April 3, 2014 by NS
I’m going to take a short break from loftier topics to vent about something that’s bothered me for many years. That is, there seems to be a widespread lack of understanding of introversion in society, which I think is caused by a bias in mainstream society to think that extraversion is good, and also caused by the lay popularity of the Myers-Briggs type indicator. Note that I think this issue is understood in the field of psychology, but this understanding hasn’t penetrated the public consciousness.
Broadly speaking, I think the issue is that there are two different reasons that people avoid social situations. One is that social situations cause them anxiety. The other is that social situations are draining. The difficulty is that the observable effects of these two things is very similar, and subjects may not even sufficiently understand themselves to make this difference clear to investigators, whether in personality tests or in interviews.
I want to start by giving some background about personality type classifications in general. We are still quite far from understanding how the brain works, so it’s basically impossible to approach the question of personality from first principles, or in other words, to come up with a model with a strong theoretical foundation and biological explanation. Personality is not a very precisely defined concept in the first place. But if personality is defined as some underlying set of traits that affect or even determine a person’s behavior in situations, it may be possible to figure out what those traits are by observing patterns in human behavior. Thus, you can ask people a bunch of different questions about how they feel or what they would do in various situations. Then you observe that certain sets of questions all tend to be answered the same way – most people who say “yes” to #10 also say “yes” to #15 and “no” to #18. So you may conjecture that those three questions all pertain to the same trait, and everyone who answers the same way to those three questions shares the trait.
The mathematical version of this concept is factor analysis. I will avoid doing any serious math, so click the link if you care about that stuff.
There is no precise way to determine the correct number of factors. There are lots of reasonable things you can try, but I won’t get into that here. Suffice it to say that there’s no standard or best practice method.
Now you’ve extracted this set of factors, so what do you do? Hopefully you can now predict a person’s behavior (beyond just how they’ll answer questions like the ones on your test). Maybe you can better identify or understand psychopathology – mental illness. The factors may be useful in a statistical sense even if they are not meaningful in the human sense, if they are good for making predictions about situations beyond the test conditions. But generally people seem to hope and expect that the factors will be meaningful traits that can be interpreted or described using the existing language of personality and behavior. Such meaningful factors are also more directly useful for interpretation and understanding.
Let’s move on to the main personality type models that are in use today. There are two: the Myers-Briggs Type Indicator (MBTI) and the Five Factor Model (FFM). There are some variations of the Five Factor Model, and a number of other models have been proposed that haven’t gotten much traction yet. There are also some personality tests that are more specialized for business purposes and attempt to determine how people work and think, in order to help managers assemble better teams, understand their employees, and match people with appropriate tasks. My intent is not to survey the field of personality testing but rather to talk about introversion, so I’m not going to try to cover everything.
MBTI, the most famous personality type model, is an attempt to realize Jung’s non-analytical theory of personality types using rigorous statistical methods. Unfortunately, it seems like Jung’s theory is not all that great by modern statistical standards – people have a tendency to get different types when they take the test twice with a gap of time in between, even if that gap is as short as a few weeks, which suggests that MBTI is not very good at capturing people’s true personalities (which one would expect to be fairly stable over the medium term). Also, the factors don’t seem to be independent. As a result, although MBTI is still very popular in the non-academic world, it is no longer in fashion in academic psychology, where models with more factors tend to dominate.
The flaw with Myers-Briggs that concerns me is its Introversion-Extraversion dimension. The definition of this scale does not acknowledge the idea that neuroticism, or a person’s tendency to worry about things or get anxious, is separate from a person’s preference between solitude and social interaction based on emotional drain. Thus, no matter why you are shy or avoid social situations, you are likely to score in the Introverted direction along the Introversion-Extraversion dimension. I think this causes a lot of people to mistake their anxiety for introversion, which potentially leads to poor self-treatment and a lack of understanding of oneself.
The Five Factor Model instead explains personality as a combination of Agreeableness, Neuroticism, Extraversion, Openness, and Conscientiousness. The dimension of Neuroticism basically separates aversion to social activity for anxiety-related reasons from aversion to social activity for effort/energy-related reasons. The only complaint I really have about the FFM is that the names of the dimensions are very far from neutral: most people would agree that being agreeable, open and conscientious is good, and being neurotic is bad. There is an implicit suggestion, therefore, that being extroverted must be either good or bad also, and is probably good. It isn’t obvious to me that there is anything inherently good or bad about introversion or extraversion. In any model, it would be preferable to choose neutral labels for the factors.
Anyway, it’s important to understand that “introversion” as it is popularly understood is not so simple as a binary shy vs outgoing distinction. There are multiple factors that can potentially explain what is going on when someone is withdrawn or aloof or avoids social contact. In fact, people who are highly neurotic and also extraverted are probably the most tortured, since those factors would drive internal conflict: simultaneous desire and fear of social interaction. Those who are truly introverted are content if they avoid social interaction, and simply prefer solitude or limited and private interactions with others. Extraverted people generally seem to be better able to identify with the former, and thus wrongly assume that people who are introverted are tortured people. Just like extraverted people, the MBTI system cannot distinguish introverts from highly neurotic extraverts. Broader knowledge of FFM type classifications would certainly benefit mutual understanding between people with different personalities.
There are of course other interesting questions to ask about personality factor models. Are five factors sufficient? There have been efforts to explain personality using many more factors. Should the factors be statistically independent? Will the statistically best factors be directly observable or testable features of people, or can they only be inferred mathematically from subtle patterns? Personally, I know I am introverted, because I very much prefer solitude, and I find social interaction, especially with strangers or in large groups, to be exhausting and draining. But I don’t understand why I have this preference. The feeling of being drained is distinctly noticeable and is sometimes quite intense, but I don’t have a clue what I am being drained of. Further study of patterns of behavior, or personality, might be able to shed light on such questions.
But in the meantime, I’d be pretty happy if the public perception of introversion moved in the direction of greater understanding of the variety and subtlety of personality that psychologists had figured out a generation ago.