Precise neurobiological implementation of brain learning algorithms

Determine precisely how the learning and processing algorithms of the thalamocortical system, the cortico-basal ganglia loops, and the hippocampal system are implemented at the neurobiological level in the brain, in order to enable faithful AI implementations inspired by these mechanisms.

Background

The paper argues that human creative cognition depends on multiple interacting brain systems—most notably the thalamocortical system (including recurrent cortico–thalamic loops), cortico-basal ganglia loops that support skill learning, and hippocampal mechanisms implicated in memory and episodic future thinking. These systems are said to use distinct learning and processing algorithms that collectively underpin creative processes.

While future AI architectures might draw more heavily on these neurobiological principles, the paper emphasizes that a key barrier is the lack of precise knowledge about how these algorithms are implemented in the brain. Without this specificity, accurately translating such mechanisms into artificial systems remains difficult.