Reductionism, Complexity, and Personality Types

Posted on September 6, 2010


There is a branch of psychology that classifies people into “types” of personality. There are a number of variations, but a common scheme is this:
Extroversion or Introversion
Sensing or Intuition
Thinking or Feeling
Judging or Perceiving
People are in one of each of these categories leading to 16 personality types. This is classical reductionism: trying to abstract a simple model out of the complexity of reality.
I find reductionism in psychology to be deeply misguided. I’ve always found it strange when reading how everyone was shocked and amazed when it turned out that you couldn’t in fact reduce the complexity of the human mind to, say, an IQ number.
More importantly though is to realize that putting people into these categories is largely useless, and the reason is Complexity.
Say the idea is to get an idea of how people will act in a situation. By having their “personality type” you can take a guess right? Wrong! Because people act based on what’s going on around them: that means other people. Who in turn act based on their surroundings. Its a dynamic feedback loop – a complex system.
Complex systems are formed of many agents interacting over time. Crucially each agent could have even the simplest set of rules – nothing as elaborate as even the anaemic “personality types” – and still the system can spin off into pure chaos with only a small change in a variable.
So to create such a simplified model of personality types without studying the dynamics of their complex interaction is a waste of time, unless all you want to know is how people behave in labs. Without studying this you can’t tell whether a pattern of behavior is stable or sensitive to tiny variation. And in the best of cases you still can’t really tell how a group of people are going to act.
The desire to reduce organic systems to these crude representations seems to stem mostly from a desire to emulate the “hard” sciences such as physics, where all the universe is condensed down to a few basic laws. Which is silly since physical laws can directly predict only the most simplified of situations (though a big enough computer can take you quite far), and also deal with things which can be described mathematically and measured numerically.