Triangulation (social sciences)

[guitarplayer]

Turns out triangulation is an awfully versatile word, spanning from arts, media technology, through mathematics and technology to social sciences and (used rather awkwardly there) psychotherapy.

A brief search on the forum reveals that the word has been used only a handful of times, in the context of finding location/maps and psychotherapy.

I remember first reading about application of triangulation in social sciences over a decade ago and it stayed with me as a good mental tool to think about the world and how we get to know it. From wiki

In the social sciences, triangulation refers to the application and combination of several research methods in the study of the same phenomenon.[1] By combining multiple observers, theories, methods, and empirical materials, researchers hope to overcome the weakness or intrinsic biases and the problems that come from single method, single-observer, and single-theory studies.

@Ego used the word once here

One Extreme.................▲................The Other Extreme
The average person decides what is "reasonable" by triangulation. They look at the two extremes and find a point somewhere in the middle that allows them to avoid standing out.

which is perhaps close to my understanding of a simple application of the 'social science' version of the word in everyday life.

In terms of the theoretical considerations on the forum, accounting for different perspectives as in yellow / metamodern view of the other stages might represent a good example of triangulation in the social science sense of the word.

From the wiki

Triangulation can be used in both quantitative and qualitative studies as an alternative to traditional criteria like reliability and validity.

which is indeed a good 'cross-sectional' or 'cross-polinating' alternative to reliability and validity which are traditionally used e.g. in psychology to indicate strength of a theory. For example, from the traditional point of view MBTI typology (or psychodynamic theories) has been repeatedly questioned by mainstream psychology, but using triangulation it could be argued that it is valid because it makes sense from other points of view, not least applicability and 'granularity'.

I have an intuition that it would also be possible to bridge (somehow, metaphorically again) triangulation in this sense or the word and ergodicity. But this thought probably needs to sit in the background for a while longer. Also, this might all be rubbish!

Anyway, I propose we include triangulation (social science) in the vernacular here.

[daylen]

Developing a system for producing drawings that match up to arbitrarily precise experimental designs in psychology has been on the back of my mind for a while now(*). I think perhaps the theory developed in 101 and 102 is solid enough at this point to start looking more deeply into what this would entail. I suspect that you are right in that something like triangulation will emerge, perhaps in various forms.

For instance, right off the bat, triangulating 1st, 2nd, and 3rd person accounts of a situation would be quite central to the routine protocol of such a system that can help connect researchers across psychology and sociology. Perhaps a general example being a triangulation between three of four different classes of interaction that can predict properties of the remaining class.

In the long run a uniform protocol and graduated scale of what "counts" could eventually suggest an ergodic attractor or basin of sorts. Which may go a long way towards setting up an invariant set of constraints from which agent variation across the lifetime (development) can be interpreted and communicated with more rigor. Maybe uniting the social sciences sufficiently for simulations that are compatible with economics yet deeply rooted enough to predict some interesting properties at a vague topological level of analysis. Though, will probably never be "powerful" enough to replace governmental agencies or anything like that.. unless something like westworld season 3 occurs in which an AI conforms people to make them more predictable while using some of the outliers to round up more outliers for indefinite hibernation (dystopia or utopia? if almost everyone does not know the difference?).

There are several short and simple tests that can be given for the various functions in a one-on-one or small group setting to reveal type likelihoods. The simplest perhaps being an Si test of remembering particulars of the recent past (e.g. describe your day, or week..). Where the more specific the description the higher Si is likely to be in the stack. ENxJ's will usually have to think for a while in order to do this (relative to someone with first or second slot Si). Other examples being logical deduction and missing generalization tests for Ti, or inference tests for Ni (e.g. hotdog, flyers, and cleats indicate what activity or setting?). Triangulating Si, Ti, and Ni into Fi while adjusting for IQ, age, culture(s), state, complexity, etc.

(*) A drawing being a static graph over which to draw causal structure which can then suggest a data topology or structure. The data structure linking with various predictions that can be tested through creative experimental design and research funding (or otherwise volunteers in a trusted network).

[daylen]

Of course, it should also be mentioned that higher-dimensional versions of triangulation will likely be required in higher-dimensional spaces of analysis (e.g. quadration). 2 to around 8 degrees being about the likely bounds for -ational(*) analysis due to the increasingly edgy data being generated (i.e. more points are near edges of distortion due to an incomplete theory).

The social science may just have attractors/basins in higher-dimensional spaces and hence we are just now really getting around to it in the last few decades or something.

