Personality Cafe banner
1 - 20 of 41 Posts

·
Registered
Joined
·
2,374 Posts
I don't think "disproven" is quite the right term, since I'm inclined to think most of the claims about cognitive functions are set up in a manner that makes them impossible to truly evaluate. But I would say that there are issues with type dynamics that make accepting its claims difficult, so it's less that the theory is demonstrably wrong so much as that there's little reason to presume it's right - not disproven, but unproven. I've said on here a number of times that type dynamics is too rigid, in that it presupposes things about people without justification ("E-S-F-P? Oh, therefore you prefer S to F, and T to N." "Your dominant is Fe? Then your auxiliary can only be Ni or Si."), which potentially prevents people finding their type if/when they fall outside of the confines of the model presented; to overcome this rigidity, and to avoid having people fall into gaps, it then often becomes too nebulous, too vague, too imprecise, such that what exactly each function does is unclear or at least incredibly malleable. And that's just one of the problems, with one specific variant of cognitive functions theory; there are many others...

As for the Reyniersean alternative? The official MBTI line is that scores on tests are merely indicative of how clearly you display your preference, not of how strong your preference is; whilst denying the link between clarity and strength outright seems foolhardy, I do think that this distinction is important. If I were to call myself, say, an ITNP - the order in which I'd say my preferences apply to me - then what I am saying is "I prefer I to E more than I prefer T to F, which I prefer more than I prefer N to S, which I prefer more than P to J". It would need to be borne in mind that no comparison is made between each continuum - nowhere in ITNP does it say "I prefer I to N", or "I prefer T to P", or even "I prefer P to S" - and I suspect the temptation to infer such things could be quite a problem (the reason it doesn't follow: let's say I prefer my judging functions generally to my perceiving functions generally. Although I prefer T to F more than I prefer N to S, it may be that I am so substantially more judging-function-oriented than perceiving-function-oriented that the part of me which accords with F is greater than that which is N - that, if I were to rank [S,N,T,F] without any imposed constraint, I would say T>F>N>S. This is decidedly not the case for me personally, but I think the possibility needs to be remembered and accounted for.)

So I would agree that there's little reason to suppose type dynamics is correct (and, even if it is correct, which variant to believe is another substantial issue that many seem to ignore), and I would say that building the theory to focus more on the relevance of degree of preference is definitely a good idea, but that the model Reynierse presents is not quite as good as perhaps it would need to be, and has potential flaws of its own to the extent that it implies any relationship between the four dimensions.
 

·
Banned
Joined
·
139 Posts
Discussion Starter · #3 ·
I don't think "disproven" is quite the right term, since I'm inclined to think most of the claims about cognitive functions are set up in a manner that makes them impossible to truly evaluate. But I would say that there are issues with type dynamics that make accepting its claims difficult, so it's less that the theory is demonstrably wrong so much as that there's little reason to presume it's right - not disproven, but unproven. I've said on here a number of times that type dynamics is too rigid, in that it presupposes things about people without justification ("E-S-F-P? Oh, therefore you prefer S to F, and T to N." "Your dominant is Fe? Then your auxiliary can only be Ni or Si."), which potentially prevents people finding their type if/when they fall outside of the confines of the model presented; to overcome this rigidity, and to avoid having people fall into gaps, it then often becomes too nebulous, too vague, too imprecise, such that what exactly each function does is unclear or at least incredibly malleable. And that's just one of the problems, with one specific variant of cognitive functions theory; there are many others...

As for the Reyniersean alternative? The official MBTI line is that scores on tests are merely indicative of how clearly you display your preference, not of how strong your preference is; whilst denying the link between clarity and strength outright seems foolhardy, I do think that this distinction is important. If I were to call myself, say, an ITNP - the order in which I'd say my preferences apply to me - then what I am saying is "I prefer I to E more than I prefer T to F, which I prefer more than I prefer N to S, which I prefer more than P to J". It would need to be borne in mind that no comparison is made between each continuum - nowhere in ITNP does it say "I prefer I to N", or "I prefer T to P", or even "I prefer P to S" - and I suspect the temptation to infer such things could be quite a problem (the reason it doesn't follow: let's say I prefer my judging functions generally to my perceiving functions generally. Although I prefer T to F more than I prefer N to S, it may be that I am so substantially more judging-function-oriented than perceiving-function-oriented that the part of me which accords with F is greater than that which is N - that, if I were to rank [S,N,T,F] without any imposed constraint, I would say T>F>N>S. This is decidedly not the case for me personally, but I think the possibility needs to be remembered and accounted for.)

So I would agree that there's little reason to suppose type dynamics is correct (and, even if it is correct, which variant to believe is another substantial issue that many seem to ignore), and I would say that building the theory to focus more on the relevance of degree of preference is definitely a good idea, but that the model Reynierse presents is not quite as good as perhaps it would need to be, and has potential flaws of its own to the extent that it implies any relationship between the four dimensions.
I agree, but I chose the term as I couldn’t pinpoint which word was best to use. Maybe I should have just gone with “an alternate view?”
And about the cognitive functions, given that there are 8, how can’t there be more than 16 personality types? Sure, the four letter preference limits them, but if one were to ignore them and simply go with the functions, wouldn’t there be more types?

It’s interesting that you point out the relationship between letter preferences. How would you solve it?

Maybe you might be interested in the following link as it leads to an article on the same topic:

Reynierse's revised type theory
 

·
Registered
Joined
·
2,374 Posts
I agree, but I chose the term as I couldn’t pinpoint which word was best to use. Maybe I should have just gone with “an alternate view?”
And about the cognitive functions, given that there are 8, how can’t there be more than 16 personality types? Sure, the four letter preference limits them, but if one were to ignore them and simply go with the functions, wouldn’t there be more types?

It’s interesting that you point out the relationship between letter preferences. How would you solve it?

Maybe you might be interested in the following link as it leads to an article on the same topic:

Reynierse's revised type theory
I definitely think that, if function-attitudes are considered workable, then any theory proposed needs to either accept "unorthodox" preference combinations or justify their exclusion better than they supposedly do currently. There is nothing intrinsically problematic with, say, Fi-Ni-Se-Te. So yes, I think if function-attitudes were sufficiently quantified and a theory built around them, you'd need more than just sixteen "types". The TypologyCentral link you gave has a link to this article given on Page 3, which somewhat deals with this issue, giving 32 types by separating out function order from J/P the dimension - making J/P an independent continuum rather than making the unsubstantiated presumption that a person's J/P score will determine which function they prefer more. This deals with one of the issues in connecting type dynamics to the "letters", but issues such as why function stacks would have to be XYXY, amongst others, remain unresolved - and, of course, precise definitions for each of the function-attitudes as well.

The relationship between letter preferences issue is not so much a problem with the mooted theory as with how it would likely be interpreted. Could a means for evaluating the relative strength of, say, T and S be developed? Even then, if ITNP refers to the degree each is preferred over its opposite, then if this person happens to prefer N to T, reflecting that proves difficult. So it may simply be that, if Reynierse's model were adopted, misinterpretations of what each preference order actually means would simply have to be endured and corrected - much as misconceptions like an S preference denoting the absence of intuition continue to be had on occasion with the current MBTI.
 
