This is a follow-up to Why rational arguments rarely change minds, which I started when it became clear that some points in that musing needed clarification.
I use the word model quite a bit in some musings, but realized I have never quoted a definition. There are several meanings, but I will quote the most relevant meaning from the Oxford American Dictionary here: “a simplified description, esp. a mathematical one, of a system or process, to assist calculations and predictions”. This is the meaning I had in mind when I wrote that Thinking, Fast and Slow is a very useful model. As Kahneman says in his book, System 1 and System 2 are made up entities that provide a useful way of thinking about thinking, not actual pieces of human brains.
A 1973 science fiction novel I've read a number of times refers to a brain model of a man on a dog on a lizard. This is called the Triune brain model. This model is now considered oversimplified and obsolete, but it was fairly widely popularized decades ago, and I suspect it is still around in popular culture in terms such as “the lizard brain”. A lot of scientific ideas and models that get replaced or refined by later research can linger in the culture at large if they were popularized before they were devalued by later science. Other examples from obsolete models include the claim that we only use 10% of our brains, that logical thinking always happens in the left side of the brain and artistic/creative thinking always happens in the right side of the brain, and that a huge percentage of our DNA is silent/unused/junk.
The popular use of the System 1 / System 2 model since Kahneman published it shows signs of misunderstanding and misuse. The two systems are often discussed as if they describe discrete/independent brain structures, which was not in the original model. I would like to suggest a reformulation of the model based on my understanding of actual brain structures.
While searching for current numbers about parts of the brain, I found The human brain in numbers: a linearly scaled-up primate brain, which I recommend for anyone who likes to dig a little deeper. The cerebral cortex is the outermost and newest part of the brains of mammals. In humans, the cerebral cortex contains about 82% of totals brain mass, but only about 19% total brain neurons, with about 16 billion neurons.
The model I am suggesting here is a layman's suggestion with no testing or experimental evidence (that I have read or heard about). I just think it is a way to make the System 1 / System 2 model slightly less prone to misuse. Note that the history of models of brains/minds/thinking is littered with whatever the most complex technology was at the time of any given writing, from gears to computers. Since my understanding of current brain science is that most brain activity involves large networks of cells, I will use a feature of the Internet for the model. I think System 1 and System 2 are labels for parts of a single network process for conscious thought. The System 1 part is older and faster and before conscious awareness. The System 2 part in the cerebral cortex is the final step of the network process and is the point of conscious awareness. I suspect that the System 2 activity happens in some subset of the neurons in the cerebral cortex, so it will involve some fraction of 19% of the neurons in a human brain. The key issue is that there are fewer neurons in the part of the network where System 2 happens, so they can run out of energy faster than the part of the network where System 1 happens. Beyond the simple fraction of total neurons, if the brain regions that support System 2 activity are very limited, the same neurons could be involved in multiple System 1 + System 2 networks, causing bottlenecks if multiple brain networks with shared neurons are activated at the same time. The Internet analogy comes in when we think about what happens when System 2 cells run out of energy or are tied up due to bottlenecks. The architecture of the Internet includes ways to work around blockages or failures. I suggest that brain networks also work around failures. The full network is always System 1 + System 2, but when the System 2 part is fatigued or partially unavailable, it looks like what the original model describes as System 1 only. So it is not two independent brain systems, it is one brain system with a safe failure mode, producing output even when the neurons supporting the System 2 part don't contribute anything. This tweaked model is consistent with what I've read about the original model and research based on it. I have no idea what percentage of conscious thinking is purely conscious, but I have a strong suspicion that research will eventually find that most conscious thinking is actually solidly built on a subconscious base in mixed System 1 + System 2 networks.
Thinking about mixed conscious and unconscious processes brought to mind a topic we discussed a few times in CCRS meetings over the years: research that questioned whether conscious decisions actually happened before the brain processes involved started. I found a good article on this: Timing and awareness of movement decisions: does consciousness really come too late?. The model I suggest above of a brain network with a mix of unconscious and conscious activity seems to fit what that research is investigating.
Note that the brain has plenty of other networks beyond the System 1 + System 2 conscious thought network. There is a lot of brain activity involved in keeping our bodies alive, with regulation of basic survival functions like breathing (though there is also some conscious control for breathing). There are reflexes like blinking to protect our eyes from approaching objects. Etc.
I also realized that my discussion of Second Level Awareness and System 2 merged two issues. This is where we are able to work to improve ourselves and our lives, but it includes (at least) two different kinds of work. The first kind is learning to relax, to clear the mind, to reduce the toll of distractions, etc.—all the ways we might keep enough energy available for the System 2 part of the network to be able to think consciously! Once we have the energy, we can devote System 2 time to choosing areas of our life we want to improve and then learning and using whatever techniques help us achieve our goals. If we only do the first part, we remain reactive rather than proactive, but our reactions include a lot more conscious thought than those of someone without any available energy for System 2.
A personal toolkit example: I established a habit years ago of using a page-a-day calendar that provides some kind of mental stimulation beyond just a picture. For the last several years, it has usually been a brain teaser or puzzle of the day, with a number of different types of brain teasers. Since I try to pass most of the pages on to my sister, whenever possible I do the puzzles in my head rather than writing on the calendar pages. I can't prove it, but I suspect that years of that practice have slightly improved my short-term working memory. We should always evaluate any tools we add to our personal toolkits to see if they actually work for us and are worth keeping and continuing to use.
A note about brain exercises that supposedly help to improve our brains and minds: based on what I've read, very few things actually produce global improvements. Most brain training or exercises, if they help at all, help develop more specific abilities, the nature of which reflect the nature of the exercise. This kind of practice is a more positive expression of the old “use it or lose it” maxim: use it to improve it!