Learning new rules is a part of life. The rules may be simple, like learning to stop at a red light. Or they might be a bit more complex. Sometimes, for instance, you can turn right at a red light, unless there’s a sign saying you can’t, but only in certain states, and not when a pedestrian is crossing the street. Most human brains learn complex rules like these, and their myriad exceptions, with seeming ease. But how they do this has puzzled neuroscientists for decades.
“So much of human behavior is guided by rules,” says Michael Hasselmo, a Boston University College of Arts & Sciences (CAS) professor of psychological and brain sciences and director of BU’s Center for Systems Neuroscience. “But how do brain circuits mediate learning rules of different types? To me, this is a central question of brain research.”
Hasselmo is the principal investigator on a five-year, $7.5 million Office of Naval Research (ONR) grant to investigate, broadly, the question of exactly how human brains learn rules, and how this might be translated into computer programs, especially for autonomous systems. “If we understand this, it could massively enhance the capability of computers,” says Hasselmo.
The grant was awarded as part of the ONR’s Multidisciplinary University Research Initiative program, or MURI, which supports team research involving more than one traditional scientific discipline, according to the Department of Defense (DOD). Most of the program’s efforts involve researchers from multiple academic institutions and academic departments. Hasselmo will oversee a team that includes BU’s Marc Howard and Chantal Stern, both CAS professors of psychological and brain sciences, as well as researchers from Brown University and the Massachusetts Institute of Technology. The award provides $4.5 million over the first three years, with the option to renew for an additional two years.
“One of the most exciting things about the MURI awards is the opportunity to work with exceptionally talented researchers across different universities,” writes Stern in an email. “For the current MURI, I am excited to be collaborating with MIT researcher Earl Miller and Brown University researcher David Badre.”
The research will focus on two areas of the brain associated with learning: the prefontal cortex, known to be critical for working memory, and for “gating” memories in and out of the rest of the cerebral cortex; and the basal ganglia, which most people associate with movement and motor function. The scientists will investigate how the basal ganglia performs the “mental action” of loading working memories into the correct part of the brain, says Hasselmo.
Hasselmo will oversee the development of computational models of neural circuits used in learning, while Badre and Stern will use functional magnetic resonance imaging (fMRI) to study the brain in humans as they learn. Badre will examine how the brain learns tasks using rules involving matching of letters and symbols, while Stern will study how people learn new rules and changes in rules by trial and error. Stern will use BU’s newly acquired MRI scanner, the centerpiece of a new Center for Cognitive Neuroimaging, to be housed in BU’s Center for Integrated Life Sciences & Engineering (CILSE), scheduled to open in spring 2017.
Stern notes that the NSF award for the MRI system arrived the same week as the MURI award notice, “which was a great start to the academic year,” she says.
For Hasselmo, the MURI award, coupled with the new neuroimaging facilities, will help open new windows onto the deepest mysteries of the brain. “To me, it’s just fascinating to try to figure this out,” says Hasselmo. “Neurons are single cells with no individual cognitive ability, yet their interactions underlie every single action we take, every single belief we have. It’s fascinating to investigate how it comes together.”