What type is most different from INFP?

Ardielley said:

If you completely reverse an INFP's functions, it's ISTP. However, I would say ESTP is the type least like INFP since Fi is not only their PolR function (the one that's most foreign to them), but the type as a whole is much more brazen and outspoken than ISTP.

Click to expand...

This, exactly. In Socionics the ESTP is considered to be the INFP's "conflictor", essentially meaning that's the type you'll have the hardest time relating to. An INFP's functions actually differ a little more from an ISTP than an ESTP in terms of sheer order, but the ESTP and INFP lead with functions that hit the more vulnerable spots in one another.

Functions, schmunctions. The notion that INFPs and ESTJs have the same functions (albeit in reverse order) is inconsistent with both Jung and Myers (as further explained here) and has no respectable body of evidence behind it.

There are decades of MBTI data pools establishing the validity of the dichotomies by demonstrating statistically significant (and sometimes very dramatic) correlations between the dichotomies (and various dichotomy combinations) and lots of personality manifestations (internal and behavioral both).

And here's the main point, for purposes of this thread: Virtually every time the correlations put people with one or more MBTI preferences at one end of the relevant spectrum, you know who's at the other end? The people with the opposite preference or preferences.

As one close-to-home but dramatic example, the spoiler has June 2013 membership stats for PerC.


Who's at the top? The INs. And they're 60% of the PerC population, even though they're only around 12% of the general population (according to the official MBTI folks).

And who's at the bottom? The ESs — just as you should expect (since they're the dichotomous opposites of the INs). And they're only 4% of the PerC population (vs. 34% of the general population).

And if you analyze those same stats from a functions perspective? Well, the INs are obviously a dramatically tidy foursome from a dichotomy-centric perspective, but they supposedly have three different dominant functions (Ni, Ti and Fi). Good luck coming up with any function-based set of influences on personality-forum membership that seems to account for much of the variation in those PerC stats.

But so what, you might say. Maybe personality-forum membership just happens to be one of those things that's dichotomy-related but not particularly function-related.

To which I'd reply: OK, but... if INFPs and ESTJs are both "Fi/Te" types and "Si/Ne" types, and if being an Fi/Te type or Si/Ne type corresponds to some significant real-world stuff, then there should be at least some data pools where Fi/Te vs. Fe/Ti (and/or Si/Ne vs. Se/Ni) turn out to be the main influential factors, and the ESTJs and INFPs are together on one side of the spectrum and the INFJs and ESTPs are together on the other side.

And the thing is ... are y'all listening? (*drum roll*) ... there really aren't any data pools like that.

And that's because the Harold Grant function model (where INFP=Fi-Ne-Si-Te) is bogus — as is any model that would lead you to predict that, if you take Person A and flip two of their MBTI preferences, the result will be someone who's more like Person A (in some significant MBTI-related respects) than if you'd only flipped one of those two preferences.

And again, that lack of data pools wouldn't have surprised either Jung or Myers, because that weird double-flip expectation isn't really consistent with either of their type models.

So, in conclusion...

Q: What type is most different from INFP?

A: ESTJ.

ESTP is the most contrary type to the INFP. Whoever says ESTJ is being shortsighted.

proto said:

Thinking of myself as the opposite of an INFP is pretty funny. That means ISTPs are the least ethical type out of all the types. Muhauhuahauha.

Click to expand...

That implies that it's a spectrum, which would also make ISTP the most logical of all the types. Huehuehuehue.

reckful said:

INFP 3723 — 21%
INFJ 2580 — 15%
INTP 2228 — 13%
INTJ 1876 — 11%
ENFP 1352 — 8%
ENTP 1112 — 6%
ENFJ 514 — 3%
ISTP 527 — 3%
ISFP 506 — 3%
ISTJ 437 — 2%
ENTJ 401 — 2%
ISFJ 314 — 2%
ESTP 159 — 1%
ESFJ 102 — 1%
ESFP 117 — 1%
ESTJ 97 — 1%

Click to expand...

But then, what about ENFJ and ISTP? ENFJ is Fe-Ni, Se-Ti, and ISTP is Ti-Se, Ni-Fe. They're together on the list.

