Wordcels vs Shape Rotators?

Fig 1: schizopost

Fig 2: schizopost

A wordcel refers to a person person who has a wide vocabulary, good reading comprehension, and is good at understanding and dealing with abstract concepts. A shape rotator is somebody who is good at visualization and understanding the concrete world. A wordcel is more likely to fail the “rotate a cow in the mind” test, while a shape rotator is most likely to not have an internal monologue.

The concept of different structures of “intelligence” having differential predictive validity – essential to the dichotomy of wordcel vs rotator – does seem to be true. This is based on a large body of research concerning verbal tilt, where researchers evaluate the verbal skills of a person relative to their other skills (usually math). Emil Kirkegaard (2020) has done a great article about the topic of verbal tilt. Here is a quick rundown of what verbal tilt predicts:

• Creativity
• Left wing views
• Preference for social sciences
• Mental disorders and neuroticism

Overall this seems to be eerily similar to the stereotypical wordcel, which is surprising given that your average twitter buffoon doesn’t know about this research. Chances are they just going off anecdotal pattern recognition and the collective unconscious.

However, I think it is worth noting that wordcels and shape rotators are not the only 2 ways of categorizing people with different ability tilts. This claim is based on the CHC model of the WAIS 4, another factor analysis of multiple intelligence tests, and a correlation matrix of the WAIS 4.

Here is hierarchial factor analysis of 3 different intelligence tests (Johnson, 2004):

Correlation matrix of the WAIS 4:

(Schmank, 2019) I – information, V – vocabulary, C – comprehension, S – similarities, PC – picture completion, BD – block design, FW – figure weights, MR – matrix reasoning, VP – visual puzzles, A – arithmetic, DS – digit span, LN – Letter-Number sequencing, Ca – cancellation, Cd – coding, SS – symbol search

CHC model. (Benson, 2010). Gc – crystalized intelligence, Gc – visual-spatial intelligence, Gf – fluid intelligence, Gsm – short term memory, Gs – processing speed

The “wordcel” component in the CHC model is clearly Gc, crystalized intelligence. The “rotator” component here is Gv, which is the component that involves block design and visual puzzles. Ultimately, there are 3 other factors here which have been ignored – most notably the Gf factor, which has a g-loading of 0.99. Notably the “wordcel” tests and “rotator tests” are still clearly correlated – most expressing a correlation of about 0.53. While spatial and verbal tests tend to share a lot of overlap, intelligence has more dimensionality to it that cannot be ignored.

What these particular subtests measure is skills, not intelligence. You can always read to learn more about the world, you can memorize vocabulary, arithmetic can be practiced, and so on. The point of measuring them all is that they all positively correlate with each other, indicating there is one or many factors that are mediating the relationship between all of them. Almost all factor analysis on IQ testing data yields an enormous general factor which explains about 50% of the variance in the entire matrix of tests (Johnson, 2004). Looking back at that factor analysis, there was a factor which was identical to g, which is Gf. Gf loaded the most on arithmetic, figure weights and matrix reasoning. Here is what the latter tests look like:

Figure Weights

Matrix Reasoning

Mostly pattern recognition tests, consistent with the theory that intelligence is broadly defined as the ability to infer information from other information.

My personal theory on this is that a wordcel is somebody who is disproportionally good at retaining learned information compared to somebody of equal intelligence (gf). This gives them some disproportionate advantages in vocabulary and long term memory, which enables them to become better writers and speakers than your average grug.

This comes at a cost: retaining all of this information increases the amount of internal instability they experience. This instability consists in uncertainty about their own knowledge and an increased amount of information they parse through on a daily basis. Postmodernism and antipositivism are the most fundamental manifestation of wordcelery, as the wordcel has parsed through so much information that he can no longer trust what he concludes from it. This is also consistent with the observation that people who are verbally tilted tend to be mentally ill.

To sum it all up, the concept of verbal tilt does seem to be scientifically validated as far as predictive validity goes. However, when it comes to performance in GCSEs or jobs verbal tilt doesn’t mean anything. Hierarchical factor analysis does seem to generate a verbal and spatial component, but it also generates 3 more components besides that. One factor – “fluid intelligence”, is identical to g.

OC

Edit 1- corrected verbal tilt definition and view of wordcel vs shape rotator

Edit 2 – minor rewriting, improved theory of wordcelery\

Edit 3 – corrected correlates of verbal tilt, misread Emil’s article