By GREGORY ZELLER //
Straight from the fusiform gyrus (bridging the bases of the temporal and occipital lobes) comes the human brain’s innate ability to recognize faces, a fairly mysterious physiochemical action – until now.
Science understands how the brain’s fusiform facial area responds to faces, or at least that it does. But the limitations of experimentation within functioning brains – that old bugaboo – have left the more precise goings-on, at the individual-neuron level, largely undiscovered.
Enter scientists from the Feinstein Institutes for Medical Research and Israel’s Weizmann Institute of Science, who dig deeper than ever before into the science of facial recognition in a new paper published this month in the peer-reviewed Journal of Neuroscience.
Welcome to the ventral temporal cortex, home of the brain’s high-level visual processing – critical to recognizing things like faces and familiar landmarks. Working with adult epilepsy patients at Manhasset’s North Shore University Hospital, researchers have gained unprecedented access to the cortex and measured electrical activity among its individual neurons.
Ashesh Mehta: Be specific.
To explain the science, lead researcher Ashesh Mehta – director of the Feinstein Institutes’ Northwell Health Laboratory of Human Brain Mapping – references the “Jennifer Aniston cell,” an oft-debated scientific precept that suggests individual concepts (the “Friends” star, certain sounds, your mom’s face) each occupy their own single neuron in the brain.
“Essentially, different brain cells activate to different things,” Mehta said. “So, show them something familiar, like Jennifer Aniston’s face – but not other faces – and sure enough, that same cell activates.”
In “Face-Selective Units in Human Ventral Temporal Cortex Reactivate During Free Recall,” Mehta et al take the concept a step deeper, revealing that the same neurons that activate when people see a face activate when people visualize a face – as in, close their eyes and picture it.
That’s an important discovery, according to Mehta: Neurologically, there’s no difference between actually engaging the sense of sight and mentally picturing an image.
“When you close your eyes and see something,” he noted, “yes, you’re actually seeing it.”
This level of sensitivity also allows researchers – with the right electrodes in place, and an established baseline linking particular electrodes to individual concepts – to literally know what someone is thinking.
Face it: Do you know me?
“We can ask someone to think of something like a face,” Mehta said. “And before they start describing it, we can … determine which neurons are firing and know which face they are envisioning.”
These breakthroughs were made possible by access to NSUH’s epilepsy patients, which Mehta framed as key to the study. Studies of similar but smaller monkey brains – “They just don’t have the hardware,” according to the scientist – and views of living brains afforded by conventional technologies like MRI exams “just don’t give you the resolution, the scale, the granularity of what you can get from recording individual brain cells.”
For this study, the Feinstein Institutes/Weizmann Institute researchers (the institutes collaborate frequently) worked with eight volunteers suffering a form of epilepsy that initially causes seizures in only one area of the brain, leaving most brain functions intact.
More importantly, this particular population offers another unique advantage critical to monitoring developments inside the ventral temporal cortex.
“The best way to map activity is to place electrodes directly in the brain, and these are the only people with whom we can do that,” Mehta noted. “It’s a risk you wouldn’t want to take unless you were doing some kind of therapeutic endeavor.”
Being able to explore mostly normally functioning brains “to a very specific level” is a big step toward future treatments for epilepsy and other brain-based conditions, according to the neurosurgeon, who ranked the hyper-accurate preservation of healthy brain tissue as one goal.
Another: new “brain machine” innovations.
“Can we use this information … to improve our ability to think? To decode? To communicate?” Mehta said. “We undoubtedly will, if we continue down this path of scientific inquiry.
“These are the little steps you have to take in order to realize some kind of more practical technology.”