And if we go by what you say, shouldn't the list look something like a mirror image of itself? Why are some of the opposites not totally opposite each other like INFP and ESTJ are?

Reluctanine said:

But then, what about ENFJ and ISTP? ENFJ is Fe-Ni, Se-Ti, and ISTP is Ti-Se, Ni-Fe. They're together on the list.

And if we go by what you say, shouldn't the list look something like a mirror image of itself? Why are some of the opposites not totally opposite each other like INFP and ESTJ are?

Click to expand...

Stats in the psychological/social sciences are not like, say, physics data, where you can (at least sometimes) expect anything like perfectly consistent and 100%-predictable results. For one thing, no one claims that MBTI preferences account for anything like all of the psychological and other factors that can come into play with respect to any particular person's attitudes, behavior, etc. And for another, there are generally a lot more sources of error (think about mistyping errors, for example) in the soft sciences than in the hard sciences.

That said, though, those forum membership stats reflect a couple of pretty dramatic correlations. The type frequency pattern is almost perfectly in line with a type-related explanation that says that (1) an N preference has a very large impact on the likelihood that someone will participate in personality-related internet forums, and (2) introversion also has a substantial impact (but not as large as an N preference).

So, from the standpoint of what personality psychologists call the "validity" of the MBTI, the magnitude of the correlations in the PerC membership stats make them a reasonably good sample in support of the validity of the E/I and S/N dichotomies.

ADDED: And to maybe clear up another misunderstanding: You asked why "some of the opposites" aren't "totally opposite each other" — but you should only expect to see that kind of pattern in a case where all four of the MBTI dichotomies are making a substantial contribution to whatever it is you're looking at.

reckful said:

Stats in the psychological/social sciences are not like, say, physics data, where you can (at least sometimes) expect anything like perfectly consistent and 100%-predictable results. For one thing, no one claims that MBTI preferences account for anything like all of the psychological and other factors that can come into play with respect to any particular person's attitudes, behavior, etc. And for another, there are generally a lot more sources of error (think about mistyping errors, for example) in the soft sciences than in the hard sciences.

Click to expand...

Ahhh okay, yes, I agree with this! So.... how do I know which trend to take into account and which one not to? I mean, it's possible that a lot of people are mistyped and the results snowball and affect things like PerC forum membership. Is there a way to differentiate which data to follow?

I mean, if there's an N-bias in personality typing online, because we all feel like/think that we have a lot of ideas, then it'll affect results.

That said, though, those forum membership stats reflect a couple of pretty dramatic correlations. The type frequency pattern is almost perfectly in line with a type-related explanation that says that (1) an N preference has a very large impact on the likelihood that someone will participate in personality-related internet forums, and (2) introversion also has a substantial impact (but not as large as an N preference).

So, from the standpoint of what personality psychologists call the "validity" of the MBTI, the magnitude of the correlations in the PerC membership stats make them a reasonably good sample in support of the validity of the E/I and S/N dichotomies.

Click to expand...

But then, all you can say from the data is that IN types like to use personality forums more than ES types. How do you know ESTJs are super different from INFPs? Like, erhm, how are they thinking differently?

Reluctanine said:

Ahhh okay, yes, I agree with this! So.... how do I know which trend to take into account and which one not to? I mean, it's possible that a lot of people are mistyped and the results snowball and affect things like PerC forum membership. Is there a way to differentiate which data to follow?

I mean, if there's an N-bias in personality typing online, because we all feel like/think that we have a lot of ideas, then it'll affect results.


But then, all you can say from the data is that IN types like to use personality forums more than ES types. How do you know ESTJs are super different from INFPs? Like, erhm, how are they thinking differently?

Click to expand...

Two points.

First: You're misunderstanding what I claimed those stats illustrated. I wasn't saying that particular data pool showed that ESTJs were the most different from INFPs. My post made the point that, if the Harold Grant functions model was correct about "Fi/Te" types and "Si/Ne" types and so on, and if the so-called "cognitive functions" that the model uses are significant things, then there ought to be data pools where — because the main influential MBTI factor with respect to whatever's being measured is "Fi/Te vs. Fe/Ti" — the FPs and TJs are on one side of the spectrum and the FJs and TPs are on the other. And instead, MBTI data pools with a two-preference combination (e.g., FPs) on one side virtually never reflect that "double-flip" result. Instead, if the FPs (for example) are on one side of the spectrum, the TJs are on the other. And the PerC forum stats were simply offered as one example of what "validity" looks like, and the fact that what MBTI data pools have generally "validated" is the notion that the more (relevant) preferences you flip, the farther apart on the spectrum the affected types get — rather than the "double flip" expectation that the Harold Grant model predicts (for aspects of personality affected by the functions).

Second: Any time you're doing studies where the results take the form of correlations, most sources of error are going to introduce noise into the data that has the effect of reducing the magnitude of the reported correlations. So if you assume there's a substantial amount of typing error involved in those stats, that actually makes the correlations more impressive (from a validity standpoint), rather than less. And in fact, if you assume that the errors are mostly of the "random noise" variety — i.e., if it's as likely that an N will mistype as an S as vice versa — then, since the ratio of N's to S's is about 4 to 1 (in the PerC membership pool), you'd expect there to be substantially more N's mistyped as S's than vice versa, in which case, if the errors were corrected, the membership type ratios would be even more lopsided than they already are. In any case, given the spectacular lopsidedness of the PerC stats (particularly against the background of S's outnumbering N's two-to-one in the general population), you can assume a pretty substantial amount of error of pretty much any kind you want (random or otherwise), and it's not going to effectively undercut the notion that, to at least some relatively substantial degree, the average MBTI N is more likely to be a personality-forum member than the average MBTI S.

If you say the type which is most different, logically it means the opposite.
INFP = Fi, Ne, Si, Te
so there will be three option:

based on the letter of MBTI:
INFP != ESTJ
However, INFPs in some condition tend to be an ESTJ and vice versa because ESTJ is not the opposite but the shadow (link). So, it means that ESTJ is not the most different.

based on the cognitive function:
Fi Ne Si Te != Fe Ni Se Ti (ENFJ) or
Fi Ne Si Te != Ti Se Ni Fe (ISTP)
ENFJ is absolutely not INFP opposite because they are in the same temperament (NF= the dreamer) and there are some character which are regularly same each other. So, I choose that the most different is ISTP.

reckful said:

And that's because the Harold Grant function model (where INFP=Fi-Ne-Si-Te) is bogus — as is any model that would lead you to predict that, if you take Person A and flip two of their MBTI preferences, the result will be someone who's more like Person A (in some significant MBTI-related respects) than if you'd only flipped one of those two preferences.

And again, that lack of data pools wouldn't have surprised either Jung or Myers, because that weird double-flip expectation isn't really consistent with either of their type models.

Click to expand...

...huh? Why is this so-called double-flip expectation "weird"? You realize you're talking about notation, not theory, right? This is like arguing that it's ridiculous to say that 09 vs. 10 are closer to each other than 01 vs. 09 are, because two digits changed in the former and only one did in the latter.

When you say ESTJ is the opposite of INFP, you're not necessarily contradicting the above model. Your model is simply more shallow, in that it only considers Fi-Ne. I'm not saying your expectation is necessarily less consistent; I'm saying that this "double-flip" nonsense has absolutely nothing to do with it.

That said, I'm also not saying your expectation is more consistent either. It seems like you're looking for the wrong kind of data. We're talking about internal processing preferences of minds, and you're looking for general statistics of activities like forum membership. Really? That isn't going to tell you personal motivations or how each member is experiencing the activity in their minds. I doubt you will ever be satisfied, though, because you can't really quantify the experiences of individual interpersonal relationships in the same way. But I can certainly tell you that my personal clashes are more function-related than dichotomy-related.

pernoctator said:

That said, I'm also not saying your expectation is more consistent either. It seems like you're looking for the wrong kind of data. We're talking about internal processing preferences of minds, and you're looking for general statistics of activities like forum membership. Really? That isn't going to tell you personal motivations or how each member is experiencing the activity in their minds. I doubt you will ever be satisfied, though, because you can't really quantify the experiences of individual interpersonal relationships in the same way. But I can certainly tell you that my personal clashes are more function-related than dichotomy-related.

Click to expand...

I'm not talking about "activites" vs. "motivations" or any other flavor of external vs. internal. The whole notion that Jung was all about internal stuff and the MBTI dichotomies are all about "shallow" external stuff is pretty much just part of the silly "cognitive functions" mythos that gets endlessly passed around in the Great Forum Echo Chamber, and it's at least 90% horseshit. At the end of the day, Jung's typology and the MBTI both deal, at their core, with internal temperament dimensions and the various ways they end up being typically manifested both internally (by way of values, motivations, thinking processes, attitudes, emotional responses, etc.) and externally (through speech and behavior). And if you're interested, you can read more about that here.

When I said there's no respectable body of data that reflects that counterintuitive Harold Grant "double flip," I wasn't saying there are no supporting data pools that relate to external/behavioral activities. I'm saying there's no data support for the double-flips of any kind. And by contrast, there's decades of data supporting the validity of the MBTI dichotomies, and it relates to internal things and external things both.

And hopefully needless to say, it's usually the case that many aspects of a person's external behavior and activities significantly reflect internal personality characteristics. And I assume you wouldn't argue that that isn't likely to be the case with respect to a lot of people's membership at personality-related internet forums.

If you're open to some serious deprogramming on the dichotomies-vs.-functions front, you could do worse than start with this post.

reckful said:

The whole notion that Jung was all about internal stuff and the MBTI dichotomies are all about "shallow" external stuff...

Click to expand...

...is not what I'm saying at all.

reckful said:

I'm saying there's no data support for the double-flips of any kind. And by contrast, there's decades of data supporting the validity of the MBTI dichotomies, and it relates to internal things and external things both.

Click to expand...

Naturally, since "double-flip" is your own made-up term. MBTI dichotomies are what we are talking about.

reckful said:

And hopefully needless to say, it's usually the case that many aspects of a person's external behavior and activities significantly reflect internal personality characteristics.

Click to expand...

Yes, but the problem is that any wide-scale external behavioral data has too many valid interpretations. You're saying that your angle provides the best view because it's the only one you can find, but you may be looking through the wrong lens in the first place.

pernoctator said:

Naturally, since "double-flip" is your own made-up term. MBTI dichotomies are what we are talking about.

Click to expand...

Well, assuming somebody's talking about the MBTI dichotomies...

If you start with an INFP with four reasonably strong preferences and you "flip" the P to J, anybody who believes the dichotomies reflect meaningful aspects of personality is going to agree that that flip is going to result in the INFJ being different from the INFP in some significant MBTI-related ways.

And if you start with that same INFP and you flip the F to T, anybody who believes the dichotomies reflect meaningful aspects of personality is going to agree that that flip is going to result in the INTP being different from the INFP in some significant MBTI-related ways.

And if you start with that same INFP and you flip both the F and P preferences (to TJ), you've certainly performed a "double flip," whether that's a particular term I've coined or not.

And if you're a believer in the Harold Grant functions model and you say that, if you flip both the T/F and J/P preferences, then you end up with somebody who's more like the original person — at least when it comes to those "Ti/Te/Fi/Fe" aspects of personality — than if you'd only flipped one dichotomy, then you've arrived at what I think most people would agree is a somewhat counterintuitive result.

But please note: Just because something seems counterintuitive doesn't necessarily make it wrong, because sometimes true stuff is counterintuitive and truth can be stranger than fiction and all that. In this case, though, that "double flip" model prediction, besides being counterintuitive, has also colossally failed to find any support in over 50 years of MBTI data gathering.

If you're looking at a data pool that relates to something on which the T/F and J/P dichotomies both appear to have a significant impact, and if the FPs are basically at one end of the relevant spectrum, it is virtually never the case that the TJs (their purported fellow-Fi/Te types) are together with the FPs at that end of the spectrum and the FJs and TPs are together on the other side. Instead, as I've previously posted, if the FPs are at one end of the spectrum, you can pretty much count on finding the TJs at the other end.