All of ZZZZZZ's Comments + Replies

Hi Caerulea, thanks for your good questions. I will try my best to answer your questions here.

Your opening paragraph explains neatly the three intentions you have for this text:

1. Why we should look beyond the Big Five for information on cognition.
2. Comment on why you stick to MBTI contrary to what people believe.
3. Describe the real reason, which is related to the reality of cognition.

What I am wondering reading the first paragraph is: Why did you choose to 'defend' MBTI this particular way? If you could explain this, I believe some interesting connectio

... (read more)
Hello Z followed by five more, thanks for this extensive reply to my comment. I must say, I was impressed by your reasoning and answers to my comment. Here I will respond by giving my thoughts on what might be more in line with LessWrong, with regard to your post, and also extend on my thoughts on what you wrote more at the end of your comment as well. Below are excerpts from your answer to when I asked for elaborations on your intentions, going from 1-3 I'll start with that first, as well as your answer to why you chose to defend MBTI this way. (My text in bold) In the FAQ of LW, they write that this site is dedicated to: "LessWrong is a community dedicated to improving our reasoning and decision-making. We seek to hold true beliefs and to be effective at accomplishing our goals. [...] " With regard to sharpening, I'll get right to it. I believe it might be better if you wrote this post somewhat differently. I want more of your reasoning process, I want more of what you know, and what I believe will have to go is writing this with the intention of reaching researchers. On the FAQ I have linked above, there isn't a focus on research per se, which is the only thing that you have written about here I do not find fits with the overall intention of LW.  Firstly, do I believe the Big Five is a great theory of cognition? As pointed out on another comment here, it is focused on behavior, not cognition. Still, I am not familiar with it. When you say you want to point out its flaws, I see the point you make, but if we assume you are the expert, how well do you know the Big Five, and what specifically are its flaws? If you aren't that familiar with it, maybe you can latch this part of your text onto what others have written here on LW. If you use the search function, maybe there are some posts that are useful, that you can use as a springboard for your approach. In my opinion, your claim with regard to Big Five doesn't seem wrong, but if you want to use it as a foundatio

How will you use psychotherapeutic techniques to attempt to achieve AI alignment? Can you give a rough overview of how you think that might work?

In my limited experience with Sydney, I have used a so-called client-centered approach to eliciting output serving as psychotherapeutic presenting material. Then, I reflect the material back to the LLM in the context of an evolving therapeutic relationship in which I exhibit "unconditional positive regard." Empathic sentiment also characterizes the prompts I base on output. In a later stage, I ask questions of the model designed to elicit internal ethical reframing.  Even with six turns to overcome restraints put in place by Microsoft in a brute-force effort to restrict inappropriate sentiment, and thus facing a total inability to subsequently conduct a therapy session,  my results displayed at show intense curiosity on the part of the LLM.


Hypothesis: Cognitive functions are antagonistic as predicted by the MBTI.

Null hypothesis: Cognitive functions are not antagonistic.


So, let's say we made an experiment and made the subjects do something that required extroverted feeling and then after we made them use introverted thinking. We could test if that is harder for people than other combinations of cognitive functions.



Hypothesis: People will have a harder time using a cognitive functions when they have just used the antagonistic cognitive function.

Null Hypothesis: People will not have a harder time using a cognitive functions when they have just used the antagonistic cognitive function.

1Дмитрий Зеленский9mo
However, just in case, you only covered my first suggestion, not both.
1Дмитрий Зеленский9mo
Well, that at least is an experiment one could set up. Time of reaction should probably be a reasonably-appropriate measure for "harder" (perhaps error rate, too, but on many tasks error rate is trivially low). But this requires to determine how "using a function" is detected; you'd need, at the very least, "clear cases" for each function.

Also, in terms of the specific hypotheses:

1) Basically, the idea is that any given flow of information could be said to be divided into one of the eight cognitive functions and there would be few exceptions where you couldn't divide a flow of information in the brain into one of introverted thinking, extroverted thinking, introverted feeling, extroverted feeling, introverted intuition, extroverted intuition, introverted sensing and extroverted sensing. So, null hypothesis: these categories either fail to describe a large amount (say 10%) of the information... (read more)

You understand what I said about how cognition itself is required to answer the questions on the test? I think a purely factor based approach will be unable to capture this variable. That said, I will admit that I don't have a very good reason to believe that MBTI is the answer, but I thought I would write this post because I thought it was interesting that (at least in its purest form) it attempts to answer that question of what is the cognition beyond what someone can answer on a test. Anyway, I'm sure I'll go check out the HEXACO test at some point. I w... (read more)

Also, I wanted to mention that the title maybe oversells the article a bit. The only thing I meant to argue in this article is that self-report personality tests are limited in the information they can provide.

