While humans still can't learn new information as quickly as artificial intelligence, a biological brain is pretty much capable of that.
A new study claims that exploratory actions can help with that a lot.
Let's find out more.
Researchers at the Sainsbury Wellcome Centre have made a breakthrough in understanding how various animals learn their spatial environment through exploratory actions.
By studying mice, they discovered that purposeful actions, such as darting towards objects, play a crucial role in efficiently learning a cognitive map of the world.
The research, published in Neuron, challenged the notion that simply observing obstacles was enough for learning.
Using various optogenetic tools to prevent mice from initiating exploratory runs, the scientists found that the mice did not actually learn, even after spending time observing the obstacles.
These instinctive exploratory actions proved essential for building a cognitive map.
To further explore the learning algorithms used by the brain, the team ran different reinforcement learning methods, finding that mice displayed both model-free and model-based behaviors.
This research has implications for developing more efficient AI agents and understanding how animals learn with less experience compared to artificial agents.
The researchers plan to investigate the relationship between exploratory actions, subgoals, and the brain's representations in future studies.