I have never been comfortable with proclamations by educators or scientists (and yes, there is a difference) about how the brain works. The logical fallacy goes something like this: “we have isolated a mechanism in the brain, learning takes place in the brain; therefore, we now know how learning works.” Whenever a psychologist says something smug like “the brain doesn’t work that way” (around 1:21), I want to pull my hair out. The latest theories about how the brain supposedly works also include huge gaps in our understanding of how the brain supposedly works and plenty of lines of research that may one day soon give us a more complete picture of how the brain supposedly works. The idea is that if we know how the brain is supposed to work, then we will somehow know how we learn. There are so many layers here though that it seems to be an impossible task. First, it assumes a purely mechanistic view of the mind and learning. Not that we have to get metaphysical, but this could be something that is so complicated that thinking of the mind as a flow chart or a network may not even scratch the surface of what is really happening. When educators talk about what neuroscience has to say about learning, we have to remember that neuroscientists aren’t even sure what neuroscience has to say about neuroscience. It is a difficult field because each year brings in a new raft of technologies that reveals more and more about the physical properties, chemical reactions, and neural connections in the brain. But I think there is some promising work in neuroscience that we should be keeping an eye on as educators. One of the more interesting lines of research includes the mathematical models around “deep learning.” I think this is finally getting at the complexity necessary to account for the complexity of thinking, language, and learning.
I think there are some promising avenues of discovery in the work of Gary Marcus that could one day help address how we learn. Gary Marcus describes deep learning this way: “Instead of linear logic, deep learning is based on theories of how the human brain works. The program is made of tangled layers of interconnected nodes. It learns by rearranging connections between nodes after each new experience.” In other words, the brain is not seen as a series of connected flowcharts but as intersecting nets of connections that create patterns.
Additionally, Geoffrey Hinton describes the brain as a holograph. Daniela Hernandez writes about Hinton in Wired saying that “Hinton was fascinated by the idea that the brain stores memories in much the same way. Rather than keeping them in a single location, it spreads them across its enormous network of neurons.”What I like about Hinton is that he says that his work involves creating computer models of intelligence and he seems to avoid the heavy handed proclamations of discovering how learning works. His work discusses “machine learning” which is an entirely different concept. I think it is very important to remember that we are talking about models and not “how the brain works.” The networks involved in learning are even more complex than his model because our layers include language, behavior, culture, society, etc. Never mind the chemical and quantum connections in the brain. It is just possible that one day Hinton’s work can speak to the complexity of the interplay of all of those networks and their seemingly infinite interrelations.
How does this shape my practice as an educator? I teach workshops on concept mapping and have used concept mapping in my classes, not because I feel that they somehow mimic the way the brain learns but because it is an engaging learning and teaching method that provides opportunities to utilize visual and kinesthetic learning modalities as well as using critical analysis. In other words, it is a method of teaching and learning that engages multiple ways of knowing. And it may also be a good metaphor for how learning may occour in networks, including neural networks. I have seen this discussion around the learning theory, Connectivism. I think we could go into any learning theory and use it, somewhat clumsily, as a way to discuss how learning arises out of the formation and interplay of network, but fortunately George Seimens and Stephen Downes have done a better job with their work around Connectivism.