Perceiving Patterns
This text is sort of an addendum to my previous post about perceiving randomness. In this one, I go a bit on a tangent and discuss things such as the evolutionary roots of our pattern-recognizing brain. I am not a scientific expert in any of these fields so I am probably bastardizing many of the valid concepts, so forgive me for this, but this is a story that makes sense to me.
The purpose of the brain is to predict the future. This statement of mine might terrify you with its boldness but bear with me for a moment. This is not your garden variety tin-foil-hat theory although my story has much to do with them.
Also, I use the word “purpose” here for the rhetoric effect only, as there is probably no particular “purpose” to the existence of any particular organ.
When I say predicting the future I don’t mean predicting who will win the lottery next month or what will be the score of next year’s Super Bowl or anything so silly and random. The function, one of many, of the central nervous system, in any creature equipped with it, is to ensure the survival of its owner by predicting the future. For a frog that future might be two seconds in advance, and survival might mean determining the size, the speed, and the direction of the movement of a large incoming object, ensuring the timely jump out of the harm’s way, Or it might be telling apart potential prey from inedible objects. A mental calculation that goes something like this. Oh, there is something small flying this way. It looks like a fly. It might be tasty. If I stick my tongue out in 3.25s it will intercept its path and I will catch it. The frog does not know anything about seconds, degree angles, miles per hour, and any such thing. Not consciously at least. But a frog’s brain does. It can do this calculation in a fraction of a second. Over a couple of million years, it has evolved to do precisely this sort of calculation. It’s a survival strategy. Of course, you don’t need a brain to survive, and I don’t have reality TV stars in mind when I say this. Plants, mushrooms, and sponges live their entire lives happy and brainless. Still, it’s a survival strategy chosen by a sort of lifeforms to which we, ourselves belong.
We, humans, are a strange bunch. We’ve pushed this strategy to the extreme. Our brains are our best weapons in our fight for survival. The motion of a swamp fly, the shape formed on a frog’s retina by light reflected from the fly’s body, is a complex pattern. Predicting its change in the next ten seconds is not a small feat. However, our human brains deal with patterns that are several orders of magnitude more complex. For us the time window of prediction might be not mere seconds but, days, months, and years up in advance. This is something that defines us as sapient creatures, distinct from other animals, that sets us apart even from our closest primate cousins.
There is a river in France. Archaeologists have uncovered the remains of two distinct Paleolithic stamens on its banks. One belonged to Neanderthals and one to Homo Sapiens Sapiens, the Cro-magnons, our direct Stone Age ancestors. They can tell them apart by the subtle difference in tools, in the way the flint blades were chiseled out of the rock, but the main difference between them was that Cro-magnons one was used only occasionally, and the Neanderthal was permanent. The two camps coexisted side by side for centuries. The river was full of salmon, a prime source of nutrients. That is why they were there. Yet salmon swam up the river only once a year, a regular yearly migration circle. And that’s the catch. They all knew that salmon will eventually come, that there is going to be fish in the river, but Cro-magnons knew their time patterns better. They knew when the fish would come back. They knew they had the time to go and pick berries and hunt rabbits in the meantime and come back to catch them at the right moment. Neanderthals had a different sense of time or none at all. They had to sit there and wait for the fish lest they would miss the moment when the salmon returns, missing on all the hunting opportunities away from the river.
Being a smart kid in high school doesn’t mean being able to answer all the questions on an ABC test. It means being able to realize that if I don’t fuck around now and sit and learn this stuff I will not get in trouble three months later when the test comes. There is a so-called marshmallow test, a long-term psychological study that shows that the people that, at age of three, were able to longer resist the urge to eat a marshmallow that was given to them, did better in life later. The size of the prediction time window is therefore a measure of intelligence.
These patterns are one or more variables that change their value over time. Mathematics calls them time series. The prediction process starts by gathering the data, observing the variables and how they change for some time, then one can hope to guess how the same variables will behave in the future, which values they will take.
Since this process is essential for our survival we are hard-wired to do it. We are hard-wired to even joy doing it. Our brains squirt out a bit of dopamine whenever we manage to do a good prediction, a small reward for doing a good job. More precisely defined the brain is a prediction error minimization organ. It constantly makes predictions about time-based patterns. Observes the results of its own predictions and rewards itself with a hit of dopamine if things go as expected.
Music is just one such pattern, complex yet regular in some way, predictable. This is why we derive pleasure from listening to it. We learn to recognize patterns in the same way machines do, by examining the patterns that we know over and over again. This is why little kids like to watch the same scene in their favorite cartoon three hundred times in a row and why some people like to watch cliche stories in movies. But the mechanism works in both directions. Our brains know when there is nothing new to learn when reexamining a pattern is not effective. This is why some people find predictable stories boring. The threshold is a personal thing. Some people learn quicker and find things that are easy to predict more boring. Maybe there is a correlation between IQ and the complexity of music and the storyline we find entertaining?
Prediction is easy if the interplay between variables that describe the system is simple. These types of systems are known as linear. Our brains are really, really good at doing this sort of prediction. This is pretty much the only type of prediction we can do. We are wired to seek out such patterns and we see them even if they don’t exist. Nature errs on the side of safety. Thus pareidolia, and the face of Jesus on a piece of toast.
But what if a system is more than a sum of its parts if its behavior is not a simple sum of values of its variables at any given moment in time. A small change in one variable can cause a disproportional change in the values of other variables in the system down the line. The proverbial butterfly effect of chaos theory. The weather, the stock market, and our own metabolism are all such nonlinear systems. This is why we are not able to predict Wall Street or the result of the next year’s Super Bowl.
The catch is that even such systems behave in a linear way in some time range or range of variable values. We can predict the weather three days in advance. This is why our whole evolutionary mechanism is useful even in the nonlinear environment that we inhabit. We can’t do these nonlinear predictions. We are very bad at them. Drinking beer from a curved glass makes you get drunk more quickly. Our brain mucks the prediction of the time it will take you to drink a pint. This is not just an imperfection of our brains. Mathematics breaks at this point. There is no easy way to handle it and we are terrified by it. We would like the world to be predictable. Small changes that result in gigantic events scare the hell out of us. This is why conspiracy theories seem so appealing. Big events need to have big causes, even if we do not see one, surely it must be a conspiracy.
