediction engine is a fascinating one, and it has important implications for our understanding of how the brain works. By studying the brain's predictive models, we can gain insights into how we learn, how we make decisions, and how we interact with the world around us. And who knows, perhaps one day we will be able to model the same kind of language model that ChatGPT uses in our own brains!

So, what does all this mean for us as humans? Well, it suggests that our brains are constantly working to make sense of the world around us, based on our past experiences and our current sensory input. By understanding the brain as a prediction engine, we can gain insights into how we learn, how we make decisions, and how we interact with the world around us.