5 ms·

That's a good question, and your reaction to the Myers-Briggs is justified! It is not evidence-based and is roundly rejected by modern personality psychology. T

That's a good question, and your reaction to the Myers-Briggs is justified! It is not evidence-based and is roundly rejected by modern personality psychology. The work the author of this piece was describing on the other-hand is one of the most well-researched phenomena in the history of psychology. It is based on a massive amount of research looking at clusters of personality adjectives found in dozens of languages that tend to cluster along five dimensions. And the same five clusters are found in virtually all languages (sometimes to include a 6th cluster often called "honesty/humility"). And to your point about types, this system doesn't assign types to people, but instead places them on a continuous dimension that is considered to be much more useful. And to your point about complexity, the 5 dimensions are higher level structures. Within each dimension is a branching tree of complexity that is where the current work lies. Further reading: https://en.wikipedia.org/wiki/Lexical_hypothesis https://en.wikipedia.org/wiki/Lexical_hypothesis https://en.wikipedia.org/wiki/Big_Five_personality_traits https://en.wikipedia.org/wiki/Big_Five_personality_traits

periheli0n 5y ago
Thank you for this deep answer, and for confirming my impression of Myers-Briggs! I can see how a continuous scale can already fix a few issues. However, isn’t it still too simplistic to say “personality feature X correlates with real-life success measure Y?” Take for example “Openness”. It’s not hard to imagine how the very same person can be very open with one group of people, but act totally differently in another group. Is Openness then really a feature of personality? Or maybe rather a description of the social behaviour of one person in a specific context?
- anbende 5y ago
  >Is Openness then really a feature of personality? Or maybe rather a description of the social behaviour of one person in a specific context? This is a great question! And it touches on what's call the "person-situation" debate, first popularized by Walter Michel[1]. Essentially there was a major debate about whether traits were a thing or we should just be focused on context-specific behavior. Mischel believed that traits weren't useful, and situations ruled. In many ways this debate has been resolved by the "person as a density distribution" [2]. People do vary a lot within, and that variation is vitally important. However, their average or "set point" upon which they vary is also important and predictive. When we talk about the correlation between a trait and some external marker, we are only looking at correlations with averages or set-points, which as you mention is only a limited picture of that person, but it is still useful and valuable to look at. For example, I'm a fairly open person. This means that on average I'm more open in more situations. And this is predictive. It's also important to note that this only loosely predicts what I'll do in any particular situation or at any particular timepoint. [1] https://en.wikipedia.org/wiki/Person%E2%80%93situation_debate https://en.wikipedia.org/wiki/Person%E2%80%93situation_debat... [2] https://personality-project.org/revelle/syllabi/classreadings/fleeson.2001.pdf https://personality-project.org/revelle/syllabi/classreading...
  - periheli0n 5y ago
    Excellent—that sounds almost like a Bayesian approach, where personality corresponds to the prior, the context in which the person behaves is the likelihood, and the posterior is a distribution of behaviours. I guess a personal goal should then be to seek the context in which one’s strengths match best the demands. I would suggest an MCMC approach but I guess the evaluation of individual outcomes is too costly to make this practical ;)
    - wizzwizz4 5y ago
      I think it's more “a personality is a function over the continuous domain of situations”.
      - periheli0n 5y ago
        Not a contradiction! Bayesian inference is fully compatible with this definition.
        wizzwizz4 5y ago
        I don't see how you could get a prior over this without some really dubious assumptions. Plus, it's not what GGP described.
        periheli0n 5y ago
        The prior is given by the score on the big five?
        wizzwizz4 5y ago
        And do you combine that with something like an Occam's Razor formalism?
- namenotrequired 5y ago
  > Take for example “Openness”. It’s not hard to imagine how the very same person can be very open with one group of people Lest this be misunderstood, note that "Openness" in this context is not about being honest. It is short for "Openness to Experience". Roughly, it's about creativity and curiosity. https://en.wikipedia.org/wiki/Openness_to_experience https://en.wikipedia.org/wiki/Openness_to_experience
  - periheli0n 5y ago
    Good point. Still, a person that radiates creativity and curiosity in an environment that they find inspiring could be the exact opposite in another environment that they find less suitable. Hence my point that this personality trait stuff is at least as much dependent on context as on personal traits.
    - planet-and-halo 5y ago
      That's widely accepted as part of personality theory. The idea is that you have some "base setting," not that you will literally try anything (or refuse to try anything). It should also be noted that personality is one aspect of your psychology. It doesn't absolutely determine how you will behave under a given set of circumstances. Example: you might be very open to experience, but as a child you were bit by a dog. Next time you see a dog, your experience-based fear could easily overwhelm your natural curiosity.
    - namenotrequired 5y ago
      Definitely! I only didn't respond the rest of your comment because anbende did so better than I could.
mycologos 5y ago
The author of the featured article has also written a defense of the Myers-Briggs system that argues against several of your points: https://dynomight.net/in-defense-of-myers-briggs.html https://dynomight.net/in-defense-of-myers-briggs.html (I'm not particularly invested in either model, but it is somewhat cute that the author of both pieces is the same.)
- yCombLinks 5y ago
  His article defends a modified myers-briggs that is much closer to the big 5. Myers-briggs with personality traits on a distribution is extremely similar to big 5 minus neuroticism.

periheli0n 5y ago

Thank you for this deep answer, and for confirming my impression of Myers-Briggs! I can see how a continuous scale can already fix a few issues. However, isn’t it still too simplistic to say “personality feature X correlates with real-life success measure Y?” Take for example “Openness”. It’s not hard to imagine how the very same person can be very open with one group of people, but act totally differently in another group. Is Openness then really a feature of personality? Or maybe rather a description of the social behaviour of one person in a specific context?

anbende 5y ago
>Is Openness then really a feature of personality? Or maybe rather a description of the social behaviour of one person in a specific context? This is a great question! And it touches on what's call the "person-situation" debate, first popularized by Walter Michel[1]. Essentially there was a major debate about whether traits were a thing or we should just be focused on context-specific behavior. Mischel believed that traits weren't useful, and situations ruled. In many ways this debate has been resolved by the "person as a density distribution" [2]. People do vary a lot within, and that variation is vitally important. However, their average or "set point" upon which they vary is also important and predictive. When we talk about the correlation between a trait and some external marker, we are only looking at correlations with averages or set-points, which as you mention is only a limited picture of that person, but it is still useful and valuable to look at. For example, I'm a fairly open person. This means that on average I'm more open in more situations. And this is predictive. It's also important to note that this only loosely predicts what I'll do in any particular situation or at any particular timepoint. [1] https://en.wikipedia.org/wiki/Person%E2%80%93situation_debate https://en.wikipedia.org/wiki/Person%E2%80%93situation_debat... [2] https://personality-project.org/revelle/syllabi/classreadings/fleeson.2001.pdf https://personality-project.org/revelle/syllabi/classreading...
- periheli0n 5y ago
  Excellent—that sounds almost like a Bayesian approach, where personality corresponds to the prior, the context in which the person behaves is the likelihood, and the posterior is a distribution of behaviours. I guess a personal goal should then be to seek the context in which one’s strengths match best the demands. I would suggest an MCMC approach but I guess the evaluation of individual outcomes is too costly to make this practical ;)
  - wizzwizz4 5y ago
    I think it's more “a personality is a function over the continuous domain of situations”.
    - periheli0n 5y ago
      Not a contradiction! Bayesian inference is fully compatible with this definition.
      - wizzwizz4 5y ago
        I don't see how you could get a prior over this without some really dubious assumptions. Plus, it's not what GGP described.
        periheli0n 5y ago
        The prior is given by the score on the big five?
        wizzwizz4 5y ago
        And do you combine that with something like an Occam's Razor formalism?
namenotrequired 5y ago
> Take for example “Openness”. It’s not hard to imagine how the very same person can be very open with one group of people Lest this be misunderstood, note that "Openness" in this context is not about being honest. It is short for "Openness to Experience". Roughly, it's about creativity and curiosity. https://en.wikipedia.org/wiki/Openness_to_experience https://en.wikipedia.org/wiki/Openness_to_experience
- periheli0n 5y ago
  Good point. Still, a person that radiates creativity and curiosity in an environment that they find inspiring could be the exact opposite in another environment that they find less suitable. Hence my point that this personality trait stuff is at least as much dependent on context as on personal traits.
  - planet-and-halo 5y ago
    That's widely accepted as part of personality theory. The idea is that you have some "base setting," not that you will literally try anything (or refuse to try anything). It should also be noted that personality is one aspect of your psychology. It doesn't absolutely determine how you will behave under a given set of circumstances. Example: you might be very open to experience, but as a child you were bit by a dog. Next time you see a dog, your experience-based fear could easily overwhelm your natural curiosity.
  - namenotrequired 5y ago
    Definitely! I only didn't respond the rest of your comment because anbende did so better than I could.

mycologos 5y ago

The author of the featured article has also written a defense of the Myers-Briggs system that argues against several of your points: https://dynomight.net/in-defense-of-myers-briggs.html https://dynomight.net/in-defense-of-myers-briggs.html (I'm not particularly invested in either model, but it is somewhat cute that the author of both pieces is the same.)

yCombLinks 5y ago
His article defends a modified myers-briggs that is much closer to the big 5. Myers-briggs with personality traits on a distribution is extremely similar to big 5 minus neuroticism.