Origins: Motus Project vs Pod'lair

[Juan E. Sandoval]

Hello,
With the migration from the old forum, much information has been buried which is relevant to the origins of Cognitive Type, and how it came to be designed. I feel this record needs to be restated in clear terms, both to set the record straight and to dispel any misconceptions about its initial construction which people may have, as we move forward into a wider audience.
You will also be able to track my choices along the way, and the deductive path that has lead to where we are now. Thank you all for reading.

August 2012 - Motus Project vs Pod'lair

^ The entirety of the video above is very relevant to how CT began. The first ever rendition of Cognitive Type was a project called Motus. It started around August of 2012, as you can see in this video's upload date. And it was atheoretical and data-driven. The project's ambition was to catalog the entirety of human motions and mannerisms, and then look for self-emerging statistical trends within the data.
First the process began by taking a video footage and cropping it into individual clips, each a few seconds long, and creating hundreds of looping GIFs. Each of these GIFs was called a "bit" and they were stored together in a pile. As the video above says, these bits were then checked against one another for naturally occurring matches, and when matches were found they were grouped together into what we call an "X", which is roughly equivalent to a signal.
Since this approach did not leave out anything, any motion that had multiple instances was grouped this way. This lead to a group of X's, and a pile of bits with no matches. The bits with no matches would await a match to arrive, as we repeated this process with more videos. As we allowed the data to lead the research, we came across correlations between X's as well. These correlations between X's were put under categories called G's (groups).
Some of this information can still be seen using Wayback Machine at: https://web.archive.org/web/20121221105157/http://physiognomy.me/catalog

Back in 2012 me and my small group of researchers understood that nothing could be assumed or taken for granted. This was an agnostic inquiry into human expression and to this day I still think this methodology is the most sound and, if funding permitted, it would be re-continued. It remains the core premise of CT; that an atheoretical, holistic investigation of human body mannerism clustering would yield a non-random distribution of clusters, and this non-random distribution would correlate to cognition based on the theory of embodied cognition. And that this organic clustering can be used to reverse-engineer an understanding of cognition.
However, we ran into labor problems. As you can see in the link provided there were 1,125 bits and 13 x's at that moment. I don't think that was the final number, but somewhere between ~1,125 - 2,000 bits the project became financially unsustainable. The project had no source of funding, and volunteer efforts grew tired. But before getting into what happened next, some context:

Motus Project vs Pod'lair

The Motus Project was a direct response to Pod'lair, and a challenge they made, as you can see by the end of the video above. They were a group of five people on youtube who claimed to have discovered the Rosetta Stone of the Human Soul -- claiming that it could be deciphered via reading physiological cues visible on the face and body.
Pod'lair asserted that Jung was only approximating, poorly, at what Pod'lair had now discovered. And that Jung got some aspects right about mojo (the phenomenon Pod'lair suggests exists, and which roughly correlates to type), but that mojo is not something he properly unearthed. Therefore Pod'lair saw itself as not being a Jungian model, but instead saw that Jung was a rough draft or intimation of Pod'lair. The claims of Pod'lair were of the highest caliber - portrayed as the greatest discovery of humanity.
Almost everyone who saw Pod'lair laughed at their high assertions, their esoteric language, and they were quickly banned from all type communities they joined. They tarnished the name of visual reading beyond recognition. For years to come, visual reading was a very taboo topic in typology circles, with fingers quickly pointing to the memory of Pod'lair as an example of its ridiculousness.
But to me, at the time age 22, I knew that nobody else really had any more claim to truth than Pod'lair and everyone was battling subjectivities. The arguments were narrow, such that disbelieving in Pod'lair's take on types had no more foundational premise than Jung's, Lenore Thompson's, Keirsey's, etc. Science gave no credence to any of them. And I felt that if you laugh at Pod'lair but not at typology more generally, you also have to laugh at yourself, because the entire typology community is swimming in a sea of near-baseless conjectures.
I was not impressed by the rebuttals to Pod'lair, which were mostly ad homs. However unseemly some people might find their presentation, they were making an objective claim. They claimed to be data-driven in their conclusions, and the only way to see if a hypothesis is correct is to actually check the data.
I investigated the data to the extent that I could and found that their methodology ("Five Gears of Mojo Reading") was qualia-based at root. In other words, it resisted strict concretization and mojo was something tacit that is identifiable only from a certain intuitive or artistic lens. After debating with them about how this methodology was fundamentally flawed, we ran into a deadlock that disallowed any further conversation to evolve.
Therefore, the Motus Project was my answer as to how to properly investigate what Pod'lair felt it had arrived at, and the rebuttal I provided to Pod'lair was that because it had not done this meticulous data-construction (such as by bits/x's/g's), it could not claim objective truth value. I asserted that their "data" was not actually data, but was only a list of people in a spreadsheet until they could go through second-by-second and provide the statistics.
Difference
For anyone curious, this is why Motus  (and later CT) was not compatible or synonymous with Pod'lair. There was a fundamental disagreement in methodology that made it so that both projects could not operate together. They could not be equated, even though both were parallel attempts to investigate a given territory. And for years after, we were both aware of each other's projects and had a rather unspoken rivalry between them.

