Intelligence & Abstraction vs Association/N

Home Forums General Psychology Intelligence & Abstraction vs Association/N

Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #20379
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    Thread split from: https://cognitivetype.com/forums/topic/a-computational-metaphor-pt-2-v-m/

    Sidenote about General Intelligence vs the "N" category:
    In the neural network video above, the author uses 4 neural network layers to give his representation. This gives me an opportunity to talk about something that is relevant to typology going forward, and for newcomers to CT; intelligence and type.
    Now it's my current understanding that the amount of "layering" a person can do is moreso a function of their IQ (G factor), than anything to do with the cognitive processes of typology. We all know that type is independent from intelligence, but we need a better comprehension of "how" this is factually true. Without a more proper understanding of this independence, the correlation of N with intellect (thanks to mbti) won't properly go away.  So I wanna try to model this problem in more detail.
    High-order topics, such as politics or philosophy, which require a sophisticated handling of objects at such high levels as "economic systems" (composed of so many other sub-components) and "epistemology", are the result of greater cortical power, which allows greater layering. And so as a semantic point, if we choose to call this layering capacity "abstraction", then abstraction is a function of IQ, not at all "N." The "N" would instead be association, which is horizontal not vertical. I'll try to show what I mean with a few diagrams:

    ^ Above is a diagram showing ever-higher order objects, as created by neural networks. It's another view of the one Kurzweil has above, with the word "Apple." How "abstract" a person can think, is a measure of how much vertical layering they can do, of this sort. But this is fundamentally different from association formation, which runs horizontally like this:
    A person's ability to associate information horizontally (such as by connecting object properties) is not the same as how densely one can "stack" objects inside objects like Russian dolls (vertically). The former is N, the latter is closer to something like IQ.
    A high IQ person with S over N priority will think at highly abstract levels (SiTe Jeff Bezos, TiSe Elon Musk), but will always be inclined to conceptualize objects (even highly abstract objects) as formed by discrete parameters; a clarity of topic, without mixing things up into a hazy horizontality of topic-boundaries. A "this is this" tendency, even if "this" is a topic as nested as quantum physics or monistic absolutes.
    Again, this is common knowledge to anyone who has any sense of what type can or cannot be, and it's annoying that it even requires mentioning, but I figured using this analogy of verticality versus horizontality may be instrumental in quantifying what the real difference is, in a way that can be modeled.

    #20394
    fayest42
    Participant
    • Type: FiNe
    • Development: ll--
    • Attitude: Unseelie

    Thanks for sharing this. I'm excited to see where else this computational metaphor idea goes. One thing I'm confused about is the little side note about the difference between vertical abstraction and horizontal association. To me, it seems like horizontal association is a necessary part of the vertical abstraction process. One way of describing abstraction (from wikipedia) is "as a compression process, mapping multiple different pieces of constituent data to a single piece of abstract data; based on similarities in the constituent data, for example, many different physical cats map to the abstraction 'CAT'." So in order to form the abstraction "cat" (vertically), you have to recognize that all the different individual cats you've seen are associated with one another (horizontally). Is that not right?

    #20412
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    edit: I've split this off into it's own thread since it was kindof off topic to begin with, but could make a good discussion on its own. 🙂
    So I really wondered if I used the right wording and went to read that Wikipedia page, and I think it is the right concept after all. I think vertical abstraction and horizontal association and quantitatively and qualitatively separate things. Let me see how I could explain...
    The page (here, for instance) gives examples of how abstraction is used in computer language to define some variable/function and then call that variable/function as a shorthand. We can say "Einstein's Theory of Relativity" to summarize "E = mc2" and its permutations and body of mathematics, or we can say "Maxwell's equations" to summarize a body of equations into a new [object]. This is abstraction, and is a sort of nesting process.
    This itself is quite different from association, as it has nothing to do with interlinking data with other data, and is instead a kind of summarization process. It's a theme I see across all examples on the page. One more example given is the summarization of information in this hierarchy schema:

