Stephen Monrad

Over the weekend, I read the book “Kluge – The Haphazard Evolution of the Human Mind” by Gary Marcus. It’s premise is that our brains are the product of evolution. The parts that have been around for a long time work pretty well. The parts that are newer have more problems. Vision and motor control are old, so they are pretty reliable. Language and rational thinking, on the other hand, are new. They have many problems.

One of the main source of problems is how evolution works. It doesn’t discard the old and start from scratch. It builds upon itself over time. For most critters, contextual memory is very useful. If I’m a squirrel and I see a tree branch shake, I want to know what is likely to happen next. The squirrel’s memory is designed to retrieve information relevant to the shaking branch when it sees the shaking branch.

Contextual memory of this sort isn’t the best type of memory for basing rational decisions on. It is better to think through all the data available when making rational decisions instead of considering only what happens to come to mind. We’d be much better off with memory similar to what a computer has. That way, we could ruffle through all of it and know we haven’t missed anything.

Our memory isn’t the only problem. Our whole brain is cobbled together from bits and pieces that have done well in our ancestors. Our expectation should be that our ability to think rationally is limited because our brain is made up of very old components that were useful for other purposes than doing mathematics and logic problems.

The thrust of the book was that if we recognize that our brains are far from perfect, we can do things to try to help ourselves make more rational decisions. I take away something else. The reward systems in our brain are located in the old portions that haven’t changed much over time. What makes sense may not feel particularly good. Our use of logic to make decisions needs to make concessions to the reality that our happiness depends on needs that are primitive.