-
Website
http://willwilkinson.net/flybottle -
Original page
http://www.willwilkinson.net/flybottle/2004/11/20/bad-theories-that-track-robust-regularities/ -
Subscribe
All Comments -
Community
-
-
Top Commenters
-
Popular Threads
Gladwell makes fun of it, but MBTI does work.
It strikes me that Gladwell has missed a critical flaw in most applications of the MBTI: obviously the test does not measure personality traits directly, but rather *self-conceptions* of such traits. While self-perception is certainly interesting in its own right, it is likely to be less variable than actual personality traits, and to (in the best case scenario) trail changes by a significant period of time as the psyche comes to acknowledge its own development.
In the worst case, self-perception can be a negative image of the mind's workings. Take Ayn Rand, who gladly accepted the moniker "Ms. Logic", shouted about "reason" all the time, and is always classified as an NTJ. Her revealed personality profile was emotional, intuitive, holistic, and hence far more compatible with F. (One site lists as a characteristic of T: "Accept conflict as a natural, normal part of relationships with people.") And yet it's easier to imagine a 5-dimensional cube than Rand anwsering a typical T/F question on the F side.
It might very well be that most libertarians test as NTs because much of the traditional ethical/political argumentation has been explicitly or implicitly premised on a society made up by such people.
well, what is "G" is an emergent property of transdomain interactions? let me elaborate, if you read the number sense by stanislaus dehaene you will note he reports that some patients with brain damage have lost the capacity for algebra by not geometry or vice versa. obviousy there is no "math module," but various nodes or domains in the brain that work together to give on a capacity for mathematics. G, like mathematical ability, is distributed normally on a bell curve, this suggests that there are various independent factors that contribute to the full expression of the phentoype. these factors in the genetic sense are various genes who come in allelic flavors. in a neurological sense they are cognitive domains.
so yes, the mind is modular, we do have an analog number sense (infants to rats have a conception of bigger and smaller). but the digital mathematical capacity is an emergent property of the number sense being grafted on to language and visuospatial thinking (to name a few factors). the normal distribution is the result of the fact that people have various levels of fluency in the various module capacities.
anyway, that's my theory.
I've never been able to conclusively decide, though, whether I am actually a libertarian or just incredibly paranoid....
New reader, BTW, I think was referred from GapingVoid but that was so many hours ago I've already forgotten.
Now possibly an ENTJ...
The big mistake that Gladwell makes is to assume that MBTI is dichotomous. Anyone who's studied it realizes that people have strong preferences in some areas, and weak in others; some people have well-developed auxiliary functions, and some don't. The 16 types aren't points; they're regions in a four-dimensional space defined by a full spectrum of possibilities for each preference.
Also, there's a reason they're called "preferences." Everyone is spontaneous sometimes; everyone plans sometimes. Most everyone can be extroverted at least on occasion, and most everyone has moments of introversion. Etc. A person's Myers-Briggs type only represents their most common mode of action. So to point out that there are exceptions, as Gladwell does, is to miss the point.
Finally, I also think that the deeper truths of Myers-Briggs types are separable from the MBTI, which is a very crude device. Ranked preference questions or degrees of agreement would be much better methods of determining type, IMHO.
Woohoo!
Goard above not only claims but displays his F. For my money the differences between T and F are the BIG political determinants, God help us when systems are run by F and HR is administered by T.
For practical cooperation, the big split is P and J. Extremes are typical odd-couple stuff, Oscar and Felix, impulsive and anal, EP off the chart on intuition and initiative, IJ dutiful on follow-thru.
This is good stuff, but has to be danced with (now if that's not an ENTP offhand indicator of a statement!)
Myers Briggs increases my predictive capabilities when dealing with others. As far as I am concerned, that means, "Check yer premises."
Before Myers-Briggs, it was all straight-up projecting. Which, of course, had the effect of filtering all but those who shared my pathologies. . . er . . .type.
I know some libertarian S's. Not a pretty sight.
I was struck by how easily it depened on the situation for a lot of the questions, mainly the ones about introverted/extroverted. It mainly has to do with how secure I feel. I'm extremely extroverted in certain interesting class discussions and more introverted in personal relationships. I also do a lot of internal analyzing. So which one am I? Who knows. It isn't that easy.
"G" is very likely just a way of the overall or coarse-grained interaction of all the modules -- in other words, any general intelligence measurement is looking at the emergent effects and overall efficiency of the cooperating and competing modular subunits.
This perspective is influenced greatly by Marvin Minsky's "society of mind" theories, and more recent elaborations of a modular perspective, but I've seen little that would suggest that a monolithic rather than a modular approach to cognition is called for.
Like others here, since I learned a little more about the MBTI, I've found usefully predicatively, so I keep it in my pocket, always mindful of its limitations.
As for g, WW, I can't follow you there, sorry.
No one can describe it - or agree on a description - no one can accurately measure it or duplicate another's measurement, no one can offer a prediction based on it, no one can tie it in any meaningful way to anatomy or brain function - it's clearly phlogiston, a ghost. It doesn't exist.