The brain constantly juggles tasks: it reacts to a sharp sound, solves an equation, or drifts into daydreams. Scientists have long suspected that switches between different "scenarios" of neural network activity underlie this, but extracting them from the chaos of electrical signals was difficult. Now researchers have applied a computational model, akin to a conductor who, from a multitude of instruments, isolates three main parts: reaction to the external world, information processing, and spontaneous activity.

The model was trained on recordings of electrical brain activity from 114 people at rest. It was able to extract hidden factors that govern network dynamics and showed that neural network oscillations in the parietal region are key to cognitive flexibility. It's like finding the master volume and timbre controls in an orchestra where each musician is a separate group of neurons.

This division helps understand how the brain switches between concentration and distractibility, between perception and reflection. In the future, this could lead to new diagnostic methods for attention disorders or to interfaces that more accurately read human intentions.