  • Like
Reactions: Realeros

·
Registered
Joined
·
3,491 Posts
Here's an article I came across on the web, and which goes over 11 points as to why type dynamics and cognitive functions (in the Jungian sense) should be abandoned for an alternate view (proposed by the same critic). Here's the article:

Cognitive Functions and Type Dynamics - A Failed Theory? | Oddly Developed Types

What do you all of you think?
As further discussed in this post, I have my doubts about some of the specific additions that Reynierse thinks can usefully be made to the MBTI at this point — including the ordering of preferences by (purported) strength that @StunnedFox mentioned — but I think he's right on target with his view that the modern "cognitive functions" model that people like Thomson, Berens and Nardi have been peddling for over 15 years now is a "category mistake."

As far as whether it's been conclusively "disproven": For the reasons discussed in this post and this post, I think the Harold Grant function stack — together with its associated "tandems," where somebody's either an "Fi/Te type" or an "Fe/Ti type" and so on — should really be considered disproven at this point by anyone who likes to think of themselves as reality-oriented.

And as for the aspects of "cognitive functions" theory that don't revolve around the Harold Grant function stack but are nonetheless inconsistent with (or go beyond) the dichotomy-centric model that says that MBTI-related personality characteristics result from (1) your preferences on the four dichotomies, and (2) preferences on multiple dichotomies where the relevant characteristics correspond to multidimensional additive effects of those dichotomies — it may not be fair to call them all "disproven" at this point, but they sure don't have any respectable body of evidence behind them. As Reynierse notes in The Case Against Type Dynamics, the third edition of the MBTI Manual cites 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."

Anyone interested in more from me on the place of the functions (or lack thereof) in the MBTI's history, and the tremendous gap between the dichotomies and the functions in terms of scientific respectability, will find it in this long "why I'm a dichotomies guy" post at INTJforum.
 

·
Registered
Joined
·
31,261 Posts
I always viewed type as a code for the functions. I don't understand the foundation behind reducing it to the letters. I think descriptions would have to be removed since there seems to be no conceivable basis to having them under that system.
Lack of empirical evidence supporting it aside, functions as defined and explained by Jung along with whatever model of Cognitive functions would "wrap together" the whole type. This explains this, blah blah, and if it's not accurate in reality then I would say scrap the system entirely if I truly did not believe it contained any worth at all.
But this system by letters alone, I just fail to see the foundation behind it. If you add up that a person gets a higher N than S, etc. to get NIPT, how does one put that together into a holistic personality? What understanding would I derive them that? Obviously the old descriptions would need to be scrapped. So something like the already existing "INTP" would have descriptions. At least in some part, it would be tied together by functional analysis in these descriptions, this letter idea would just be putting "N preference traits, big picture thinking, whole to part" together with other preferences creating a list of disconnected traits that don't really contain any value, unless perhaps prediction for a simple career path. As an example, saying INTPs rely on "introverted thinking" and INTJs rely on "extroverted thinking", the idea is differentiated by functions (obviously the ideas could be denied as well to be fair).
Take an idea of INTJs being more "controlling" of their environment into rational structure and INTPs trying to "create theoretical structures and strongly value consistency". Using letters, INTJ and INTP (and I know there would be other orders but these still would be possible orders), I don't see any way of making that distinction. I don't see how putting together these preferences alone could lead to that distinction between them. The way I see it, this system has less to say. Maybe it's more consistent with studies or this or that, but as far as I can tell, it has a lot less to say and therefore older concepts of what an INTP and INTJ is would have to be scrapped, the differences and relationships between two types in MBTI often go beyond mere letter comparison, these functional attitudes are considered. Well I agree those could be denied and scrapped, so would much of the descriptions as well, much of it would be basic.
The Big Five is similar. I only took the similarminds quizzes and to my understanding there are no SLOAN descriptions like there are MBTI type descriptions. While Big Five may be more testable, it is mostly descriptive in a sense, it doesn't really explain anything. When I do the Big Five quiz on similarminds, I get back a list of traits. Traits that go unexplained, they are just put through the computer and regurgitated, the end result may be accurate to what is found in later study, but what does that even matter? It reveals nothing, there is no explanation, there just is. That which may be obvious. At least in Earth Science we have casual relationships between phenomena and natural laws, patterns. In Big Five, we just take a test and get some "empirically verified traits". Maybe it's my prejudice but seems to be void of any meaning or explanation, there is no system underlying it, it's just a compilation of data. It may have practical uses but there is nothing I can derive from it in terms of explanation.
 

·
Banned
Joined
·
139 Posts
Discussion Starter · #7 ·
I definitely think that, if function-attitudes are considered workable, then any theory proposed needs to either accept "unorthodox" preference combinations or justify their exclusion better than they supposedly do currently. There is nothing intrinsically problematic with, say, Fi-Ni-Se-Te. So yes, I think if function-attitudes were sufficiently quantified and a theory built around them, you'd need more than just sixteen "types". The TypologyCentral link you gave has a link to this article given on Page 3, which somewhat deals with this issue, giving 32 types by separating out function order from J/P the dimension - making J/P an independent continuum rather than making the unsubstantiated presumption that a person's J/P score will determine which function they prefer more. This deals with one of the issues in connecting type dynamics to the "letters", but issues such as why function stacks would have to be XYXY, amongst others, remain unresolved - and, of course, precise definitions for each of the function-attitudes as well.

The relationship between letter preferences issue is not so much a problem with the mooted theory as with how it would likely be interpreted. Could a means for evaluating the relative strength of, say, T and S be developed? Even then, if ITNP refers to the degree each is preferred over its opposite, then if this person happens to prefer N to T, reflecting that proves difficult. So it may simply be that, if Reynierse's model were adopted, misinterpretations of what each preference order actually means would simply have to be endured and corrected - much as misconceptions like an S preference denoting the absence of intuition continue to be had on occasion with the current MBTI.
Yes, starting with the cognitive functions and building a type theory from them would make sense. I'm guessing once one knows their cognitive functions apart from all it, one could measure the strength and/or preference of them, and supposing the strongest is Ni or Ti, Assume the leading letter is either N or T (and then assume opposing one)? It would define the extremes and allow one to trouble-shot the remaining letters to discover one's type code.

Like you said, the trouble would come in figuring out how to measure "strength". Even the notion of "preference" needs revisiting. There would be a new found dependence on interpretation when it seemed as though that's what typology has been trying to escape from. Almost like traveling in a circle.
 

·
Registered
Joined
·
2,374 Posts
@Grandmaster Yoda - I think the key is in distinguishing between necessary and contingent truths. The idea that INTJs might be interested in controlling their environment into rational structure whilst INTPs "create theoretical structures and strongly value consistency" is fine, if the claim can in some way be supported, but it becomes problematic if you start trying to say that it holds true for all members of the class - that every INTP strongly values consistency, &c., because making such assertions narrows the amount of people who actually fit those categories, and leaves people without a category by making the "criteria" for being a particular type too specific. You're right, we would have to scrap many type descriptions, since they're predicated on assumptions unjustifiably drawn (such as that all INTJs prefer N to T, and all INTPs T to N), but that only renders the system basic in that it scraps unfounded assertions from it; theoretically, a set of general truths about type can be built up in a more justified manner. As it stands, I can't see what it is that justifies, for instance, a person scoring as E and J and the information "You prefer T/F most" being returned...

