Fe dominant types and inner emptiness -- Interviews Helia

I conducted this interview of Helia:



I asked Helia to do this interview with me to explore themes of inner emptiness, inner change, and variable sense of self. Before interviewing Helia, I already understood, or at least strongly suspected, that she was a EIE, as I typed her in this interview. And she also knew that I wanted to explore these themes in the interview.

But even so, I don't believe the rapidity at which her self concept became a discussion about the variability and indescribability of self, is doctored or really very prompted. In response to the very first "Tell me about yourself question" at 1:40, Helia struggles to describe herself, saying "I don't think there really is a true self, I think I'm a combination of actions, history choices." And 2:35 "It's hard to describe myself with a few words, because I'm a lot of things."

The whole interview showcases the themes of conscious transience and self confusion that often describe the Fe dominant types and especially so the EIE that is more divulging of the entire range of their emotional world, positive and negative both, than the ESE tends to display.

My description of the EIE has drawn some attention for its claims that EIEs sometimes see themselves as empty inside, or indescribable, or having a unclear sense of themselves. Nowhere does the description say that all EIEs have an unclear sense of who they are, but it does say that this theme presents itself in cases with moderate frequency. This claim has attracted some attention over the years, with some arguing that this is not so, that EIEs are not empty or confused about themselves inside, not as a typical theme. The best way to advance the discussion, then, is to provide casework and examples.

Let's be clearer about what Helia does and doesn't say in her self description (to the extent that we can). Helia doesn't describe herself as empty inside, generally explicitly rejecting the idea "No [I don't see myself as empty inside]" (9:20),  "It's complicated; I'm not empty, but I'm not full" (10:25), and instead describing herself as full of contents inside, rather than empty. But, she does say that in the past, she has been empty (14:00). And she explains herself extensively, definitely struggling to describe herself in any consistent ways, emphasizing her transience of character, and emphasizing some confusion and clarity about who she is inside "at times I have an unclear sense of self" (6:25), "I view myself almost through an external lens" (7:00). Generally, (my summary), she is emphasizing that the process of self-discovery is highly variable depending on what we might experience.

Other EIEs have sometimes assented because of a similar process of transient emotional states that they feel empty inside. In some cases they shared their assent in interviews although for reasons of privacy and also, sadly, some technical issues with my recordings that failed to process correctly, I am not able to share all of my interviews. The whole interview is worth listening to to explore the theme, although the first 16 minutes are the most topical. 

As an aside, since I noticed it while reviewing timestamps here, we can see the somewhat classical way Helia is ideologically pushy, in the case about the rather abstract condition of human fullness, around (12:00). The channelling and depth of fervor to her causes is also apparent at (28:00). Helia shows off her vibrancy and intentional control over her emotionality at 39:40. I didn't make this post to show off every confirmatory detail of the interview as it went along. This type is pretty obvious, my usual readers probably read her right away.

Model L is beyond trash

I was asked, or challenged, or both, to write criticisms of Model L in a formal way to reflect the record and perhaps, in lieu of bitching loudly about it on social media. A sensible or at least understandable request, although not one I relished to accept, for several reasons. One important reason is because the creator of Model L, Kimani White, has proven to be quite a belligerent asshole, who vainly seems to expect recognition for his accomplishment, and has shown nothing but hostility and contempt towards others (not only myself) who are not quick to kiss his feet and who do not care about his extremely vapid dichotomy structure (I wish to add that Model L's former co-creator, Aleesha Lowry, has absolutely not engaged in any of this vanity or bullying at all, she is an old friend of mine and not an asshole). As such, I have some reservations about even bringing attention to Model L, although likely towards whatever audience reads this blog I am preaching to the choir. But probably an even more significant reason not to bother with this post is because Model L is overwhelmingly dearth of descriptive content, and there is very little of substance to actually respond to, and what substance does exist will be exceedingly technical.

Anyway, since I'm here with questionable judgment writing this regardless, let's start by discussing what Model L is and what it isn't.

Model L is a 16 element model, whose elements (which I will call L-elements, and whose function I will call L-functions) are the combination of an IM element and a jungian axis. For example, the leading function of the LSE in Model L, is Te(S), or Te colored by the ST club, or alternately, Te+Si+Se. Therefore, unlike my block models which have blocks as their elements, Model L is not a block model since L-elements are not blocks, since a block consists of one rational element and one irrational element, and clearly Te+Si+Se does not consist of one rational and one irrational element. In fact, L-elements consist of two ordered blocks, not one. Each of the 16 socionics types has exactly one unique base function in Model L. As such, Model L is homomorphic to the set of socionics types and to other 16 element models such as Model A16, but does not emphasize in its design the isomorphism between the model and the set of intertype relations. More on this a bit later.