    Spoiler

    (1) a publication

    (2) a newspaper

    (3) The San Francisco Chronicle

    (4) the May 18 edition of The San Francisco Chronicle

    (5) my copy of the May 18 edition of The San Francisco Chronicle

    (6) my copy of the May 18 edition of The San Francisco Chronicle as it was when I first picked it up (as contrasted with my copy as it was a few days later: in my fireplace, burning)

     

    [collapse]

    Here we see an [object] nesting, with the highest levels (1/2) being more abstracted. It moves away from an abundance of specific details, to fewer details. But this is not the same thing as an association (two+ separate things are not being joined). The symbol "a publication" is itself still a detail and is discrete (S) even though it has within it larger layers of discrete [objects].
    So there isn't very much about abstraction that is associative, although I have to describe this better below. It seems instead to be about shorthands, generalizations of already existent categories.

    J?

    As for whether it requires association, I think so, but I think it generally requires all mental faculties, and association doesn't seem to be particularly important to this.
    If I had to say whether it has any stronger connection to other operations, I think it has more to do with object-definition/classification as (J). Languages, mathematical or linguistic, are very J-heavy activities, with Je being the Articulator and Ji being the semantic/definitional critic. The creation of object-categories that have nested objects within them appears to require at least the participation of both Judgment and Perception. But the layering of objects cannot happen unless we use some sort of linguistic system to systematize it and give it hierarchy.
    And if object delineation is a property of J, then J can also be largely responsible, via self-similarity, to do that at higher levels. The hierarchy of objects is created by J, which is why Ji (as per this post) is responsible for making sure the layering/stacking is coherent up and down this layering.
    This would also explain why we see J-leads (or a J development) more associated with an investment in language systems, whether they be economics, physics, metaphysics, ideology, etc.

    Association (N)

    Association, appears to be quite a separate operation. It seems to have a sort of 'low threshold' for identifying commonalities between objects or events, via focusing on shared properties and features.
    N interlinks objects together, but it doesn't itself summarize that into a higher order object. That actually goes against the way the P functions work, which is holistic and expansive, rather than reductive. Pure N, if it was possible for it to exist in pure form, would interlink all objects together indefinitely, being able to trace threads from one to another, as one continuous object. But it would not create hierarchy, which requires differentiation/"cutting".
    For example, suppose that N says that bats are like birds, which are like butterflies, which are like moths, which are like airplanes. (J) would perhaps quantize this and say "you're describing an object category called [things with wings]". But N would actually be annoyed with that, because as soon as you delineate, you cut the flow of association. N may have wanted to continue saying that moths are also like grasshoppers, and grasshoppers and like ants. J may then interject and say "you're now talking about [things that are insects]", but N saw that an airplane isn't an insect.. , nor do ants have wings. And the arbitrary delineation of borders in information will seem like tyranny to N.
    Therefore, N/association must work horizontally across data points, but not vertically, because verticality requires an orthogonal process to it, which cuts and frames things by logical coherence. Therefore abstraction is information-eliminating (summary), handling only the concept without the data itself, and this is quite different from N. N, like S, is all about the data and is about feeling-out data conjunctions. It's not reductive, but expansive. By contrast, abstraction is the omission of data for the construction of grammatical objects, which themselves have lower objects beneath them.

    #20418
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    Ugh, having said the above, I just realized I've made quite the same error early on, (in word choice), that other typologies have made. N is not abstraction, yet it has been used as an adjective to describe it. @jelle you were totally right, lol.
    Apparently, in that wiki link it even says that Jung knew this, and considered abstraction something separate from the other processes, but which can happen to all of them:

    There is an abstract thinking, just as there is abstract feeling, sensation and intuition. Abstract thinking singles out the rational, logical qualities ... Abstract feeling does the same with ... its feeling-values. ... I put abstract feelings on the same level as abstract thoughts. ... Abstract sensation would be aesthetic as opposed to sensuous sensation and abstract intuition would be symbolic as opposed to fantastic intuition. (Jung, [1921] (1971): par. 678).

    It's curious to me how he even signals out "abstract sensation", which he calls aesthetic, rather than sensuous.
    Abstraction really is conceptually a different operation than association. UGH, old man Jung's still taking me to school.

