Chinese Scientists Uncover Single-Neuron Computational Mechanism Behind Flexible Brain Decisions

China News Service reported that a research team from the Center for Excellence in Brain Science and Intelligence Technology of the Chinese Academy of Sciences has revealed the computational mechanism of brain neurons in flexible decision-making. This finding fills a long-standing gap and may inspire underlying algorithms for future artificial intelligence. To explore brain flexibility, researchers established an auditory decision experiment where mice learned hidden patterns from sounds. Once the mice mastered a pattern, scientists quietly switched the rules to track how the animals adapted to new logic. Using two-photon microscopy on auditory cortex neurons, researchers observed that apical dendrites handle inferred abstract rules while the cell body receives sensory signals. Both streams of information are compared inside a single cell to determine whether a hypothesis holds. The experiment demonstrated that an individual neuron operates as a small computer, carrying out decision-related cognitive computations through interactions between its dendrites and cell body before synapses gradually reinforce confirmed rules.

Current artificial intelligence simulates only synaptic connections between neurons. Incorporating this newly discovered single-neuron computational mechanism into future algorithms could potentially enable artificial intelligence systems to generate hypotheses rapidly and infer new rules from small sample sizes.