The 1972 Nobel Prize winner neuroscientist Gerald Edelman asserted that a complex brain that tells alot of information about the world must be highly differentiated and yet highly integrated. This statement may seem intuitive and convoluted but it’s really interesting to understand this idea using Information Theory.

The brain works the same way. Firstly, Edelman’s states that the brain must be highly differentiated. Each part of the brain is so attuned to responding and processing only certain sets of signals. This specialisation refers to us having distinct brain centers for vision, olfaction and motion. This differentiation is pertinent because if every part of the brain did the same thing, then it’s wasting a lot of resources not telling us as much as it can.

However, beyond this differentiation and specialisation of cells, our brain needs to be integrated. The information each part of the brain knows needs to be shared with other parts of the brain, via re-entrant signals and rich recurrences.

Edelman worked on this conclusion using Information Theory (more specifically the measure of Mutual Information). It’s amazing how neuroscientists are looking to theoretical fields to structure research.

In the same way, computer scientists borrow ideas from cognitive science to model the brain. Neural networks, backwards propagation, reinforcement learning are really concepts and theories originally describing how brains work. They are now the buzzwords of machine learning and artificial intelligence engineers.

It’s heartening how approaching problems from multiple different angles (Cognitive Psychology, Neurobiology, Artificial Intelligence) can mutually reinforce ideas. This inter-discliplinary nature of learning is increasingly important as our world has favoured very specialised knowledge.

In the Lyceum, it was perfectly normal for one to be a geologist, astronomer and mathematician all at once. In the Modern World, you can’t really say you’re a geologist at a conference without specifying the type or region of rocks your research is focussed on. (Maybe you can actually do that.)