    #20423
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    A few more notes...?
    Thinking even more on this, association (N) really is directly opposed to abstract verticality, as association disrespects object boundaries, which are the main feature of hierarchical abstraction.
    This is why Ne, for instance, leaps across disparate objects that belong to different object-groups in a given abstracted hierarchy and pays little mind to it. And it's also why Ni has no problem associating/equating things at a wide-scale as well, without differentiating the elements it crosses across.
    And of course I do think that, when an association-chain is created, it can then be quantized into a [higher object] within a hierarchy with the help of J, but this has to be a different process than N, because N doesn't stop itself there. So it will just as easily go around and violate that J category in order to interweave more later; continuing to see that the object is continuous across more data.

    * * *

    I do want to say a few more things about this though, since I'm seeing the need for more clarity on how J and S differ, and how N differs from both.
    (J) Category -- A (language) 'category' is created by J. An abstract category can be, for example, [cat]. In an ideal world, J creates categories overlayered atop of manifest reality. The best systems (i.e. physics) work because their categories map onto manifest reality well. But the abstract system is not equal to reality as-such.
    (S) Discrete Data Parameters - But S is discrete or literal data, which is not to say physical, but finite. It is not a category, it is a finite set of adjacent information. For example an Si anecdote is discrete in that it has a beginning and end, and you know what is and isn't part of a given set. An Se observation (tunneling) leads to a chain of adjacent discrete data points, which create an object but that object "is what it is", and exists not as a category, but as its own thing. Superimposing a category onto it actually reduces and eliminates the data. Because not all cats are the same.
    There is no way to describe discrete perception other than what it is, as such. It's non-linguistic. It cannot be reduced without talking about 'something else' fundamentally. A proxy. So S is quite different than what J works with, even though both have different types of 'limits' to what they work with.
    (N) Associative Data Parameters - N is an associated data-chain. Rather than being linked by literal data adjacency or non-adjacency, it's linked by certain features/properties, where the features are the defining parameters, even though the features themselves are non-verbal and non-categorical, but data contingent. For example, a certain kind of eye-shape may be noticed across data sets and someone might say "you look like [x celebrity]". If asked why that is, it may be difficult to articulate because the reason is not coming from a shared linguistic category, but a data group.
    Good use of J will also create categories around S and N information types alike, even though these information types will never be synonymous with categories.

    #20424
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    S and N working within higher abstracted layers
    Now, it gets a bit interesting when you have S and N working within very abstracted language layers. Because then the "information" available is in the form of linguistic objects.
    S then has to content with this object, but it does so the same way it would content with sensory information; by looking for discrete parameters. So a strong S type may be able to tell when an abstract object (as as topic) is no longer being talked about, because we've moved away from the object as-is. This translates to a keenness at identifying the natural limits of any idea, and not over-extending.
    Oppositely, when N is working within abstracted language layers, it associates other abstract objects (topics) by shared features, perhaps without noticing the topic has changed (or feeling it hasn't). So for example, the parameters of an object (topic) may glide across minimum wage to universal basic income to potlucks to codependency to child-rearing to meme culture to social media and then back to capitalism. And you get from a convo about minimum wage to one about meme culture in 2 minutes.
    (This slip-n-slide across topics is something that plays out differently in Ni and Ne, with Ni traveling cross-country on a train railway, while Ne is road-tripping and hitchhiking. But the effect is similar.)
    Having observed people like TiSe Jelle and SiFe Phibious discuss abstract topics, it is characteristically different than listening to Umbi and Saf, for example. And I know Jelle likes to point out how I am "slippery" in debates ;p because I apparently swerve topics without noticing, and this comes across as a sort of avoidance that violates intellectual integrity in a debate setting. I always found that funny, heeh.

    #20432
    Ninth
    Participant
    • Type: TiSe
    • Development: l--l
    • Attitude: Directive
    Spoiler

    An Se observation (tunneling) leads to a chain of adjacent discrete data points, which create an object but that object “is what it is”, and exists not as a category, but as its own thing. Superimposing a category onto it actually reduces and eliminates the data. Because not all cats are the same.

