How is it that we can use our brains, which evolved over hundreds of millennia, to manipulate abstract concepts like gravity, refraction, and inertia that we’ve only defined in the last few hundred years? It turns out that we use structures we developed long ago, networks we already use to process much more basic things like rhythm and language. Scientists from CMU report that the brain handles scientific abstractions “using inherent, re-purposed brain systems,” according to Robert Mason, coauthor of their latest study.
The researchers recruited nine advanced students of physics and engineering, put them into an fMRI at CMU’s Scientific Imaging and Brain Research (SIBR) Center, and asked them to contemplate high-level concepts like diffraction, displacement, radio waves, and centripetal force. Then they fed those brain scans into a machine learning model, and found that the brain lit up in certain places in response to certain physics concepts, regardless of which person was doing the contemplating.
Then they asked the model to identify what concept each of the students was thinking about, based only on those scans. It answered correctly, with ease. When trained on data from the other participants, it even knew exactly what an excluded participant was thinking of.
The model also figured out that under the hood, the brain is able to learn physics because it can understand the four fundamental concepts of “causal motion, periodicity, energy flow, and algebraic (sentence-like) representations.” We handle abstract concepts by adding complexity to neural firing patterns and “repurposing” the same structures we use to handle basic, intuitive concepts. Changing the software is cheap and easy compared with swapping out new hardware, here.
Brain systems that process rhythmic periodicity when hearing a horse gallop also support the understanding of wave concepts in physics. Similarly, understanding energy flow uses the same system as sensing warmth from a fire or the sun. Grasping how one concept relates to others in an equation uses the same brain systems that are used to comprehend sentences describing quantities.
Breakthroughs like this could change how we teach physics. Knowing what the brain is doing when it’s learning physics concepts is a powerful tool. It lets instructors tailor their curricula to what they’re trying to accomplish.
“If science teachers know how the brain is going to encode a new science concept, then they can define and elaborate that concept in ways that match the encoding,” said Mason. “They can teach to the brain by using the brain’s language.”