The brain’s center of memory and navigation, once considered too disorganized to decode, may soon be unlocked. Using a brain scanner, researchers were able to determine the location of people standing in a virtual room from the activity in their brains.
“We could read their spatial memories, so to speak,” said study co-author Eleanor Maguire, a University College, London, cognitive neuroscientist. “There must be a structure to how this is coded in the neurons. Otherwise we couldn’t have predicted this.”
Maguire’s team focused on the hippocampus, a region of the forebrain responsible for processing spatial relationships and short-term memories. As people move, hippocampal activation helps them know where they are. In Alzheimer’s patients, disorientation and memory loss go hand in hand.
But animal studies haven’t been able to link specific hippocampal activities with memories, and rat studies suggested that spatial memories were actually stored randomly. There seemed to be no pattern, at least not a pattern that scientists could decipher and apply.
Maguire’s study, published Thursday in Current Biology, challenges that notion. And though it’s far too soon to pull memories directly from a brain, the findings suggest future avenues of research on Alzheimer’s and other forms of dementia.
“How these millions of hippocampal neurons work is a fundamental question in neuroscience,” said Maguire. “We still don’t know how the hippocampal neural code is organized to support memory and activation.”
The researchers used an fMRI machine to measure hippocampal blood flow in four subjects who navigated a room in virtual reality. They focused on groups of neurons identified by Maguire in an earlier study of London taxi drivers, whose hippocampi were hyperdeveloped by years of mental navigation through the city’s mazelike streets.
After analyzing activation patterns and correlating them with a record of test subjects’ movements, Maguire’s team found that patterns could actually be used to predict location.
The results “are an intriguing first step toward using fMRI to read out information about visuo-spatial scenes,” said Arne Ekstrom, a University of California at Los Angeles cognitive neuroscientist who was not involved in the study.
Ekstrom cautioned that the findings, relying on a bird’s-eye fMRI view of just one part of the hippocampus, don’t explain what’s happening in individual neurons or across the entire structure.
Further studies will incorporate more test subjects than the four men included in this study, and involve other types of memory than spatial.
Though the findings fit with earlier demonstrations of visual memory’s reconstruction from visual cortex activation patterns, study co-author Demis Hassabis, a London-based artificial intelligence researcher, cautioned that full-blown mind reading is still decades away.
More relevant, said Hassabis and Maguire, are potential insights into how memory deteriorates.
“We’re learning more and more about how memory is laid down,” said Maguire. “We can begin to understand how pathological processes erode memories, and think about how we might help patients in a rehabilitation context, to make the most of what memories they have left.”