(*) That is, translational? Perhaps there is a triangulmation and quadramation for transformational movement.

[Ego]

@Ego used the word once here

From an evolutionarily perspective accurate navigation is an important ability. Triangulation works when navigating because it uses fixed and/or known points to determine location. Those points are not variable. The mountain is stable. The paths of the stars, once known, are immutable.

Did the ability to triangulate provide an evolutionary advantage? Did it become one of those innate skills that flourished among those who survived? Is there a portion of the brain that grew larger in those who could navigate well? Did it somehow change from a useful skill to something we must do?

I don't know, but the problem I was highlighting in that thread is that I believe people have a compulsion to triangulate. By that I mean that they have to use the various extremes in their landscape to determine their place and positions in the world. They (we?) can't not do it.

Trouble is, those extremes positions are not fixed. In fact, the positions people use today for triangulation are often manufactured by others with the intention of manipulating their destination.

Many years ago I managed a resort that had a convenience store with a soda fountain. A year after leaving the job I returned to find that the soda company had changed the cup sizes so that the cup that was the largest before I left was now the smallest. The medium size cup, which at the time appeared absurdly large, had not existed the year before. I have not visited the store for years. If I did I would not be surprised to find that the massive medium has now become the smallest.

[daylen]

Trouble is, those extremes positions are not fixed. In fact, the positions people use today for triangulation are often manufactured by others with the intention of manipulating their destination.

Right, as if we try to use what is fixed to draw conclusions (the stars), then anything goes for node interpretation (astrology). Location on earth being one thing, social navigation being another. I suspect conscious manipulation isn't really all that common except in limited cases where the synchronicity of a situation does most of the leg work. Subconscious manipulation is quite readily common in that basically everyone constructs a narrative of reality with them at the center and expects everyone else to conform to it (else they get angry or anxious or whatnot and either stubbornly persist or change their narrative).

In the world of social science and more generally the postmodern era of anything goes, no positions are fixed beyond a narrow slice of measurables in heavily constrained contexts. People do not really know what to think anymore so it is often easier to find a few highly disagreeable people who likely exhibit subconscious manipulation of others in order to get a grip on the social dynamics of the world stage playing out.

This is partially a motivator for me in finding a more solid inference engine for social interaction that just about anyone can understand and doesn't come loaded with ideological bias. This is easier said than done, of course. Typically, people do develop their own models of social interaction quite early, though I often wonder if such will be enough to shield against all the misinformation aka subconscious manipulation going on. Which I am likely doing right now as I string together slightly less variable nodes with slightly more variable nodes.

I encourage everyone to doubt everything I say and push back on what doesn't add up. Otherwise, how would I know if I am going off the rails, ha.

[Ego]

The human compulsion to triangulate evolved together with the compulsion to find order in chaos. Complex religions sprung up to provide what appeared to be solid inference points to help triangulate and navigate the chaos.

Some of the solid inference points they created where accurate. Some played the percentages and were accurate for most of the people, most of the time. Many were obviously wrong but were created with good intentions to explain the unexplainable and to lure people toward goodness and kindness. Others where pure manipulation. All were presented as fixed, unwavering, often impenetrably complex, but perfect.

At this strange moment in history science is killing religion while simultaneously creating technologies that can find actual truths among the chaos. But the technology cannot explain how it determined the truth in a way we are capable of understanding. That impenetrable complexity sets up a modern version of the situation I outlined in the paragraph above.

In this new world our compulsion to triangulate may be our saving grace or the thing that enslaves us.

[daylen]

In this new world our compulsion to triangulate may be our saving grace or the thing that enslaves us.

Agreed. That is why it is so important to make clear that any model built will never be perfect. The map is not the territory. Though, to me it seems like collective suicide to give up mapping territory. I like to keep one foot in doomer optimism and another in technological path dependency.

I would push back a little on scientific truths not being explainable from the algorithms themselves. This would tend to presume that agents cannot become an outer fit where computational procedures speak for themselves (i.e. algorithmic determination or ST). Becoming an inner fit looking out at the vast and complex external world aligning more with NF. This NF-ST divide seems to be quite stark in the world at large right now where faith is split between the internal and external worlds. Some using the inner world to deconstruct algorithmic determination and others using the outer world to deconstruct context.

Healing this dichotomous split in the social fabric would require putting each on equal footing through the flexible drawing of a boundary between the inside and outside with overlaying casual graphs that blur where agency begins and ends.

