Miraculously, our brain is able to recall images of things we had seen in the past in a way that still feels real. A study led by Cedars-Sinai Health Sciences University, published in the journal Science on April 9, found some insights that gets us closer to understanding this unique behavior of the brain.
In the study, 16 adults with epilepsy, a brain condition causing repetitive seizures, had electrodes temporarily implanted in their brains to diagnose their seizures. They were asked to view a series of images of faces and objects. Then, a subset of these participants was asked to imagine the same images from memory. During this process, researchers recorded the electrical activity of hundreds of individual neurons in each participant’s brain. They mainly focused on brain activity in the ventral temporal cortex (VTC), an area of the brain involved in identifying and categorizing visual stimuli.
When the participants were viewing the images, neurons, or brain cells, specifically those situated in the fusiform gyrus, an area of the brain crucial for high-level visual processing, especially for faces, were activated. For 80% of the visually responsive neurons recorded in the study, the researchers determined which features of the images triggered their responses, revealing their neural code, or neural patterns.
Later, when participants were imagining the images, roughly 40% of those neurons reactivated using the same code, recreating the pattern of activity that occurred when they initially viewed the images.
“We generate a mental image of an object that we have seen before by reactivating the brain cells we used to see it in the first place,” said Ueli Rutishauser, PhD, director of the Center for Neural Science and Medicine and professor of Neurosurgery, Neurology and Biomedical Sciences at Cedars-Sinai Health Sciences University, and the study’s joint senior author. “Our study revealed the code that we use to re-create the images.”
Mental imagery is our brains’ ability to create percepts, emotions, and thoughts in the absence of external stimuli. It allows us to make art, mentally rehearse actions and outcomes, recall previous experiences, and imagine new ones. However, uncontrolled mental imagery can lead to mental health conditions including anxiety, schizophrenia, and post-traumatic stress disorder. Fortunately, this study may lead to further investigation to provide relief from them.
“Further insight into this neural process has the potential to open pathways toward developing new therapies for post-traumatic stress disorder, obsessive-compulsive disorder, and other mental conditions that involve uncontrolled vivid imagery,” said Adam Mamelak, MD, director of the Functional Neurosurgery Program and professor of Neurosurgery at Cedars-Sinai, and co-author of the study.
Artificial intelligence (AI) played an important role in this study, as well.
“Advanced artificial intelligence tools were critical to our investigation at all stages,” said Varun Wadia, PhD, a postdoctoral scientist in Rutishauser’s laboratory and first author of the study. “We used deep visual neural networks to create numerical descriptions of objects so that we could understand the neurons’ code. We then verified the code by using generative AI to create never-before-seen images and correctly predict the brain’s responses to these images.”
The research builds on the work of Doris Y. Tsao, PhD, of the University of California, Berkeley, who is co-senior author on the study. She identified the neural code for object recognition in nonhuman primates. The current study reveals that the same neural code is present in humans and that it explains visual imagination.
“These findings support the idea that imagining and seeing share a common neural code and may have important implications for understanding psychiatric disorders marked by disruptions in mental imagery and reality discrimination,” said Hermon Gebrehiwet, DrPH, program officer at the National Institutes of Health.
Avenues for future research include what triggers the neural reactivation the investigators found, and how memories lead to reactivation of just the right subset of neurons needed.