Because L-elements contain two blocks and not one, and because these blocks are from adjacent quadras, (e.g. Te+Si is a delta block, and Te+Se is a gamma block), each L-element does not belong to any one quadra. The L-elements are isomorphic to types, and these types each belong to exactly one quadra, but the L-elements considered by themselves do not belong to any one quadra.

Why was the choice made to make L-elements into two blocks and not one? I don't know, but this structure emphasizes clubs and dimensionality, and it seems likely to have been invented by someone who was primarily concerned with clubs and dimensionality, and who sees the creative and demonstrative functions as type defining, in other words, the type defining blocks are 1+2 and 1+8, instead of 1+2 and 1+6 as they are in the theory of quadras.

Therefore, by emphasizing clubs and dimensionality, and by including no quadra structure in the basic formulation of the model," Model L is fundamentally incompatible with Western Socionics. Recall that Western Socionics is the idea that socionics is largely, although not necessarily entirely, a theory of quadras. Model L has more in common with Model G and other eastern systems of Reinin dichotomies than it does with Western Socionics. Both Model L and Model G emphasize the 1+2 and 1+8 blocks as the type defining blocks, unlike in Western Socionics, and both of them rely heavily on Reinin dichotomies for their formulation and analysis, which are incompatible with quadras.

Kimani's long, dense, non-descriptive document entitled "Socionics Tweaks" has a section describing quadra priority from the presence cube, but it does not very clearly integrate priority into the model. For example, the "ordinals" in Model L are grouped (bizarrely) based on dimensionality and into clubs, such that the ego and id blocks are paired together in an orbital, the super id and super ego blocks are paired together, and the other two ordinals consist of L-elements isomorphic to adjacent quadra blocks. It seems like the section with priority and dimensionality exists to define priority and dimensionality operators over the L-functions, that is, each L-function is given a dimensionality and priority score. The derivations of the operators are actually not explicitly given in "Socionics Tweaks," and oddly 75% of the L-functions are given neutral dimensionality and priority. It is not exactly easy to understand why this would be. It is, in the first place, not easy to understand why the 1+2+8 function (called the A1 function in Model L) has a dimensionality of 4 and a priority of 4 given that the dimensionality of the functions 1+2+8 average to 3.67, and the priorities of these functions average to 3. There is some explanation for the extension of dimensionality onto L-functions in section 7 of "Socionics Tweaks," having to do with dichotomy and dimensionality structure of the club membership of L-elements, which doesn't seem intuitive at all but okay fine -- priority is not addressed there. The priority operator in "Socionics Tweaks" therefore at best is opaquely defined and, more realistically, is just merely defined badly, which is to say, the definition is wrong. Model L's only item pertaining to the quadra structure of Western Socionics is its priority operator. After excluding the priority operator due to both its questionable definition and overall low profile in the Model, there is no remaining quadra structure in the whole of Model L. There is a section near the beginning of "Socionics Tweaks", section 3, called Quadra currents, and another section at the end, section 13, called Quadra summaries, but both sections are very short and there is nothing in them except some logically chunked labels with very little elaboration or description. There is slightly more description in section 4, expanding on section 3, but almost all of section 4 is devoted to describing why the classical blocks (or maybe the L-elements, I can't really tell) are either aristocratic or democratic in nature. In any case, neither of these sections appear to pertain directly to the Structure of Model L. So in conclusion, no part of Model L is quadral, so it does not not by definition belong in Western Socionics.

Model L was not built with an isomorphism to the intertype relations in mind. Instead, it was built with an isomorphism to the Reinin dichotomies in mind, each L-function maps to a unique Reinin dichotomy (or equivalently and without loss of generality to the Tencer dichotomies). The Reinin dichotomies are shown in sections 9 and 10 of "Socionics Tweaks." More specifically, section 9 defines the Reinin dichotomies as Model L function dichotomies, and section 10 defines these same dichotomies over the socionics types and therefore these are literally Reinin dichotomies although the section is entitled something else for some reason. These sections are confusing because the Reinin dichotomies, as well as the Reinin equivalent L-function dichotomies, are all given different names than the usual conventional names used to describe Reinin dichotomies. This attitude of renaming established terms with no explanation and little description, which clearly reflects logical chunking, is a pervasive disease throughout the entire "Socionics Tweaks" document. The new names and short, insubstantial descriptions, certainly do nothing to promote any practical understanding of socionics, and like the Reinin dichotomies before them, do not very clearly apply to the types to which they purport to describe. Let's look at one example, the Anticipatory/Circumspect dichotomy, which is exactly the Carefree/Farsighted Reinin dichotomy. In a hyperlink from "Socionics Tweaks", the carefree (circumspect) types are described as having their "sphere of focus be largely spatial, whether focusing on actual or possible conditions", and the farsighted types are instead described as having the "perceptual sphere be largely temporal, whether of eventual or imminent occurrences." This is obviously very abstract, and it is not explained what it means to have a temporally focused attention on "imminent occurrences," as attributed to the ESXX types, like the erstwhile example type LSE. The definition is so abstract and so vague that it could really be rationalized to apply either way to any type. There is a name for the effect where people misattribute highly vague features to themselves: The Forer effect, or Barnum effect. The kind of definition that triggers the Barnum effect has a name too: garbage.

Throughout the entirety of "Socionics Tweaks" there are really only three things that appear to receive any significant quantity descriptive attention beyond logically chunked labels, or one sentence abstract descriptions (instead these features have short paragraphs of abstract descriptions that are still too complicated to understand). These are, dual-seeking pairs of IM elements, "co-dimensional function pairs" (in other words, unordered and ordered skew blocks when in the 1=8 or 1+8 position), and L-functions. The Reinin dichotomies and L-elements are not sufficiently well described (at least, I couldn't find any detailed description of L-elements) to qualify for this category of description, despite their fundamental importance in Model L. So, as I said at the outset, there is not that much to talk about in Model L that makes any sense, since the entire model is so abstract and the explanation consists of so much abstract garbage and many tables of diagrams pointing out Reinin structure, the same Reinin structure that I have previously called meaningless.

In his presentation about Model L on Jack's channel, Kimani discusses that there is a significant point of departure between Model A and Model L, in that the asymmetrical intertype relations are now symmetrical in Model L. This is an odd claim, which is not published in "Socionics Tweaks" as far as I can see. Because it's unpublished and not very elaborated in the presentation, I am not sure exactly what is being claimed, but in the video, Kimanis states that instead of supervisors projecting onto a supervisee's vulnerable function, both supervisors in Model L project into one another's negligent functions, which is function C2 in Model L. As such it is actually the radial orbitals themselves that are asymmetrical. In order for this intertype relation to be symmetrical, the C2 functions must consist of different block functions. In other words, the LSE has Te(S) as its leading L-function and it's supervisee SEI has as it's leading L-function Si(F). Therefore in order for these L-elements to be in the C2 position, the C2 position must consist of the 2+3 and 2+5 blocks for the LSE, and 4+1 and 4+7 for the SEI. At first, I thought this was so absurd that it was a mistake, but upon further researching the issue, I see that this is actually true in Model L. In section 9, it describes that the B ordinal always is the club with the same rationality as the type, and the C ordinal is the club with the same irrational axis as the type. For example, the LSE, an ST type, has the NT club as its B ordinal and SF as the C ordinal. While this does make supervision symmetrical, it leads to some bizarre properties. As one example, kindred types no longer always have the same relations with each other. The LSE is kindred to the LIE, and these types have each other's base L-function as B1. However, the SLI and SEI are kindred, and these types have each other's base function as C1. Therefore these types have different intertype relations with each other in Model L. I guess the reason for this change is that it makes the Reinin math work better, but it makes the block math broken. These differences in intertype relations, in fact suggest that Model L is not compatible with Model A from classical socionics, in contrast to Kimani's frequent and presumptuous claims that Model L is fully backwards compatible. It is possible that Kimani would acknowledge this. By the way, since there is not a single example of a type and all of its L-functions with the correct L-elements in "Socionics Tweaks", it was rather hard to understand the L-functions on the radial ordinals. The fact that there is no example of any type with all of its L-functions fully listed is a remarkable omission, and speaks both to the eccentricity of the model and to Kimani's negligence as a communicator. 