[Ego]

I would push back a little on scientific truths not being explainable from the algorithms themselves.

An example would be how AI is now able to predict with terrifying (!) accuracy whether a person with a terrible rental history has made the necessary changes in life to ensure that they will be an excellent tenant in the future. It has access to granular spending data, location data, employment data, contacts, sleep details, mood information, schedules and a million other snippets that somehow allow it to make astoundingly accurate decision.

It extracts order from chaos and eliminates my need to look for inference points to help me navigate the decision. It just gives me the decision in the simplest form, yes or no. But it is incapable of explaining exactly how it reached the decision.

Kinda like an all-powerful, all-knowing, unquestionable deity.

[daylen]

AI models are just more sophisticated deterministic algorithms that can be represented on a turing machine which reads and flips bits along a tape. In other words, if enough time was spent understanding how the algorithm was coded, all the magic would disappear. Of course, in practice the algorithm often acts as a black box in which the human interpreter feeds the machine some input and receives output while ignoring the implementation.

A simple example being counting:

i = 0

while True:
i = i + 1
print i

<- that being a high level language of what is actually going on with the turing machine underneath. Within the ST paradigm, the algorithm (no matter how represented) is self-explained. Essentially, a tracking device with the potential to do work (transform something into something else).

[7Wannabe5]

@daylen:

I think this might get a little fuzzy with learning programs with variable inputs. For example, a plant identification app can become much more sophisticated in its output without additional algorithmic programming.And the developer of the app might not be able to determine the means by which the app developed areas of greater sophistication without doing research. For instance, the developer might have no clue that humans who are interested in roses are much more exacting in their identification specifications than humans who are interested in pumpkins, but through the feedback from the rosarians, the apps powers in that realm soon become much more developed, etc. etc. all sorts of chaotic, random or complex stuff in the real world informing an AI through its sensors or user input.

[7Wannabe5]

Or another way to look at it would be that an AI does not have the sort of limits to attention that will cause a human to not see the guy in the gorilla suit if she is concentrating on counting change in the cash register.

[daylen]

@7w5 Right, no disagreement there. Generally, pseudo-randomness can be injected into the determination to the point of no individual human being able to understand what is going on even if they had billions of years. Yet, it is important to keep in mind that the randomness is actually generated from a sensitive mathematical function that pulls from steady input (i.e. time or space or sensors) to output apparent chaos. This apparent chaos perturbing the data space to give emergence to the appearance of learning.

The NF paradigm then being born out of the acceptance of some indeterminacy or randomness for all intensive purposes.

[jacob]

i = 0

while True:
i = i + 1
print i

During the mid 1980s ...

Code: Select all

10 PRINT "JACOB WAS HERE"
20 GOTO 10
RUN

... was common in store display computers. This later devolved into various PEEKs and POKEs and delayed sound sirens

"Kilroy was here" unto new walls. History is written on walls. Yet many don't care to write on walls.

[guitarplayer]

[...]

Maybe uniting the social sciences sufficiently for simulations that are compatible with economics yet deeply rooted enough to predict some interesting properties at a vague topological level of analysis.

[...]

This will not happen on a large scale IIRC my social sciences classes, I attended a nice variety. But we can play here

There are several short and simple tests that can be given for the various functions in a one-on-one or small group setting to reveal type likelihoods.

[...]

(*) A drawing being a static graph over which to draw causal structure which can then suggest a data topology or structure. The data structure linking with various predictions that can be tested through creative experimental design and research funding (or otherwise volunteers in a trusted network).

Yep those tests, developing this seems possible practically. Also, I will be taking a course 'graphs networks and design' for my maths and stats degree in October so I will be able to better converse about graphs soon.

I don't know, but the problem I was highlighting in that thread is that I believe people have a compulsion to triangulate. By that I mean that they have to use the various extremes in their landscape to determine their place and positions in the world. They (we?) can't not do it.

Trouble is, those extremes positions are not fixed. In fact, the positions people use today for triangulation are often manufactured by others with the intention of manipulating their destination.

On some level all the organisms determine their position against something, I don't think there is other way to do it. Moving on to a metaphorical use of the word, the extreme positions / boundary conditions are not fixed because, well, we define the boundaries (or let others do it like you say). This is largely what the ERE book is about I think, redefining boundaries (extreme positions).