Cognitive Type

Going back to the matter of unsustainable labor, the Motus Project had to be terminated and the group disbanded. An attempt to derive an atheoretical model of human mannerisms from unstructured data proved to be something unfeasible at the time/age. It was too expensive to do. And as it said in the Motus project page:

...the eventual aim is to create a computer algorithm that can read film and catalog thousands of human expressions/motions and aggregate similar motions together, and find what trends exist within the collective human body language.

But the technology also was not ready at the time. A.I. computer vision is only now starting to get nuanced enough to be able to detect the sort of micro-expressions and muscle movements necessary to catalog facial expressions at the accurate, rate and price point required. As of 2020, I believe the premises of CT can now be tested objectively.
However, at the time the decision I made was to go a theoretical route -- using Jung's original descriptions (a very flimsy starting point, methodologically) and seeing if data would naturally cluster around my own interpretation of what he meant (because interpretations were infinite). If it worked, I naturally had in mind to refine the flimsy starting points. The idea would not be to rely on Jung for this, but to see if Jung's descriptions were accurate enough to reveal something organic which could then be defined more precisely and objectively than what he did.
And that's how the present methodology was formed. But the need to divorce taxonomical dependence on Jung's Psychological Types Chapter X was always necessary. So what emerged from the investigation was a visually-defined set of signals that corresponded to a certain psychology. But technically speaking even Cognitive Type is not equivalent to Jung, because what CT means by "Fe/Ti" (and so forth) is a hypothesized visual signal cluster + psychology conjunction that stands apart from any strict or purist interpretation of Jung.
So to make this clear, other names were given: V [Ni/Se], M [Ne/Si], ALE [Fe/Ti], EDA [Te/Fi]. These have remained unused for the past years but they may need to be used more in the future, as it becomes necessary to say that, for example, what we mean by "V" is not 100% equivalent to anything else. And it can be defined very precisely. Precision in terms is imperative when doing research and providing evidence.
In Motus terms, "V" would be an example of a "G" (or group of statistically correlated X's ) - such as raised outer edges, taut eye area, lowered upper eyelids, etc. And that's all. So Cognitive Type will go back to something like the Motus project's methodology eventually, as it approaches rigorous scientific examination. Because if something like type is ever to be discovered as being factually true, it won't survive into science with the same names that we know them by (i.e. "Fe", "Fi"). Even in the best case scenario, if there's something biologically true behind these conjectures, whatever that is will be called something else by science, and be defined with much more robotic taxonomy. It may be that a person's type is later called something like: neural circuits A05 + C33 + 554-1, or NS-A05/C33/554-1.
That is the first and final goal of CT, bringing it back full-circle to its origin. Some people split hairs over Cognitive Type and Socionics or what-have-you, and this-and-that term, but this is missing the bigger picture. The whole point of CT starts here: to question and/or provide evidence that there even are such things as types in the first place, from a biologically objective and non-heuristic angle. Until CT, or Socionics, or anyone, can provide a methodologically acceptable and reproducible experiment for that to be instantiated, one's definitions for function X are meaningless.
And perhaps CT isn't there yet, but so far as I can see, neither is anyone else. But it's only with this goal in mind (to investigate an objective reality, if there is one) that it makes any sense to debate type systems. And unless we're debating with data & methodology, we have nothing, because semantic debates have been happening for a century now and nobody yet has substantiated their claim to types. And that will continue to be the case until raw data is rigorously organized in a way that can reveal what is true.

[fayest42]

Thank you for putting this together. It's really helpful to see the origins of CT. Such a bummer that the original project wasn't feasible. Think of all the things humans could know if only we had unlimited time and resources to gather/analyze data! I do think that starting from an atheoretical viewpoint would be best.
It's interesting to see that you had the idea of using an EEG to see what is actually going on in their brains at the moment they display certain visual signals. I have had the same idea myself! Maybe you could test out the idea by getting an EEG done by Dario Nardi and filming yourself during the experience so that you could view the EEG and your own visual signals simultaneously later. If I ever go out to CA to get an EEG done, I'll let you know 🙂

[Alice]

Wow, this is really incredible and highly informative! Thank you so much for writing this! I really admire the devotion to purity of methodology at every turn and iteration. I've come across and researched a number of typologies at this point, and the assertion that CT isn't at it's ending point yet is the main reason I'm most interested in this theory right now. As far as I have explored, I've never come across another typology that has an end goal, or even another theory that admits it's own incompleteness.

[Author]

Posts