    So a strong S type may be able to tell when an abstract object (as as topic) is no longer being talked about, because we’ve moved away from the object as-is. This translates to a keenness at identifying the natural limits of any idea, and not over-extending.

    I don't know if I count as a strong S type, given my dev levels, but that resonates totally.

    [collapse]
    #20435
    safsom
    Participant
    • Type: NiTe
    • Development: ll-l
    • Attitude: Unseelie

    Very insightful comments and good posts being made on this thread. Having studied Jungian typology for a long time, and being immersed the process of constructing my own typological system, I too, have had to deal with the layers of "abstraction" that certain functions can deal with. The monopoly that intuition and thinking seems to have on "abstract discussion" or "intellectual topics" in other typology systems (of which MBTI is the most prominent in perpetuating these biases) seems totally wrong to me, because all functions can deal with abstract (non-concrete) data parameters. The only difference lies in the degree to which the data is removed from the situation and taken as either part of another situation or all other situations. Auburn and I seem to be in agreement on this for the most part.
    I would like to expand a little bit by creating a dichotomy that I think would help expand the architecture that Auburn has laid out in his post above, and that dichotomy is what I call Constrained/Unhinged dichotomy. Seems that a variant has already been alluded to above (when the differences between N and S were discussed in terms of association), but that J has been described as a singular unit. My aim in this post is to prove to you why a combination S+T is what closest aligns with society's standard of intelligence and even that on measured tests like IQ tests - along with producing the most potential for scientific thinking. It should also be apparent that CT's NTs are more into intersectional science (with philosophy or other disciplines) rather than pure, empirical sciences.
    At heart S+F and N+T are opposites not because the latter is the combination that has a proclivity towards intellectualism (indeed you see many intellectually inclined SFs like @hackphobia and @bruhh and many frankly stupid NTs like Charlie Sheen), but they're opposites because of the situational constraints on each function. As Auburn has mentioned above, S is focused as it recognises the differences between each object (Constrained dichotomy) whereas N is Unhinged and thus does not really contribute to this.
    T/F is a more complex matter because as a process, being biotic at heart, both F functions require more situational details to account for biotic contingencies whereas T functions don't necessarily have this orientation towards the collection of those details, meaning that T as a process is inherently more Unhinged than F and able to encompass a wider variety of situations. What this means is that the approach will differ on the basis of being either situational and axiomatic (F), as opposed to generalized and principled (T). The feeling type is essentially bound by the details and burdened by the task of perfecting those (seen in many of the Fi revisors here on the forum too), where the thinking type is much less so.
    This understanding allows us to finally accurately frame MBTI's interpretations of SFs and NTs. Simply put, in CT, NTs would theoretically (by these criteria) be the most generalising and least situationally constrained types whereas SFs would be most attentive to the boundaries of both situations and abstract concepts, perhaps making them better analysts in some areas (like philosophy or ethics, where a combination of S + J is useful for precisely painting a picture of an object). Similarly, this also debunks MBTI's "NT scientist" correlation as the degree to which intuition and thinking are divorced from situational thinking is unlikely to make an ideal combination for weighing empirical data, instead it is far more suited to fields where abstraction and structuring those are required, so areas like epistemology, ontology and possibly the intersection of science and panpsychism (which you see in almost every NiTe sample) are better areas for NTs.
    All this being said, I think that the types most likely to score high on an IQ test would be STs and not NTs as other typologies seem to think, because of the combination of an awareness of the beginning and ends of objects (which is a property inherent to both S functions) combined with a simultaneous ability to take those object understandings and frame them in terms of more general axioms for accurate deduction (which both T functions deal with) -  a skillset required for virtually ever modern IQ test. For the same reason, I also posit that a combination of S+T lends itself to being the best scientists, becuase of that situational awareness and analytical power. CT NTs are more likely to be Teal Swan. This is why I place my estimate for the most intelligent CT type on average to be the SeTi, a combination which lends itself to astute observation and lightning accurate analysis, perfect for IQ tests and even science.