In this thread I was thinking about triangulation as a reaffirmation of some thread of thought by looking at it through various lenses. For example, when you have a theory to explain something, the theory might tell you that people will do x, say x and feel x under some circumstances. So you can do qualitative content analysis of a recorded interview / focus group, then you can design a questionnaire and send it across the internet to some people and then you can have an experiment measuring time reaction to some stimulus, and they should all point in the same direction. This is triangulation in social sciences.

In personal life, what Carl Rogers talks about when he talks about congruence, when (1) who you are is (2) who you think you are and (3) who others think you are is a loose version of triangulation in my mind, as in using triangulation as a model.

Some of the solid inference points they created where accurate. Some played the percentages and were accurate for most of the people, most of the time. Many were obviously wrong but were created with good intentions to explain the unexplainable and to lure people toward goodness and kindness. Others where pure manipulation. All were presented as fixed, unwavering, often impenetrably complex, but perfect.

At this strange moment in history science is killing religion while simultaneously creating technologies that can find actual truths among the chaos. But the technology cannot explain how it determined the truth in a way we are capable of understanding. That impenetrable complexity sets up a modern version of the situation I outlined in the paragraph above.

Riding the current forum wave, this is because the Postfaustian and Modern codes have a glitch in that they have no function to deal with relative truths.

Ecclesiastes 1:9
“What has been will be again, / what has been done will be done again; / there is nothing new under the sun.” (pun intended)

Agreed. That is why it is so important to make clear that any model built will never be perfect. The map is not the territory. Though, to me it seems like collective suicide to give up mapping territory. I like to keep one foot in doomer optimism and another in technological path dependency.

I spent so many hours in the past thinking ontologically about the territory. Dealing with people, one can think that maps are the territory that is being mapped. 'your truth', 'my truth'. I have one coworker who sincerely hold that humans have come to the planet earth from outer space. I am not gonna argue with him about this, I am going to navigate this situation. Back to the OP, triangulation would probably help to confirm or revise the coworker's map, if they wanted to apply it.

I think @Ego's compulsion to triangulate was elsewhere described as humans being 'meaning making' systems.

[Ego]

I think @Ego's compulsion to triangulate was elsewhere described as humans being 'meaning making' systems.

I used the word compulsion but that isn't really the right word for it. I think of it like how our body regulates blood pressure. It just kind of happens. We may be able to influence it but it is primarily automatic.

viewtopic.php?t=4323

[daylen]

Within the stack system, outside of deep sleep and flow, [duality and..] triangulation or the higher-order variants are a pulling of acts/things/words from out there and in here (when a self boundary is drawn). These pulling mechanisms being associated with quadra A or Ti/Fe/Si/Ne where agents typed as such would tend to consciously determine the path between instances of deep sleep or flow through triangulation. The opposing quadra of C using Te/Fi/Se/Ni pushes acts/things/words outwards and inwards between instances of deep sleep or flow with triangulation being subconscious. A and C invert each other aka flip the conscious and subconscious. Though, these inverted vectors into the past and future are not perfect, more like at an acute angle. These apparent diversions can re-meet or re-synergize (narrow angle) though mutual understanding at various points in the past and future (this can happen in the 1st, 2nd, or 3rd person perspective).

It is as if "you" start off unfolded and indecisive and then fold "yourself" a bunch of times. These folds will tend to follow patterns that can be pulled to be "manipulated" or played with, or the patterns can keep on involuting fractally as a pushing that tends to synchronize starts and ends between instances of deep sleep or flow.

Or something like that perhaps.

[guitarplayer]

I used the word compulsion but that isn't really the right word for it. I think of it like how our body regulates blood pressure. It just kind of happens. We may be able to influence it but it is primarily automatic.

viewtopic.php?t=4323

I get you, I meant meaning-making in the same sense.

[...] quadra A or [...] quadra of C [...]

Please could you give me a reference to this I don't follow.

[daylen]

High-level overview with some descriptors I think are correlated based on intuition (may turn out to be misleading when abstracting up from functional level): viewtopic.php?p=257230#p257230

Here is how functions align with pushing and pulling. The agent curve/circle is drawn by you to include all that you know or own (in space and/or time and/or dimension X, Y, or Z): viewtopic.php?p=247812#p247812

Excess pull and push correlating to quadra A and quadra C first hand experiences which can be simulated through imagining self or sensorium as mapped onto agent curve (what do you see outwards and inwards? what can be pushed away or pulled in with arms?): viewtopic.php?p=247817#p247817

Pushing being an expansion and pulling being a contraction of self in the mediums that facilitate the transfer of acts, things, or words. Can sorta be thought of as metaphysical breathing.