Your Brain Literally Replays Reality When You Imagine Things

The Brain's Shared Code for Seeing and Imagining

A landmark study published on the ninth of April in the journal Science has revealed that the human brain uses many of the same neurons for both seeing and imagining objects, providing the first single-neuron evidence of a shared neural code between visual perception and mental imagery.

Researchers at Cedars-Sinai Health Sciences University and the California Institute of Technology recorded the activity of individual neurons in sixteen adults with epilepsy who had electrodes temporarily implanted in their brains as part of routine seizure diagnostics. Patients were shown a series of images of faces and objects while scientists monitored neurons in the ventral temporal cortex, centred on the fusiform gyrus, a region essential for high-level visual processing.

A Distributed Axis Code

The team discovered that roughly eighty percent of visually responsive neurons encoded objects through what they call a distributed axis code, where each neuron responds proportionally to a specific dimension of an object's visual features. When participants were later asked to imagine those same images from memory, approximately forty percent of these axis-tuned neurons reactivated using the same code, and at comparable strength to the original viewing.

Remarkably, the reactivation was strong enough for researchers to reconstruct what a patient was imagining, a first for this type of study.

AI-Powered Validation

Artificial intelligence played a central role throughout the investigation. The team used generative AI to create novel synthetic images designed to maximally activate specific neurons, then verified the predicted neural responses, effectively validating the axis code they had discovered.

Therapeutic Potential

The findings carry significant clinical weight, with researchers suggesting that deeper understanding of this neural process could open pathways toward developing new therapies for post-traumatic stress disorder, obsessive-compulsive disorder, and other mental conditions involving uncontrolled vivid imagery. The research may also contribute to defences against memory loss conditions and inform the development of more efficient artificial intelligence systems.