    #20439
    Auburn
    Keymaster
    • Type: TiNe
    • Development: l--l
    • Attitude: Adaptive

    Saf, I like the dichotomy constrained/unhinged. The terminology certainly comes closer to what we've had before, which should be removed asap:

    Spoiler

    I've crossed these out for now.

    (Next time I'm doing a round of updates on articles, I'll be removing "abstract"/"concrete" from everything.  As well as "logical"/"ethical". For now I've put a notice on this article saying it's outdated.)

    [collapse]

    In what will be the new revisions, I think N vs S can be described as Discrete vs Associative, Constrained vs Unhinged, and I think I heard someone suggest Discontinuous vs Continuous. I'm not entirely sure these are the final adjectives we should land on, and maybe we'll find better adjectives, but we're definitely moving in the right direction.

    My aim in this post is to prove to you why a combination S+T is what closest aligns with society’s standard of intelligence and even that on measured tests like IQ tests – along with producing the most potential for scientific thinking. It should also be apparent that CT’s NTs are more into intersectional science (with philosophy or other disciplines) rather than pure, empirical sciences.

    I've noticed something like that too. Specifically, the lifepaths of the database TeSi's and SiTe's tend to follow stronger scientific or academic focus and success, if not also financial/economic. But I tend to think ST Deltas are better equipped for "formal sciences," as a specific topic, because of the nature of sciences when it comes to actually doing the work on a desk.
    While ST Betas are more equipped for another type of excellence, typically in a mastery of method, leading them to be specialists. TiSe's in particular are acute like a sniper who can hone into just the right problem and have an "empirical" (in the sense of observation-based) means to frame that problem and then solve it. This could be anything from skating like TiSe Tony Hawk, all the way up to building rockets like TiSe Elon Musk. But yes, their cognition directs itself to some discrete and constrained domain, which makes their Ti able to hone itself, lock-on and perfect the approach to the problem.
    (I like the term "intersectional", as it does seem to capture a certain phenomenon I see in high N users, and it contrasts neatly against both "specialists" and "scientists.")

    Temperaments?

    But, overall I see function quadrants (Alpha, Beta, Gamma, Delta) playing heavily into differences within temperament categories like S+T or S+F. I think "SF" varies a lot if we're talking about Alpha or Gamma, for instance.
    However, perhaps now with these developments of a more coherent understanding of S, N and T, F, the temperament categories can ease back into focus? I think you mentioned on Discord that it was unfortunate that they were omitted, which was because they were poorly framed before. The older view of temperaments was discarded when samples/types failed to fit those categories. The function axes have proven to be far more consistent in what they predict, in many more details and emergent nuances.
    But if we can find a framing of them that reflects our more recent understanding of them, I'd be curious to see what emerges.

    #20504
    safsom
    Participant
    • Type: NiTe
    • Development: ll-l
    • Attitude: Unseelie

    Insightful reply, Auburn, and I agree with many of your observations of the temperaments - though they do have a certain validity to them, function quadrants as a whole seem to capture the epistemic tendencies of certain types far more accurately, and far more astutely. Indeed most of the explanations that were previously done via the temperaments are now elegantly and concisely explained by the development levels (though those do suffer the predicament of not having independent pages to describe them on the CT website - some people may not be able to discern them purely through vultology).

    But if we can find a framing of them that reflects our more recent understanding of them, I’d be curious to see what emerges.