Does Model L come with any examples or cases? "Socionics Tweaks" doesn't refer to any examples or cases nor does Kimani have any publically facing typing work, at all, let alone typing work where he discusses or applies Model L. I had a couple of (contentious) conversations with Kimani, and he suggested a couple of examples of his typing work, including seeing Elon Musk as ILE, and Dan Tubb as LIE, the latter case one where he disagreed with Jack, who typed Tubb as ILE. In my view Elon Musk is clearly LIE and I looked at the linked interview for Tubb, who seems to me clearly a beta ST and not LIE or ILE, my typing is SLE. So, despite a small sample size, I find the application of Model L to casework filled with club and dimensionality stereotypes, and wholly unimpressive. Overall, public typing work, whether of famous people or ordinary people, does not seem to be a priority for Kimani in lieu of continuing to produce more hyper-abstracted written modeling.

In summary, Model L is extremely bad. The creation of a 16 element model is not bad. The design choice to emphasize clubs and dimensionality is bad, and a step away from western socionics. The design choice to make L-elements a combination of two blocks instead of one block is bad and bizarre. The design choice to emphasize the isomorphism between functions and Reinin dichotomies is extremely bad. The design choice to make ordinals vary (ie, their blocks from Model A16 are different) depending on the rationality of the type is very bad and very bizarre. The descriptive quality of "Socionics Tweaks" is extremely bad, all of the descriptions in the document are extremely vague and empty of meaning. The lack of focus on practical typing from practitioners of model L is unfortunate, although common. The sample of typing work of famous people that I have seen from Kimani is bad, although of small sample size. There is no positive feature or redeeming quality in Model L. 

Lecture: Western Socionics Modeling

The second lecture is about modeling, it primarily focuses on Western Socionics and discussing the differences between Western Socionics modeling and other modeling, and also differences within some practitioners of Western Socionics. I titled that lecture Comparative Modeling in Socionics, but that's really a very bad title that I shouldn't have selected, it's not very much about modeling that isn't Western Socionics.



This lecture features a lot of original material about the modeling that I personally use, which I call the Theory of Quadras and which I am preparing for publication in an upcoming book which will be entitled "The Theory of Quadras." In preparation for editing and publishing that book, I am very interested in comments from anyone who might watch this lecture about the material. I also created this google form to solicit some comments. You could also leave comments here on this blog.

Lecture: How to do Typing Work

In the past couple months I did some lectures on some topics in socionics for a discord audience. I recorded the lectures and present them here for anyone who is interested and prefers an audio or perhaps podcast format.

The first lecture is on typing methods, and an overall explaination my machine learning science project, why I started it and what it's trying to do.

Another way to qualify for being typed without doing my survey

Some of you are lazy and don't want to fill out my survey. Apparently some people have complained that it is too long and I have been mostly unsuccessful at shaming those complainants into completing it anyway. 

Today I have a new method for you to qualify to be typed by me even if you do not wish to fill out my survey. Instead, I'll type you if you bring me data another way, in the form of referrals. Get four of your friends, acquaintances, or enemies (who have not already taken my survey) to fill out my survey and complete a typing with me. And do tell them that my interviews are recorded so that they are not surprised. If you refer four people, after their typings are completed, I will type you regardless of whether you complete my survey or not.

Credit for this idea goes to Blah, I am trying it out.

I will post a mini update on how my project is going sometime soon. 

Inter-system correlations

The question of whether type X in system A correlates with type Y in system B is as old as typology. Nonetheless, this topic holds an especially renewed interest in pop psychology in the last few years, with very many people posting their own original content of spreadsheets suggesting which combinations of common typologies they deem possible, or unlikely, or impossible (or sometimes "possible only in fiction".)

Virtually every single inquiring mind who engages with this question approaches the topic incorrectly. The answer to the question of whether a type 4 in the enneagram can be a seemingly dissimilar type such as the SLE is yes, by definition since these systems are independent of one another.

A more clearheaded, useful question is: How common is it for a type 4 to be an SLE? Or, addressing the question of seemingly dissimilar combinations' existence in practice, as opposed to in principle: Have you ever seen an SLE 4? If you have, how confident were you about both types? And, if you have can you show me about them so that I can judge the types for myself?

If nobody has any examples of the combination, that speaks for itself about its frequency. And if people do have candidate examples to share, even candidates that they are not very sure of, then the discussion and debate can proceed in earnest. As with anything else in typology, the proof is in the examples and the case work.