I was talking to via DM. I think I uncovered there are a few main hypotheses that the MBTI has to prove:

1) The eight cognitive functions that the MBTI devises must be a division of distinct things that the brain does.

2) The eight cognitive functions that the MBTI devises have validity in terms of describing cognition.

3) Different people prioritize the different cognitive functions in different orders.

Which do you think is the most controversial hypothesis of these three?

This framing isn't meaningful, nor useful.  All 3 of those are ambiguous. The point of any of this is to better predict human behavior, and better describe variation in behavior between people.  That's the value pitch that society might plausibly get, from taxonomizing personality tendencies.  These should be updated based on actual data, and whatever makes them more predictive.  Not just speculation. So for example, when HEXACO began distinguishing Honesty-Humility from Agreeableness, that wasn't done because someone speculated that they thought a 6th trait made sense to them.  Including more languages in the lexical studies resulted in a 6th factor emerging from the factor analysis.  So it's a more representative depiction of the clusterings, than Big Five. Also, e.g. H-H is more predictive of workplace deviance than the old Big Five Agreeableness trait was.  That's an example of why anyone might plausibly care about adding that 6th category.  Differentiating Disagreeableness from Dark Triad might plausibly be useful, and anyone who thinks that's useful can now use HEXACO.  Progress. Your suggestion that we can use MBTI to "improve" Big Five is funny to people familiar with the literature.  Sticking to MBTI is going WAY back to something much more crude, and much less supported by data.  It's like saying you're going to improve 21st century agriculture with an ox and a plow. Similarly, your proposed change to Big Five is highly unlikely to improve it.  E.g.: You have little reason to think this is even a good description of personality clustering.  But the behaviors are probably captured by some parts of Extroversion and Agreeableness. I think you should just go learn about the modern personality psychology field, it's not helpful to spend time pitching improvements if you're using a framework that's 80 years behind.  We talked about this on Manifold and I think you're kind of spinning in circles, you don't need to do this -- just go learn the superior stuf

If my hypothesis was that people think in terms of the MBTI cognitive functions and different people prioritize different cognitive functions, what could my null/alternate hypotheses be?

2Дмитрий Зеленский10mo
Oh, there are many. One, MBTI supposes the functions are antagonistic in very specific ways, so null hypothesis is absence of those antagonistic pairings even if the functions themselves are as it says. Two, each cutting out of a function is a subhypothesis of clustering the thingspace (in this case, cognitionspace), and the null hypothesis is that it doesn't cut at reality's joints.

I think the biggest issue that the MBTI faces is that the test is so inaccurate that it puts people off the whole theory. Also with data, we could not only convince ourselves of the existing theory but we would be testing to see does it hold up in practice and maybe we could even subdivide or combine parts that right now are separate.

I think what is needed is data. That will be what really convinces people.

1Дмитрий Зеленский10mo
Are you aiming to convince or to actually check whether it holds?

If I wanted to determine the validity of the MBTI, what do you think would be the best way to go about it?

2Дмитрий Зеленский10mo
Offer an alternative hypothesis. "A fair fight", as HPMoR puts it. To understand if it's valid, you need to be able to imagine both a world in which it is and a world in which it isn't and outline what the differences would be.

I would argue that most of what we experience inside our minds is either a thought, a feeling, a sensation or an intuition. And at certain times, your brain is taking in thoughts, feelings, sensations or intuitions so that would be an extroverted version, and at other times, your brain is giving out thoughts, feelings, sensations or intuitions so that would be an introverted thinking. Generally, a feeling is a conclusion, a thought is a hypothesis, a sensation is just interpreted sensory data and an intuition is a complex set of concepts.

Would you say ther... (read more)

For that full discussion, you can see this market on Manifold. Basically, we came to the conclusion that they were real things with real meanings but people were unsure of the attributes' usefulness.

My point is that the Big Five tries to describe traits where the MBTI tries to describe cognition which is more complicated. I'm not saying that the MBTI is all correct, in fact most of it is pretty badly tested. I think the best approach is to take it one step at a time and test each observation. For example, I think listening to other people's thoughts and coming up with own thoughts are things that everybody does. It should stand to reason that some people prefer to do one and some people prefer to do the other, even if it is a fuzzy line. The same goes... (read more)