I do think the lack of holistic engagement between preferences is a deficiency in Reynierse's model, and perhaps it would be preferable to discern some means of engaging across preferences - after all, we can prefer one thing to another in relation to things that aren't directly opposed as well as things that are, so why not also I/P, N/F, E/T questions and the like? That way, we could make more holistic determinations about a person's personality without needing to imply it on the basis of unsubstantiated assumptions. If the results seem not specific enough - if they become just a list of common traits, a picture of the average person like that rather than fitting really well - then I think that only speaks to the problem of trying to apply general truths about categories you're a member of to yourself. Just as I don't fit all the characteristics of the average male, or the average Australian, or the average PersonalityCafe member, so I don't fit all the characteristics of the average INTP. To the extent, then, that type dynamics tries to propose otherwise - which I would argue it does by mandating that a person with I, T and P preferences must prefer T to S/N, and things of that ilk - I would say it is as flawed as trying to make similarly universal claims about the other categories: unless there's some reason to suppose it necessarily must follow, we oughtn't to assume that it does.

- - - - -
@reckful - I can't claim to know exactly what it is Grant, Beebe and the like actually directly claim, but I would say your linked posts only disprove such models to the extent that their claim requires a function to operate in the same way as tertiary/inferior as you would expect it to as dominant/auxiliary. Certainly, if that's not their claim, then the onus is on them to argue how it remains the same function, and why it would manifest so differently, and things such as that, but it not being their claim would also make the disproof based on TJs being opposite to FPs rather than alike insufficient for the purpose of disproving it. I would say the bulk of studies that do show type dynamics to be flawed are those that engage directly with the MBTI system, since it is there where some evaluable assertions are actually made: it is they who expect T to be greater than N for all INTPs, whereas other functions theorists could easily dispute the MBTI testing mechanisms ability to get at their idea of the functions.
 
  • Like
Reactions: Grandmaster Yoda

·
Registered
Joined
·
31,261 Posts
@Grandmaster Yoda - I think the key is in distinguishing between necessary and contingent truths. The idea that INTJs might be interested in controlling their environment into rational structure whilst INTPs "create theoretical structures and strongly value consistency" is fine, if the claim can in some way be supported, but it becomes problematic if you start trying to say that it holds true for all members of the class - that every INTP strongly values consistency, &c., because making such assertions narrows the amount of people who actually fit those categories, and leaves people without a category by making the "criteria" for being a particular type too specific. You're right, we would have to scrap many type descriptions, since they're predicated on assumptions unjustifiably drawn (such as that all INTJs prefer N to T, and all INTPs T to N), but that only renders the system basic in that it scraps unfounded assertions from it; theoretically, a set of general truths about type can be built up in a more justified manner. As it stands, I can't see what it is that justifies, for instance, a person scoring as E and J and the information "You prefer T/F most" being returned...

I do think the lack of holistic engagement between preferences is a deficiency in Reynierse's model, and perhaps it would be preferable to discern some means of engaging across preferences - after all, we can prefer one thing to another in relation to things that aren't directly opposed as well as things that are, so why not also I/P, N/F, E/T questions and the like? That way, we could make more holistic determinations about a person's personality without needing to imply it on the basis of unsubstantiated assumptions. If the results seem not specific enough - if they become just a list of common traits, a picture of the average person like that rather than fitting really well - then I think that only speaks to the problem of trying to apply general truths about categories you're a member of to yourself. Just as I don't fit all the characteristics of the average male, or the average Australian, or the average PersonalityCafe member, so I don't fit all the characteristics of the average INTP. To the extent, then, that type dynamics tries to propose otherwise - which I would argue it does by mandating that a person with I, T and P preferences must prefer T to S/N, and things of that ilk - I would say it is as flawed as trying to make similarly universal claims about the other categories: unless there's some reason to suppose it necessarily must follow, we oughtn't to assume that it does.

- - - - -
This is a fair point to make. If a theory of cognitive functions is assumed then certain concepts such as INTPs valuing consistency would be in a way "a priori" truths, "INTP, by definition, values this mindset". I was actually wondering early on how these descriptions were created. In time it became clear that they were written as a sort of deduction or extrapolation of cognitive functions. I had a very empirical mind (which somewhat shifted over time) and thus I wondered about the descriptions. It was my opinion that you should make a test, then create a description. More like a hypothesis then experiment, rather than a deduction from assumed concepts. I think that would be better, but it sounds somewhat odd. One site that gives "best-fit-types" or at least mentions them, use first person descriptions to describe the general ideas of interviewed peoples of certain types. For example an NT type description may be something like "I was curious and wanted to figure out the fundamental truth behind reality" (that was just a general example I made up). So there would be these descriptions without specific reference to the functions, taken from self-understanding of tested subjects. There's nothing too profound about them, but they do fit a more object based understanding. In which case there is some room for a more testable MBTI, the descriptions may have to be based on general resultants of the tests (similar to the list of traits in Big Five), if there were to be descriptions at all. I think if there were to be descriptions there would have to be this external confirmation of interviews with "types" and a sort of justification based on the model to explain these correlations. It cannot be treated as a dichotomy, we may measure forces of physics in the real world, but that does not mean we cannot explain them or systemize them, noting there is a theory of relativity, rather than just a collection of data, a "wrapping paper".
My main point is that the full on "letter approach" doesn't seem to provide an adequate wrapping paper to me. I haven't come up with a "compromise" though. A version of the system that is both empirically testable but can also attempt to more holistically explain the personality, but that does not need to rely on "pure reason" that congintive functions could be called.
 

·
Registered
Joined
·
2,374 Posts
This is a fair point to make. If a theory of cognitive functions is assumed then certain concepts such as INTPs valuing consistency would be in a way "a priori" truths, "INTP, by definition, values this mindset". I was actually wondering early on how these descriptions were created. In time it became clear that they were written as a sort of deduction or extrapolation of cognitive functions. I had a very empirical mind (which somewhat shifted over time) and thus I wondered about the descriptions. It was my opinion that you should make a test, then create a description. More like a hypothesis then experiment, rather than a deduction from assumed concepts. I think that would be better, but it sounds somewhat odd. One site that gives "best-fit-types" or at least mentions them, use first person descriptions to describe the general ideas of interviewed peoples of certain types. For example an NT type description may be something like "I was curious and wanted to figure out the fundamental truth behind reality" (that was just a general example I made up). So there would be these descriptions without specific reference to the functions, taken from self-understanding of tested subjects. There's nothing too profound about them, but they do fit a more object based understanding. In which case there is some room for a more testable MBTI, the descriptions may have to be based on general resultants of the tests (similar to the list of traits in Big Five), if there were to be descriptions at all. I think if there were to be descriptions there would have to be this external confirmation of interviews with "types" and a sort of justification based on the model to explain these correlations. It cannot be treated as a dichotomy, we may measure forces of physics in the real world, but that does not mean we cannot explain them or systemize them, noting there is a theory of relativity, rather than just a collection of data, a "wrapping paper".
My main point is that the full on "letter approach" doesn't seem to provide an adequate wrapping paper to me. I haven't come up with a "compromise" though. A version of the system that is both empirically testable but can also attempt to more holistically explain the personality, but that does not need to rely on "pure reason" that congintive functions could be called.
Well, I think "INTP, by definition, values this mindset" is an unclear one. The theory asserts that I-N-T-P by definition makes Ti-Ne-S-Fe, but it's not clear the extent to which, say, Ti and desiring logical consistency is a necessary connection. And the problem there is obviously that there's no reason - inductive or deductive - for I-N-T-P to lead to Ti-Ne-S-Fe, but the lack of clarity as to the exact nature of the other information presented - what's said to be true a priori for each function vs. what is merely true by correlation, for example - is a challenge in itself.

So, in general, I'm not convinced that the cognitive functions really represent a "pure reason" approach, so much as they're not supported either empirically or rationally: if anything, they're notionally empirical, in that Jung's work is based upon what he "observed" in clients/patients. Pure reason would be fine if the assumptions we started with could themselves be justified... I would agree that stripping the theory back to solely the letters would leave us with an inadequate "wrapping paper", a series of connections between data that lacks anything more, but I also think that it at least gives us a solid, if skeletal, framework on which to try and build the more holistic theory, which isn't really possible in a theory built on unjustified assertions.

Descriptions sort of need to be general truths about a category by their very nature, unless any necessary truths can be reasoned out. So the main thing is that they ought to be clearer about that, so that people understand that they're reading about what tends to be true for people like them, rather than what necessarily is true for them individually. Again, it's sort of the same thing - stripping back the flawed aspects of the theory as it stands, leaving it perhaps a bit empty but giving a foundation on which more justified assertions could be used to flesh it out, rather than the unjustified ones that are currently employed.
 

·
Registered
Joined
·
1,958 Posts
The people who worked on Socionics did test their theory. That's the advantage of Socionics - it actually makes testable predictions, unlike MBTI.
 

·
Registered
Joined
·
3,491 Posts
@reckful - I can't claim to know exactly what it is Grant, Beebe and the like actually directly claim, but I would say your linked posts only disprove such models to the extent that their claim requires a function to operate in the same way as tertiary/inferior as you would expect it to as dominant/auxiliary. Certainly, if that's not their claim, then the onus is on them to argue how it remains the same function, and why it would manifest so differently, and things such as that, but it not being their claim would also make the disproof based on TJs being opposite to FPs rather than alike insufficient for the purpose of disproving it. I would say the bulk of studies that do show type dynamics to be flawed are those that engage directly with the MBTI system, since it is there where some evaluable assertions are actually made: it is they who expect T to be greater than N for all INTPs, whereas other functions theorists could easily dispute the MBTI testing mechanisms ability to get at their idea of the functions.
No, the disproof — by way of lack of validity, in over 50 years of MBTI data pools — certainly doesn't hinge on any kind of oversimple expectation that a so-called "cognitive function" (e.g., "Si") will "operate in the same way" when it's in the dominant position (for an ISJ) as it does when it's in, say, the tertiary position (for an INP). But assuming there's anything at all to the idea that SJs and NPs are "Si/Ne types" that share certain significant personality characteristics that they don't share with SPs and NJs (the "Se/Ni types"), then whenever you have a study where what's being correlated with type is an "Si" thing, you should expect to find the SJs and NPs on the same side and the SPs and NJs on the other side. If the way Si manifests in INPs (with Si tertiary) means that INPs show significantly less relevant Si influence than in ISJs (with Si dominant), but still more than in the Se/Ni types, then you'd expect to find the introverted types lined up along the relevant spectrum like this:

INJs----ISPs---|---INPs----ISJs

Or maybe the ISPs would be at the far left with the INJs to their right (but still on the left side of the spectrum), or maybe they'd both be more toward the middle of the left side. Supposedly neither one is an "Si type" at all, right? — unless you're talking about some kind of Beebean "shadow" manifestation.

In any case, the point is that, whichever of those possible Harold-Grantian patterns you might be looking for, they virtually never show up. Instead, if the SJs are out at one end of the spectrum, you can reliably expect to find the NPs out at the other end.

In the first spoiler are membership stats for Personality Cafe and Typology Central. For each type, the first percentage is the percentage of that type at the forum, the second percentage (in parentheses) is the estimated "general population" percentage from the official MBTI folks (from this page), and the final number on the right is the self-selection ratio for that type — i.e., the ratio of the forum percentage to the general population percentage.

 
November 2014 membership stats for Personality Cafe:

INFJ — 9133 — 15.7% (1.5%) — ssr: 10.5
INTJ — 7307 — 12.6% (2.1%) — ssr: 6.0
INFP — 11865 — 20.4% (4.4%) — ssr: 4.6
INTP — 7825 — 13.5% (3.3%) — ssr: 4.1
ENTP — 3709 — 6.4% (3.2%) — ssr: 2.0
ENTJ — 1681 — 2.9% (1.8%) — ssr: 1.6
ENFJ — 1904 — 3.3% (2.5%) — ssr: 1.3
ENFP — 4915 — 8.5% (8.1%) — ssr: 1.0
ISTP — 1926 — 3.3% (5.4%) — ssr: 0.6
ISFP — 1986 — 3.4% (8.8%) — ssr: 0.4
ISTJ — 2094 — 3.6% (11.6%) — ssr: 0.3
ESTP — 635 — 1.1% (4.3%) — ssr: 0.3
ISFJ — 1374 — 2.4% (13.8%) — ssr: 0.2
ESFP — 620 — 1.1% (8.5%) — ssr: 0.1
ESFJ — 573 — 1.0% (12.3%) — ssr: 0.1
ESTJ — 542 — 0.9% (8.7%) — ssr: 0.1

November 2014 membership stats for Typology Central:

INFJ — 1782 — 16.1% (1.5%) — ssr: 10.7
INTJ — 1437 — 13.0% (2.1%) — ssr: 6.2
INTP — 1958 — 17.7% (3.3%) — ssr: 5.4
INFP — 2016 — 18.2% (4.4%) — ssr: 4.1
ENTP — 781 — 7.0% (3.2%) — ssr: 2.2
ENTJ — 298 — 2.7% (1.8%) — ssr: 1.5
ENFP — 1156 — 10.4% (8.1%) — ssr: 1.3
ENFJ — 321 — 2.9% (2.5%) — ssr: 1.2
ISTP — 304 — 2.7% (5.4%) — ssr: 0.5
ISFP — 256 — 2.3% (8.8%) — ssr: 0.3
ISTJ — 278 — 2.5% (11.6%) — ssr: 0.2
ESTP — 100 — 0.9% (4.3%) — ssr: 0.2
ISFJ — 181 — 1.6% (13.8%) — ssr: 0.1
ESFP — 84 — 0.8% (8.5%) — ssr: 0.1
ESTJ — 74 — 0.7% (8.7%) — ssr: 0.1
ESFJ — 65 — 0.6% (12.3%) — ssr: 0.05

You couldn't ask for a much better set of results, from the standpoint of the validity of the MBTI dichotomies, than those forum type frequencies. The MBTI doesn't ask people if they use internet forums (or the internet), and it doesn't ask them if they're interested in personality (or psychology). But the type frequency pattern at both forums (involving relatively large samples) 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).

But from the standpoint of the functions? The top four types are the INs, and they make up 60% of Forum A's members and 62% of Forum B's members (but less than 12% of the general population), and they're a tidy foursome from a dichotomy-centric perspective but they have three different dominant functions (Ni, Ti and Fi). Good luck coming up with any function-based set of influences on MBTI forum membership that does nearly as good a job of explaining the frequencies as the obvious dichotomy-based explanation.

In the second spoiler are the self-selection ratios that Myers reported for a study involving 705 Cal Tech science students:

 
INTJ 3.88
INFJ 2.95
INTP 2.92
INFP 1.97
ENTJ 1.56
ENTP 1.42
ENFP 1.09
ENFJ 1.08
ISTJ 0.68
ISTP 0.50
ISFP 0.49
ISFJ 0.43
ESTP 0.22
ESTJ 0.12
ESFJ 0.18
ESFP 0.02

Stat spectrums that tidy are what you call a typologist's dream. What they indicate (and the sample size was pretty large, at 705) is that, like personality forum membership (but this time with offically-determined types), the MBTI factor that has the greatest influence on somebody's tendency to become a Cal Tech science major is an N preference, and the MBTI factor that has the second greatest influence is introversion, with the result that the spectrum tidily lines up (from bottom to top) ES-IS-EN-IN.

And the third spoiler has the self-selection ratios from a study of 5,700 gifted adolescents that serves as another example of the way the types line up in cases where E/I and S/N are the main MBTI-related influences on the characteristic(s) being correlated with type.

 
INTP 3.4
INTJ 2.87
INFP 2.68
INFJ 2.67
ENTP 2.32
ENFP 2.03
ENTJ 1.49
ENFJ 1.26
ISTJ 0.99
ISTP 0.78
ESTP 0.49
ISFJ 0.40
ISFP 0.40
ESFP 0.28
ESTJ 0.26
ESFJ 0.24

That's how validity gets established for typological categories in the scientifically respectable districts of personality psychology. And those are the kinds of patterns that have steadfastly failed to show up — for 50 years and counting — for the function- and tandem-based type categories that correspond to the Harold Grant function stack. I am here to tell you that I will be very surprised if any forum Grant-stack aficionado can point me to any set of data where — because Fi vs. Fe (and/or Ti vs. Te, and/or Ni vs. Ne, and/or Si vs. Se, and/or Fi/Te vs. Ti/Fe, etc.) is purportedly the relevant personality component coming into play — the TJs and FPs are more-or-less sharing one half of the spectrum and the TPs and FJs are sharing the other half (or the NJs and SPs are more-or-less sharing one half of the spectrum and the NPs and SJs are sharing the other half). Instead, if the data pool is one where INTJs are at or near one end of the spectrum, you can pretty reliably expect to find the ESFPs at or near the other end, because that's the way the dichotomy combinations play out, regardless of which dichotomies are the influential ones, and regardless of whether they comprise a combination that's supposedly function-related.

INs at one end? ESs at the other.
SJs at one end? NPs at the other.
TJs at one end? FPs at the other.

Alas for Harold Grant, and faithful followers like Linda Berens and Dario Nardi, that's just the way the real MBTI dimensions virtually always seem to play out.

That Reynierse article ("The Case Against Type Dynamics") that I'm always linking to (and that's one of the articles discussed in the OP link) is one of a series of articles that Reynierse (and Harker) published in the Journal of Psychological Type (which is published by the official MBTI folks) and that strongly argued against the Harold Grant model and scoffed at its lack of validity, as well as taking a sharp swipe or two at Naomi Quenk for her ongoing support of "type dynamics." And Quenk is about as "establishment MBTI" as you can get, having authored or co-authored lots of official MBTI materials, including the Step II Manual. Outside her "official" MBTI work, as Reynierse pointed out, she's been a pretty big cognitive functions person (although she's remained agnostic on the issue of the tertiary's attitude), and you'd better believe she's someone who has ready access to the vast trove of MBTI data that's been gathered over the last 50 years.

Reynierse's articles caused quite a stir in the MBTI community, as I understand it. And all Naomi Quenk or Linda Berens or Dario Nardi or any of those other cognitive functions people needed to do to refute his assertion that the functions are just a "category mistake" — not to mention provide, at long last, some respectable support for the functions — was to go through the vast stores of existing MBTI data and find a respectable body of results reflecting one of those patterns (TJs/FPs on one side and TPs/FJs on the other, or SJs/NPs on one side and SPs/NJs on the other). Because if either of those patterns — which are decidedly inconsistent with what Reynierse calls "preference multidimensionality" (i.e., the simple additive effects of the four preferences) — ever turned up in a respectable body of MBTI data, well, that's what validity is all about.

And instead, as I understand it, the response to Reynierse (as far as the validity issue goes) was... *crickets*

And at this point, given the overwhelming lack of evidentiary support, I'd say the aspect of the Harold Grant functions model that says that an INTJ has "tertiary Fi" (i.e., has MBTI-related aspects of personality that they share with INFPs, and that INFJs don't share with INFPs), and that an INFP has "tertiary Si" (i.e., has MBTI-related aspects of personality that they share with ISTJs, and that ISTPs don't share with ISTJs) is past the point of being able to respectably claim "not yet proven" status — and instead, as I already said, should really be considered disproven by anyone who likes to think of themselves as reality-oriented.
 

·
Registered
Joined
·
2,374 Posts
No, the disproof — by way of lack of validity, in over 50 years of MBTI data pools — certainly doesn't hinge on any kind of oversimple expectation that a so-called "cognitive function" (e.g., "Si") will "operate in the same way" when it's in the dominant position (for an ISJ) as it does when it's in, say, the tertiary position (for an INP). But assuming there's anything at all to the idea that SJs and NPs are "Si/Ne types" that share certain significant personality characteristics that they don't share with SPs and NJs (the "Se/Ni types"), then whenever you have a study where what's being correlated with type is an "Si" thing, you should expect to find the SJs and NPs on the same side and the SPs and NJs on the other side. If the way Si manifests in INPs (with Si tertiary) means that INPs show significantly less relevant Si influence than in ISJs (with Si dominant), but still more than in the Se/Ni types, then you'd expect to find the introverted types lined up along the relevant spectrum like this:

INJs----ISPs---|---INPs----ISJs

Or maybe the ISPs would be at the far left with the INJs to their right (but still on the left side of the spectrum), or maybe they'd both be more toward the middle of the left side. Supposedly neither one is an "Si type" at all, right? — unless you're talking about some kind of Beebean "shadow" manifestation.

In any case, the point is that, whichever of those possible Harold-Grantian patterns you might be looking for, they virtually never show up. Instead, if the SJs are out at one end of the spectrum, you can reliably expect to find the NPs out at the other end.
None of this refutes my point. For that conclusion to hold, you have to assume that it is claimed that the feature you're looking for only correlates with Si - I suspect many functions theorists would claim otherwise, and say that, for instance, the characteristic is partially shared by other introverted functions, partially shared by the other sensing function, &c. In any case, many would say that what it is that associates with INP types' tertiary Si shares only the characteristic of being Si - of being "subjective perception of concrete information" or some other such definition. The evidence you point to only holds as an argument against those who would claim that characteristics align particularly with Si such that SJs evince it most, then NPs, then others: if the claim is different to that, as I would say the claims of many who adhere to the functions theories are, then the argument doesn't actually refute those claims.

That really is the crux of the point I'm making - evidence such as what you're presenting is only a disproof to those claims which it actually contradicts. A theory which asserts that, say, "Si is different in NPs than in SJs, such that, although it is the same function, it manifests totally differently, the only commonality between them is that their perception of concrete information is subjective in character" has a vastly different problem to address - namely, why bother focusing on that commonality if it has no wider implications? - but the salient point here is that such an argument would in no way be "disproven" by the argument you're presenting.

In the first spoiler are membership stats for Personality Cafe and Typology Central. For each type, the first percentage is the percentage of that type at the forum, the second percentage (in parentheses) is the estimated "general population" percentage from the official MBTI folks (from this page), and the final number on the right is the self-selection ratio for that type — i.e., the ratio of the forum percentage to the general population percentage.

 
November 2014 membership stats for Personality Cafe:

INFJ — 9133 — 15.7% (1.5%) — ssr: 10.5
INTJ — 7307 — 12.6% (2.1%) — ssr: 6.0
INFP — 11865 — 20.4% (4.4%) — ssr: 4.6
INTP — 7825 — 13.5% (3.3%) — ssr: 4.1
ENTP — 3709 — 6.4% (3.2%) — ssr: 2.0
ENTJ — 1681 — 2.9% (1.8%) — ssr: 1.6
ENFJ — 1904 — 3.3% (2.5%) — ssr: 1.3
ENFP — 4915 — 8.5% (8.1%) — ssr: 1.0
ISTP — 1926 — 3.3% (5.4%) — ssr: 0.6
ISFP — 1986 — 3.4% (8.8%) — ssr: 0.4
ISTJ — 2094 — 3.6% (11.6%) — ssr: 0.3
ESTP — 635 — 1.1% (4.3%) — ssr: 0.3
ISFJ — 1374 — 2.4% (13.8%) — ssr: 0.2
ESFP — 620 — 1.1% (8.5%) — ssr: 0.1
ESFJ — 573 — 1.0% (12.3%) — ssr: 0.1
ESTJ — 542 — 0.9% (8.7%) — ssr: 0.1

November 2014 membership stats for Typology Central:

INFJ — 1782 — 16.1% (1.5%) — ssr: 10.7
INTJ — 1437 — 13.0% (2.1%) — ssr: 6.2
INTP — 1958 — 17.7% (3.3%) — ssr: 5.4
INFP — 2016 — 18.2% (4.4%) — ssr: 4.1
ENTP — 781 — 7.0% (3.2%) — ssr: 2.2
ENTJ — 298 — 2.7% (1.8%) — ssr: 1.5
ENFP — 1156 — 10.4% (8.1%) — ssr: 1.3
ENFJ — 321 — 2.9% (2.5%) — ssr: 1.2
ISTP — 304 — 2.7% (5.4%) — ssr: 0.5
ISFP — 256 — 2.3% (8.8%) — ssr: 0.3
ISTJ — 278 — 2.5% (11.6%) — ssr: 0.2
ESTP — 100 — 0.9% (4.3%) — ssr: 0.2
ISFJ — 181 — 1.6% (13.8%) — ssr: 0.1
ESFP — 84 — 0.8% (8.5%) — ssr: 0.1
ESTJ — 74 — 0.7% (8.7%) — ssr: 0.1
ESFJ — 65 — 0.6% (12.3%) — ssr: 0.05

You couldn't ask for a much better set of results, from the standpoint of the validity of the MBTI dichotomies, than those forum type frequencies. The MBTI doesn't ask people if they use internet forums (or the internet), and it doesn't ask them if they're interested in personality (or psychology). But the type frequency pattern at both forums (involving relatively large samples) 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).

But from the standpoint of the functions? The top four types are the INs, and they make up 60% of Forum A's members and 62% of Forum B's members (but less than 12% of the general population), and they're a tidy foursome from a dichotomy-centric perspective but they have three different dominant functions (Ni, Ti and Fi). Good luck coming up with any function-based set of influences on MBTI forum membership that does nearly as good a job of explaining the frequencies as the obvious dichotomy-based explanation.

In the second spoiler are the self-selection ratios that Myers reported for a study involving 705 Cal Tech science students:

 
INTJ 3.88
INFJ 2.95
INTP 2.92
INFP 1.97
ENTJ 1.56
ENTP 1.42
ENFP 1.09
ENFJ 1.08
ISTJ 0.68
ISTP 0.50
ISFP 0.49
ISFJ 0.43
ESTP 0.22
ESTJ 0.12
ESFJ 0.18
ESFP 0.02

Stat spectrums that tidy are what you call a typologist's dream. What they indicate (and the sample size was pretty large, at 705) is that, like personality forum membership (but this time with offically-determined types), the MBTI factor that has the greatest influence on somebody's tendency to become a Cal Tech science major is an N preference, and the MBTI factor that has the second greatest influence is introversion, with the result that the spectrum tidily lines up (from bottom to top) ES-IS-EN-IN.

And the third spoiler has the self-selection ratios from a study of 5,700 gifted adolescents that serves as another example of the way the types line up in cases where E/I and S/N are the main MBTI-related influences on the characteristic(s) being correlated with type.

 
INTP 3.4
INTJ 2.87
INFP 2.68
INFJ 2.67
ENTP 2.32
ENFP 2.03
ENTJ 1.49
ENFJ 1.26
ISTJ 0.99
ISTP 0.78
ESTP 0.49
ISFJ 0.40
ISFP 0.40
ESFP 0.28
ESTJ 0.26
ESFJ 0.24

That's how validity gets established for typological categories in the scientifically respectable districts of personality psychology. And those are the kinds of patterns that have steadfastly failed to show up — for 50 years and counting — for the function- and tandem-based type categories that correspond to the Harold Grant function stack. I am here to tell you that I will be very surprised if any forum Grant-stack aficionado can point me to any set of data where — because Fi vs. Fe (and/or Ti vs. Te, and/or Ni vs. Ne, and/or Si vs. Se, and/or Fi/Te vs. Ti/Fe, etc.) is purportedly the relevant personality component coming into play — the TJs and FPs are more-or-less sharing one half of the spectrum and the TPs and FJs are sharing the other half (or the NJs and SPs are more-or-less sharing one half of the spectrum and the NPs and SJs are sharing the other half). Instead, if the data pool is one where INTJs are at or near one end of the spectrum, you can pretty reliably expect to find the ESFPs at or near the other end, because that's the way the dichotomy combinations play out, regardless of which dichotomies are the influential ones.

INs at one end? ESs at the other.
SJs at one end? NPs at the other.
TJs at one end? FPs at the other.

Alas for Harold Grant, and faithful followers like Linda Berens and Dario Nardi, that's just the way the real MBTI dimensions virtually always seem to play out.
I'm not so sure the personality forum statistics can be granted too much weight, since it's clear that people are working from a variety of different personality theories to arrive at their type. There's no consistent means of evaluation between them - one "INFP" might have just done a test and listed themselves as INFP, another has looked into Beebe and decided that they're "sure" Ni is their "senex function" and Fi their "hero/heroine", another has taken the MBTI Step II and revised their reported type from ISFP to INFP because the description seemed more accurate to them and they "know" they're more Original and Imaginative than they tested as, another did a "type me" on here and got told they were an INFP... quite plainly, there's nothing consistent about that such that you could say the data is solid enough to work with.

With that said, the other sources don't have that same problem, and you're right in that they display quite clear trends towards particular dichotomy preferences. But what this tells us is that dichotomies prove better predictors of these characteristics; again, the extent to which this shows up something actually incorrect in the functions theory, as opposed to merely something unsupported, is only to the extent that the functions theory in question actually asserts that you can expect to find TJs/FPs or whatever sharing said common traits. Absolutely, functions theories that can't substantiate their claims don't have scientific validity, but to state that this is a disproof of them is to overstate what the evidence actually demonstrates.

That Reynierse article ("The Case Against Type Dynamics") that I'm always linking to (and that's one of the articles discussed in the OP link) is one of a series of articles that Reynierse (and Harker) published in the Journal of Psychological Type (which is published by the official MBTI folks) and that strongly argued against the Harold Grant model and scoffed at its lack of validity, as well as taking a sharp swipe or two at Naomi Quenk for her ongoing support of "type dynamics." And Quenk is about as "establishment MBTI" as you can get, having authored or co-authored lots of official MBTI materials, including the Step II Manual. Outside her "official" MBTI work, as Reynierse pointed out, she's been a pretty big cognitive functions person (although she's remained agnostic on the issue of the tertiary's attitude), and you'd better believe she's someone who has ready access to the vast trove of MBTI data that's been gathered over the last 50 years.

Reynierse's articles caused quite a stir in the MBTI community, as I understand it. And all Naomi Quenk or Linda Berens or Dario Nardi or any of those other cognitive functions people needed to do to refute his assertion that the functions are just a "category mistake" — not to mention provide, at long last, some respectable support for the functions — was to go through the vast stores of existing MBTI data and find a respectable body of results reflecting one of those patterns (TJs/FPs on one side and TPs/FJs on the other, or SJs/NPs on one side and SPs/NJs on the other). Because if either of those patterns — which are decidedly inconsistent with what Reynierse calls "preference multidimensionality" (i.e., the simple additive effects of the four preferences) — ever turned up in a respectable body of MBTI data, well, that's what validity is all about.

And instead, as I understand it, the response to Reynierse (as far as the validity issue goes) was... *crickets*

And at this point, given the overwhelming lack of evidentiary support, I'd say the aspect of the Harold Grant functions model that says that an INTJ has "tertiary Fi" (i.e., has MBTI-related aspects of personality that they share with INFPs, and that INFJs don't share with INFPs), and that an INFP has "tertiary Si" (i.e., has MBTI-related aspects of personality that they share with ISTJs, and that ISTPs don't share with ISTJs) is past the point of being able to respectably claim "not yet proven" status, and should really be considered disproven by anyone who likes to think of themselves as reality-oriented.
I certainly think it's incumbent upon people directly involved with the MBTI, or with functions theories generally, to make claims that can be substantiated and tested for; to the extent that the Reynierse articles, and indeed other articles in the Journal of Psychological Type and elsewhere, make valid criticisms that impact upon any given theory, the onus is on those who hold to those theories either to renounce them, or to improve them such that the criticism no longer applies, or to demonstrate why the criticism is wrong. I'd say a number of functions theorists don't claim scientific validity, so the lack of it is thus no detriment to their claim.

It may be worth linking to another pair of articles from the JPT, by McPeek and Martin in 2012, assessing the validity of the XYYY function stack (in two parts, here and here). Like Reynierse and Harker, their testing lent scant support to the functions theory they were seeking to evaluate. But, as they point out towards the end of the second article, in the section headed "Limitations of the Study":
McPeek and Martin said:
There is always the possibility that type theory is correct but that type measurement is inaccurate. [...] a conservative conclusion is that the current study illuminates two possible ways to explain MBTI results. The first, more parsimonious, is to work from a preference-based framework. The other is the traditional FA model - which in and of itself is not disproven by this study.
[emphasis mine]

The lack of evidence for functions theorists means that a prudent scientist would have no reason to accept the model - that much, I think, holds. But to claim that said theories have been actively disproven would be unreasonable. The MBTI, and many other functions theorists, have not been sufficiently clear about their hypotheses such that these tests could be said to show said hypotheses false; that we have no reason to suppose them true should not be, in and of itself, reason to conclude that they are not. I would hardly say that's not "reality-oriented" - indeed, I think it acknowledges the reality of the situation better than the approach you're proposing, that a lack of evidence for doesn't necessarily constitute evidence against. Unproven, instead, but not disproven.
 

·
Registered
Joined
·
3,491 Posts
That really is the crux of the point I'm making - evidence such as what you're presenting is only a disproof to those claims which it actually contradicts. A theory which asserts that, say, "Si is different in NPs than in SJs, such that, although it is the same function, it manifests totally differently, the only commonality between them is that their perception of concrete information is subjective in character" has a vastly different problem to address - namely, why bother focusing on that commonality if it has no wider implications? - but the salient point here is that such an argument would in no way be "disproven" by the argument you're presenting. ...

I certainly think it's incumbent upon people directly involved with the MBTI, or with functions theories generally, to make claims that can be substantiated and tested for; to the extent that the Reynierse articles, and indeed other articles in the Journal of Psychological Type and elsewhere, make valid criticisms that impact upon any given theory, the onus is on those who hold to those theories either to renounce them, or to improve them such that the criticism no longer applies, or to demonstrate why the criticism is wrong. I'd say a number of functions theorists don't claim scientific validity, so the lack of it is thus no detriment to their claim.

It may be worth linking to another pair of articles from the JPT, by McPeek and Martin in 2012, assessing the validity of the XYYY function stack (in two parts, here and here). Like Reynierse and Harker, their testing lent scant support to the functions theory they were seeking to evaluate. But, as they point out towards the end of the second article, in the section headed "Limitations of the Study":

[emphasis mine]

The lack of evidence for functions theorists means that a prudent scientist would have no reason to accept the model - that much, I think, holds. But to claim that said theories have been actively disproven would be unreasonable. The MBTI, and many other functions theorists, have not been sufficiently clear about their hypotheses such that these tests could be said to show said hypotheses false; that we have no reason to suppose them true should not be, in and of itself, reason to conclude that they are not. I would hardly say that's not "reality-oriented" - indeed, I think it acknowledges the reality of the situation better than the approach you're proposing, that a lack of evidence for doesn't necessarily constitute evidence against. Unproven, instead, but not disproven.
Well, I'd say we're somewhat talking past each other. To the extent that what you're saying is essentially that no lack of correlational patterns involving personality characteristics can really be viewed as disproving functions if "cognitive functions" are characterized as things so inchoate and subtle that they really can't be expected to have any impact on somebody's, you know, personality, then it's hard to disagree with that. But that certainly wasn't Jung's perspective. He went on and on and on about the noticeable and often dramatic impacts that the functions (including in different positions) had on people's personalities and behavior.