Instead of asking questions based on examples, discussions in the pop typology community usually proceed to argue about definitions and descriptions, with such comments as "A 4 cannot be an extroverted type since it is in the withdrawn triad", or "The type 4 is described as being in touch with their emotions according to description X by author Y, and the SLE is described as the opposite by author Z, so they must be incompatible." Of course, such arguments depend on descriptions and author interpretations, and abstractions of categories into language. The interpretations of different authors are not in absolute concordance, just as the typings of different practitioners are not absolute concordance. More importantly, descriptions in text are maybe interesting to examine and interesting for beginners to get a foothold in a system, but they are not as effective a way of conveying understanding of types as examples in the flesh that can be seen, heard, and tasted. Which, is the kind of content that this blog was created to disseminate and will still disseminate after this theoretical interlude regarding issues in practice.

Dimensionality and Axial Post-Dimensionality

This post does not have examples of cases, I continue writing about thorny theoretical things that are not instructive examples. It involves the topic of dimensionality, a framework for thinking about socionics types that I resoundingly reject. This topic is only very little better than sheer nonsense. If theoretical nonsense does not interest you, feel free to ignore this post. However, the idea of dimensionality has a foothold in western socionics content from other authors, so it might be of interest to navigate.

The presence cube model

Other authors, including Jack Aaron and Ibrahim Tencer define functions as expressing different strength and value parameters. Specifically, the ego and superid blocks are considered valued, while the ego and id blocks are considered strong. This distinction also exists in Augusta's work, although in Augusta's work it seems very little emphasized; it seems to me that the emphasis on the different flavors of functions is more modern. Generally, valued functions are seen as psychologically rewarding to attend but not necessarily skillful, and strong functions are considered skillful to attend, but not necessarily rewarding. (Throughout this post, I will refer to the idea of separation of strength and value as "the presence cube", although the presence cube is a model developed by Ibrahim Tencer linked above. Other authors like Jack Aaron do not use the same terminology, but his idea similarly involves separate strength and value and I will also refer to his modeling, and other authors' that I will not name here, as the presence cube). 

Dimensional Strength, or just dimensionality was an addendum theory developed by Vladimir Ermak and Aleksandr Bukalov in the 90s. It characterizes gradations of strength within the weak and strong blocks, and also suggests that the skilled use of functions varies in generalizability; high dimensional (high strength) functions have access to more domains of reflection than lower strength functions. Almost nobody in the western community pays attention to the specifics of these domains regarding whether people have access to experience, norms, situations, and time; instead they regard dimensionality as mere gradations of strength.

It is important to recognize that dimensionality is aquadral. The creative function and ignoring function have identical dimensional strength, despite the ignoring function being out of the quadra. In the presence cube, the value dimension is called "priority" and is a simplification, but a reasonable approximate measure of which quadra blocks are emphasized in the different types within a quadra. As such, the presence cube model is a combination of quadras and aquadral dimensionality. Western Socionics in general is the idea that socionics is a theory of quadra values. The presence cube model with both quadra values and dimensional strength is thus a kind of hybrid model, that socionics is a theory of quadras but not entirely so, although Tencer, Aaron and other authors tend to regard quadras as the more important variable than dimensionality, which makes their models recognizably Western. However, the model that follows, which views socionics as entirely a theory of quadras and entirely ignores dimensionality, is more Western.

Axial Post-dimensionality model

The presence cube making use of dimensionality alongside quadra values has been heretofore the more popularized flavor of Western socionics. I wish to point out some flaws of dimensionality and the presence cube and present a different and simpler model.

What is a strong function, anyway? Socionics is a model of information metabolism, which means that it is a model of how different people differently pay attention to different sorts of information. The idea of strength and value as separate, independent parameters of functions implies more, that not only is there a hierarchical rank of some domains that are more attended and some that are less attended, in addition there is an assumption that these hierarchically ranked attentions are qualitatively different in their skillfulness and in their rewardingness. Likewise, some functions (e.g. the ignoring) are skillful but unrewarding, and hence, the theory goes, ignored and hardly ever utilized, while others (e.g. the mobilizing) are rewarding and therefore subject to frequent practice, but are inherently unskilled, and are never fully grasped as skillfully as those people who have this function as a stronger function.

The presence cube model has the following bad features:

1. It assumes that some functions are rarely used, but nonetheless skillful, which is hard to measure (i.e., in principle how do you observe someone being skillful at something that they rarely or never do?).
2. There are two parameters of function quality, value and strength. Two parameters are harder to measure, and harder to understand, than one.
3. Having additional free parameters in a model dilutes the explanatory power of that model.

In my model, there is only one parameter to measure per function, degree of attentional visibility, which I call strength. With fewer independent things to measure, it is more straightforward to measure change in function strength over time, which led me to my idea that the superid functions level up over time, and thus change their strength as people grow and work on their weaknesses.

In this model, the mobilizing function is a strong function, and the ignoring function is a weak function and blind spot, equivalently weak or weaker than the vulnerable function. The fact that dimensionality describes the mobilizing function as weaker than the ignoring function is the most glaring issue with the concept. I can not understand what property the ignoring function has that makes it stronger or more skillful than the mobilizing function in any respect whatsoever, as dimensionality predicts. It is very hard for me to understand why the ignoring function which is rarely or never used could be considered strong in any way. The classical view that the mobilizing function is weak, makes more sense, and the classical notion of the mobilizing function as "stubborn" or misguided can be seen sometimes, usually in young people that have not developed much confidence and not developed their superid functions very well. But usually, people have a lot of confidence in their mobilizing functions, and emphasize this function a lot, and make it externally visible -- and by using it frequently and leveling it up early on in their lives, they usually become skillful at it pretty quickly.

I came up with the following simple math to translate between my axial view of model A, and the parameter of function strength output:
strength = axis*balance

In other words, each of the four function axes (e.g. Ti and Te) has a overall strength level and an imbalance towards one end. The logic axis of the LSE is very strong, as it is the dominant axis, but extremely weighted towards Te, in other words extremely imbalanced. The logic axis of the EII, by contrast, is much weaker overall, and more balanced.

As an example of this model for proof of concept, with sort of ad hoc made up numbers:

LSE

Str(Te) = Str(logic) * balance(logic) ≈ 0.99*0.99 ≈ 0.99

Str(Si) = Str(sensorics) * balance(sensorics) ≈ 0.9*0.7 ≈ 0.63

Str(Ne) = Str(intuition) * balance(intuition) ≈ 0.7*0.97 ≈ 0.7

Str(Fi) = Str(ethics) * balance(ethics) ≈ 0.5*0.45 ≈ 0.23

Str(Fe) = Str(ethics) * (1-balance(ethics)) ≈ 0.5*0.55 ≈ 0.28

Str(Se) = Str(sensorics) * (1-balance(sensorics)) ≈ 0.9*0.3 ≈ 0.27

Str(Ti) = Str(logic) * (1-balance(logic)) ≈ 0.99*0.01 ≈ 0.01

Str(Ni) = Str(intuition) * (1-balance(intuition)) ≈ 0.7*0.03 ≈ 0.01

This pretty well matches the function visibility weights that we normally see in people. Note that these parameters, axis strength and axis balance, are 8 parameters, same number as if we treated each function as having independent strength (and fewer parameters than the presence cube in which each function has an independent strength and value, so 16 parameters to measure). 

The "developmental parameters" which change over time can be attributed to the balance of the suggestive functions axis, and possibly the axis strengths for the weaker axes will become better developed as people work on their weaknesses and improve them. Most of the parameters will likely be more stationary. Importantly, axial post-dimensionality allows for modeling these developmental changes as nonstationary measurements over time without introducing more parameters. We can simply model the development of people improving their weak points as increases in axis strength over time, while leaving the axis balance stationary. 

The post-dimensionality axial model is better than the presence cube model.

1. Axial post-dimensionality has fewer parameters to measure than the presence cube.
2. It explains why equivalently valued functions in the presence cube have different strengths.
3. It explains the property of the role and demonstrative function as being "half valued" -- which the presence cube also models, but with different numerical outputs. (specifically, the presence cube considers the demonstrative 4D+2P as much stronger than the role 2D+2P, which I think is dubious, and the ignoring 3D+1P as equal or roughly equal to the role 2D+2P. Tencer suggested to me a modification, where P (valuedness) is weighted more highly than D (dimensional strength, which allows for the role to be stronger than the ignoring; however, unless the coefficient for dimensionality is zero or very close to zero, the strength of the ignoring is estimated too highly for me)
4. It more clearly describes what parameters are changing than the presence cube when developmental changes occur as people improve upon their weaknesses.

Dimensional strength is a bad theory and should be discarded. The presence cube which incorporates dimensionality is a bad model and should be discarded.