    I think that the problems with the previous temperaments lie directly in the problems with the previous standards of the functions (which you described in your post) - a lot of them sound rather MBTI-esque in giving functions like intuition and thinking an edge towards abstraction while discounting the capabilities that sensing and feeling have in those areas too - leading to rather behaviorist categorizations that almost echo MBTI descriptions. I think currently, the definitions of the functions that we are using are far more fit to base a temperament theory on because of their property of being abstracted from behavioral traits, that is, the level at which they are defined is more epistemic than it is behavioral.
    Let's consider the Ti+Se example that you used and do a comparative analysis of how it may be perceived depending on how the symbols "Ti" and "Se" are defined. If we are to use the older definitions of "Logical" and "Reality", you have the implication (since that is what we are assessing in this paragraph) of someone that is attuned to solving immediately "real" problems through the application of logic - however, with our current understanding of energetics, we know that these are not particularly Ti+Se traits but in fact Je traits (since that is the function that is most attuned to externally perceptible inefficiencies). Compare this to when you frame this as "abiotic" and "literal" - there are not many behavioral implications here, but rather the implication that someone is able to take an abiotic and delineated view of anything - whether that be fixing motorcycles (a la the ISTP mechanic) or solving metaphysical problems (a la Jelle). The current understandings of the functions are abstracted without behavioral implications.
    This is perfect for framing a new understanding of the temperaments as that was their key problem - they were primarily behavioral descriptions with little focus on ontological/epistemic aspects of an individual's experience and so the behaviors described within the temperament pages often did not align with a person's behavior or psychology. However, now that there are more abstract definitions of the functions to work with, what can be done is defining not behavioral but epistemic temperaments - that is, taking aspects of common function pairings and attempting to categorize epistemology and possibly behavior/interests following that. Each temperament can then have variations depending on the lead energetic (Conductor/Revisor).
    I'll give an example by redefining one of the temperaments that existed prior to the development levels system - the ENT temperament. Rather than describing them as a nerdy, somewhat heartless dork (as these are behaviorisms) it is best to define ENT as a temperament where the dominant energetic (not necessarily the dominant function, moreso the ego-orientation that the synthesis of conscious functions yields) within the individual is externally oriented and either N+T is at a natively high or very conscious slot for the individual, with one of these two variants (accompanied by mock archetypal portraits for each one):
    Conductor ENT (native conductor functions + extroverted (Je/Pe) energetics)

    The conductor ENT, though natively being an articulate and decisive type, may have tendencies towards expansion and integration of a myriad of intellectual interests, due to their largely external orientation. They will be constantly voicing opinions on any encountered datasets and will also be on a search for new datasets to integrate into their stores of knowledge. Sometimes, they will take a synthetic approach these frameworks and propose new frameworks to solve problems with confidence in their knowledge and knowledge-gathering abilities - leading to a reputation as a "devil's advocate" or perhaps even arrogant.
    Revisor ENT (native revisor functions + extroverted (Je/Pe) energetics)
    The revisor ENT exists in a frenzy of gathering information and attempting to understand the logical truths behind it - they want to be divorced from their own perceptions of the information itself, and understand the objective roots of what the information implies. They exist in a contradictory state, though they are natively constantly on the search for new information and thus will refrain from stating opinions, at times, you may catch them making a reflexive judgement off of their own guard. It is this state of accumulation of information and half-baked articulation/hashing that will earn the revisor ENT the reputation of being a ditz, albeit a genius one.
    As an NiTe ||-| by the current codifier, I would fit the Conductor ENT temperament rather squarely (native proclivity towards conductor functions + heavily extroverted energetics). Now, with this structure outlined, I'm pretty sure you can figure out the rest of the temperaments. What would yours be?

    #20543
    fayest42
    Participant
    • Type: FiNe
    • Development: ll--
    • Attitude: Unseelie

    @safsom I think, if I'm not mistaken, that the reason the old temperament descriptions are written in terms of behavior rather than cognition is because they were not necessarily meant to line up with your CT type. I think they were a way of acknowledging that someone might feel like a typical MBTI "NT" type while still truly being a, say, TiSe l--- (or whatever other combination).

    #20951
    Supah Protist
    Participant
    • Type: SeTi
    • Development: ll-l
    • Attitude: Directive

    @safsom is another way to rephrase the constrained vs. unhinged dichotomy as differences vs. similarities between objects?
    Also, if NT vs. SF is Unhinged vs. Constrained, is there another dichotomy for ST vs. NF?
    Edit: Analytic vs. Synthetic maybe another way to conceptualize what you're describing.

Viewing 12 posts - 1 through 12 (of 12 total)
  • You must be logged in to reply to this topic.
A forum exploring the connection between Jungian typology and body mannerisms.

Social Media

© Copyright 2012-2021 Juan E. Sandoval - Use Policy
searchhomecommentsenvelopegraduation-hatbookearth linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram