I suspect you've been exposed to more than a handful of forum posts (not to mention other sources) claiming that Jungian/MBTI type is largely about the so-called "cognitive functions," and/or that the dichotomies mostly deal with more superficial stuff, and/or that you should think of the dichotomies as "letter codes" that need to be
decoded to lead you to the deeper stuff.
But I'm here to tell you that those posters have been taken for a ride. Not even Jung himself prioritized the eight functions in the way that a lot of forumites do. In fact, Jung spent more of
Psychological Types talking about the things he thought extraverts had in common and introverts had in common than he spent talking about all eight of the functions put together; and in the Foreword to a 1934 edition of the book, he bemoaned the fact that too many people were inclined to view Chapter 10 (his function descriptions) as the essence of the book, while noting that he'd stuck those at the back of the book for a reason.
And in any event, and regardless of what Jung's perspective was, it's been close to 100 years since
Psychological Types was published, and the modern MBTI reflects countless adjustments and improvements to Jung's original concepts.
Contrary to what you sometimes hear, and notwithstanding that there are important distinctions to be made between "hard sciences" and "soft sciences," the four MBTI dichotomies now have decades of data in support of their validity and reliability — and a combination of meta-review and large supplemental study in 2003 concluded that the MBTI was more or less in the same category (if not on a par) with the Big Five in terms of its psychometric respectability. And if you're interested, you can read more about that — and about several other issues often raised by people claiming to "debunk" the MBTI — in
this post and the post that follows it.
Carl Jung — mystical streak notwithstanding — was a believer in the scientific approach, and Isabel Myers took
Psychological Types and devoted a substantial chunk of her life to putting its typological concepts to the test in a way that Jung never had, and in accordance with the psychometric standards applicable to the
science of personality. Myers adjusted Jung's categories and concepts so that they better fit the data she gathered from thousands of subjects, and by the start of the 1960s (as the leading Big Five psychologists have acknowledged), she had a typology that was respectably tapping into four of the Big Five personality dimensions — long before there really was a Big Five. And twin studies have since shown that
identical twins raised in separate households are substantially more likely to match on those dimensions than genetically unrelated pairs, which is further (strong) confirmation that the MBTI dichotomies correspond to
real, relatively hard-wired underlying dimensions of personality. They're a long way from being simply theoretical — or pseudoscientific — categories with no respectable evidence behind them.
Buuut contrary to what some of the function aficionados would have you believe, the scientifically respectable side of the MBTI is the dichotomy-centric side — and the dichotomies differ greatly from the so-called "cognitive functions" in that regard. The functions — which James Reynierse (in
"The Case Against Type Dynamics") rightly characterizes as a "category mistake" — have barely even been studied, and the reason they've barely been studied is that, unlike the dichotomies, they've never been taken seriously by any significant number of academic psychologists. Going all the way back to 1985, the MBTI Manual described or referred to somewhere in the neighborhood of 1,500 MBTI studies, and as I understand it,
not one of the many study-based correlations reported in the manual were framed in terms of the functions. The third edition of the MBTI Manual was published in 1998 and, as Reynierse notes in that same article, it cited a grand total of
eight studies involving "type dynamics" (i.e., the functions model) — which Reynierse summarizes as "six studies that failed, one with a questionable interpretation, and one where contradictory evidence was offered as support." He then notes: "Type theory's claim that type dynamics is superior to the static model and the straightforward contribution of the individual preferences rests on this ephemeral empirical foundation."
As I'm always pointing out, the modern function descriptions you'll find in Thomson, Berens, Nardi, etc. differ in many ways (large and small) from Jung's original concepts, and appear to be a set of descriptions more or less jerry-rigged to match up reasonably well with the MBTI types they purportedly correspond with. (As one dramatic example, and as described at length in
this post, the description of "Si" you'll find Thomson, Berens, Nardi and Quenk using bears little resemblance to Jung's "introverted sensation" and is instead a description made to match MBTI SJs. And you can read about the changes Myers made to Te in
this PerC post.)
So... since "Ne" descriptions are set up to match NPs (extraverts and introverts both) reasonably well (since the IN_Ps are "Ne-aux" types) and "Ti" descriptions are set up to match TPs (extraverts and introverts both) reasonably well (since the E_TPs are "Ti-aux" types), it's not surprising that INTPs and ENTPs both read those modern Ne and Ti descriptions and feel like they relate reasonably well. (Although I can't help noting that, as discussed in the spoiler in
this post, INTJs often relate pretty well to Ne and Ti descriptions as well....)
As a general matter, in other words... if you're looking at those modern cognitive function descriptions, and you're applying them to the types who purportedly have them as their
dominant or auxiliary functions, you're likely to get quite a bit of
piggybacked validity, because if an "Ne" description is largely made up of things that NPs tend to have in common, it's obviously going to be reasonably valid for NPs.
But none of that has anything to do with whether the functions — as components of a four-function "type dynamics" model — beat Reynierse's "category mistake" rap.
As Reynierse notes, virtually all the respectable data/studies point to the conclusion that the actual, at-least-semi-genetic, underlying components of MBTI personality are the four dichotomies, and that personality characteristics that result from (or are influenced by) a combination of two or more of the dichotomies are affected in a simple, additive way. So what I call the "Real MBTI Model" — partly because, if you ignore the Jungian lip service, it was essentially Isabel Myers's model — looks like this:
INTP = I + N + T + P + IN + IT + IP + NT + NP + TP + INT + INP + ITP + NTP + INTP.
INTJ = I + N + T + J + IN + IT + IJ + NT + NJ + TJ + INT + INJ + ITJ + NTJ + INTJ.
ESFJ = E + S + F + J + ES + EF + EJ + SF + SJ + FJ + ESF + ESJ + EFJ + SFJ + ESFJ.
And as already explained, type analysis that's supposedly based on the cognitive functions often "works" — in the sense of matching up reasonably well with the corresponding types — to the extent that it simply overlaps with the Real MBTI Model. But
on the other hand, and as Reynierse emphasizes, if what you're talking about is assertions about the types that are based on "cognitive functions" analysis and that are
inconsistent with what the Real MBTI Model would lead you to expect, those kinds of assertions have never had any respectable body of data support behind them, and in fact have tended to be
contradicted by the results of relevant MBTI studies.
As one example, those
inconsistent assertions include the idea that an INTP's personality tends to exhibit
both "Ne" characteristics (so far, OK)
and "Si" characteristics — with the Ne and the Si working together "in tandem."
The Real MBTI Model for INTPs and ESFJs (see above) says that INTPs and ESFJs have
no MBTI-related personality characteristics in common, because they're opposites on all four dimensions — and therefore also opposites with respect to all the relevant
preference combinations that make contributions to their personality. According to the Real MBTI Model, the INTP is indeed an "Ne" type (to the extent that your "Ne" description is made up of characteristics that NPs tend to have in common) but is
not an "Si" type (to the extent that your "Si" description is made up of characteristics that SJs tend to have in common).
By contrast, the cognitive functions model — in its tandem-based incarnation (the Harold Grant function stack) — says that INTPs and ESFJs share quite a bit in the way of MBTI-related personality characteristics because they're both "Si" types (not to mention "Ne" and "Ti" and "Fe" types), and that the types who tend to
not share those "Si" personality characteristics are the "Se" types (e.g., ENTJs with "tertiary Se").
Assuming that Ne, Ni, Se and Si correspond to significant aspects of personality — and if they're used to refer to stuff that NPs, NJs, SPs and SJs, respectively, have in common, they certainly do — and given that we now have data pools (thousands of them, over 50 years) correlating the MBTI types with a huge variety of things (including countless personality dimensions as measured by other personality typologies),
if it was true that NJs and SPs are both "Ni/Se types," and that NPs and SJs are both "Ne/Si types," you'd expect there would be some significant body of data pools where Ne, Ni, Se and/or Si were the most significant MBTI-related influences on whatever the study was looking at — and where, accordingly, the NJs and SPs showed up on one side of the correlational spectrum and the NPs and SJs showed up on the other.
But that correlational pattern — like all patterns that are inconsistent with the Real MBTI Model — virtually never shows up. Instead, regardless of what aspect of personality it is that somebody's study may be focusing on, if the SJs show up at one end of the spectrum, look for the NPs to show up
at the other end — just as the Real MBTI Model would lead you to expect.
And there's quite a bit more on that in the "bogosity of the tandems" post linked to below.
As another example of type dynamics leading to expectations that go beyond the Real MBTI Model, the function-centric perspective says that, comparing an INTJ and an INTP, the INTJs' N will generally play a greater role in their personality than their T and the INTPs' T will generally play a greater role than their N — because dom/aux! — and that notion, too, has no respectable validity. INTJs and INTPs both have N and T preferences, with all that those entail, and whether the N or the T plays a greater role in any NT's personality will basically depend on whether one of those two preferences is substantially stronger than the other — and the data suggests that the N preference is no more likely to be the stronger one for an INTJ than for an INTP.
And meanwhile, the fact that function-centric analyses of personality tend to
shortchange (if not completely miss) many significant MBTI-related aspects of personality is at least as substantial a shortcoming as those data-challenged assertions that the popular function models make. I previously noted that type analysis that's supposedly based on the cognitive functions often "works" — in the sense of matching up reasonably well with the corresponding types — to the extent that it simply overlaps with the Real MBTI Model. Buuut when I say it "works," it's important to note how grotesquely
incomplete an INTP profile is going to be if it stops at (or is overly dominated by) "Ti" (TP) + "Ne" (NP) — since it means that the profile is likely to be missing (or to shortchange) quite a lot of the stuff that's common to all introverts, and all N's, and all T's, and all P's, and all IN's, and all NTs, and so on.
If you're interested in reading more about the INTP=Ti-Ne-Si-Fe model — a function stack that's inconsistent with both Jung
and Myers and has never been endorsed by the official MBTI folks — and about the relationship between the dichotomies and the functions and the place of the functions (or lack thereof) in the MBTI's history, you can find quite a bit of further discussion in these two posts:
The bogosity of the "tandems"
Why I'm a "dichotomies guy"