Similarly, it's very much typical for discussions of "Fe types" (or "Fe/Ti types") and "Fi types" (or "Fi/Te types"), for example — both in sources like Thomson, Berens and Nardi and in forum threads — to focus on recognizable and noteworthy personality characteristics that those types purportedly share.

So in other words, my point is that if the cognitive functions were really associated with the wide gamut of personality characteristics that Thomson, Berens and Nardi ascribe to them, and if INTJs and ESFPs were both "Ni/Se types" and "Te/Fi types" in the way that they're said to be using the Grant function stack, the existing 50 years of data pools — correlating officially-measured MBTI types with everything from interests to occupations to other personality measures to you-name-it — would surely include at least some respectable body of samples where, because the specific thing being correlated with type was substantially affected by Ni, or Te, or Ni/Se, or whatever, the INTJs and ESFPs were on one side of the divide and the INTPs and ESFJs were on the other. (And they don't.)

And if, on the other hand, the Grant function stack referred to cognitive stuff so deeply buried and variable that its impact on MBTI types' personalities didn't really manifest in those kinds of ways, then I agree with you that the relevant theorists could rightly be accused of having failed to meet their scientific (albeit soft-scientific) burden to "make claims that can be substantiated and tested for."

And I'm happy to agree to disagree on whether the Grant stack, at this point, is better viewed as in the "no respectable evidence yet" category or the "effectively disproven" category, as long as you agree (and I think you do) that the burden is really on the Grantians to prove their functions/stack (if they want any serious attention) rather than on Reynierse or anyone else to somehow conclusively disprove it.
 

·
Registered
Joined
·
2,374 Posts
Well, I'd say we're somewhat talking past each other. To the extent that what you're saying is essentially that no lack of correlational patterns involving personality characteristics can really be viewed as disproving functions if "cognitive functions" are characterized as things so inchoate and subtle that they really can't be expected to have any impact on somebody's, you know, personality, then it's hard to disagree with that. But that certainly wasn't Jung's perspective. He went on and on and on about the noticeable and often dramatic impacts that the functions (including in different positions) had on people's personalities and behavior.

Similarly, it's very much typical for discussions of "Fe types" (or "Fe/Ti types") and "Fi types" (or "Fi/Te types"), for example — both in sources like Thomson, Berens and Nardi and in forum threads — to focus on recognizable and noteworthy personality characteristics that those types purportedly share.

So in other words, my point is that if the cognitive functions were really associated with the wide gamut of personality characteristics that Thomson, Berens and Nardi ascribe to them, and if INTJs and ESFPs were both "Ni/Se types" and "Te/Fi types" in the way that they're said to be using the Grant function stack, the existing 50 years of data pools — correlating officially-measured MBTI types with everything from interests to occupations to other personality measures to you-name-it — would surely include at least some respectable body of samples where, because the specific thing being correlated with type was substantially affected by Ni, or Te, or Ni/Se, or whatever, the INTJs and ESFPs were on one side of the divide and the INTPs and ESFJs were on the other. (And they don't.)

And if, on the other hand, the Grant function stack referred to cognitive stuff so deeply buried and variable that its impact on MBTI types' personalities didn't really manifest in those kinds of ways, then I agree with you that the relevant theorists could rightly be accused of having failed to meet their scientific (albeit soft-scientific) burden to "make claims that can be substantiated and tested for."

And I'm happy to agree to disagree on whether the Grant stack, at this point, is better viewed as in the "no respectable evidence yet" category or the "effectively disproven" category, as long as you agree (and I think you do) that the burden is really on the Grantians to prove their functions/stack (if they want any serious attention) rather than on Reynierse or anyone else to somehow conclusively disprove it.
My point is something like that - that a good many presentations of cognitive functions theory make their claims too unclear for the evidence to actually directly refute them. And I would agree that, to the extent theorists claim things like "Fi/Te types do this, and Fe/Ti types do that", and "this/that" are evaluable things, the evidence against such claims is clear enough. Certainly, the burden of proof is on the person asserting the theory, and there is no real burden to disprove - but I think the difference between something directly shown to be false and something lacking in evidence for its truth is an important one that needs to be maintained.
 

·
Registered
Joined
·
3,376 Posts
I dunno, I'm not seeing anything particularly new in the Reynierse model. Maybe I'm too much of a crotchety old man in the MBTI world.

I'm obviously on board with looking at this stuff in new ways, but what I read from the link seems to be an alternate way of describing what many of us have already found ways to describe using tools ranging from MBTI to Cognitive Functions to Enneagram.

Hell, the main reason I got interested in the functions in the first place was to get past this archaic, misleading nonsense about preference for "thinking" or "feeling" and "intuitive" or "sensing," and so on. It's all testable, but as Grandmaster Yoda touched on, it largely doesn't tell us shit.

As far as I'm concerned, we're still in the dark ages of personality theory when we have to resort to phrases like "sees the big picture" to describe intuition and "gets drained by social interaction" to describe introversion. And that seems to be where we're usually stuck when people don't want to use the cognitive functions. I understand why not, don't get me wrong.

But I don't see any getting past the fact that with our current limitations in exploring cognition, there's going to have to be some leaps of reasoning if we're to get at complexity at all. The important thing when doing this is to not lose sight of the fact that the leaps are for the purpose of exploring the mind and the personality, much like the hammer is for the purpose of hitting a nail into a wall, and are not to be treated with the respect that empirical science gets simply because they sound cool.

The old models should be challenged. It's the only way for this shit to grow. But I feel like there's something arrogant in revering the testability of preferences like thinking and feeling. The tests themselves seem to be nothing without the personality descriptions. And where do we get the descriptions? From the test results? From observing people?

How do we know whether somebody has a preference for thinking or feeling based on observation? What the hell are thinking and feeling?

The whole thing lacks credibility at its core. There's a reason psychology has a DSM manual thingy that changes over time - because nobody knows what the fuck they're doing, when it really comes down to it. Psychology needs neuroscience badly. Very badly.
 

·
Registered
Joined
·
3,330 Posts
Hell, the main reason I got interested in the functions in the first place was to get past this archaic, misleading nonsense about preference for "thinking" or "feeling" and "intuitive" or "sensing," and so on. It's all testable, but as Grandmaster Yoda touched on, it largely doesn't tell us shit.
@LostFavor @Grandmaster Yoda

I disagree. But why do you think that?
 

·
Registered
Joined
·
3,376 Posts
@LostFavor @Grandmaster Yoda

I disagree. But why do you think that?
I don't know if I can give you an extended answer, considering that it's 4:00 in the morning.

The brief answer is: Because it's too damn simple at every level, yet it's trying to encapsulate something complex.
 
1 - 20 of 41 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top