« The Honor of a Lady | Gene Expression Front Page | Immigration @ Reason.com »
September 14, 2004

The problem with cooperative analysis

One of Robin Dunbar's key points about human social intelligence is that we have an especially advanced Theory of Mind (ToM). ToM is basically the ability to conceive of what someone else is thinking about, that is, "I believe you believe I believe...." and so on. Another term for this is intensionality. Humans are usually thought to be able to keep track of 6 orders of intensionality. Often, autistic people don't have this capacity, which is why they act like jack-asses all the time without meaning to. What comes naturally for a normal human is something they have to learn (as if it's math).

ToM works pretty well in social situations. Yes, there are misunderstandings, but usually they can be corrected later on down the road, and other cues (mannerisms, gestures, etc.) give you a pretty good idea what someone is meaning to say. But one problem that I think people often have is they try to extend this ToM to topics that are removed from the social world, especially in abstract fields.

Abstract concepts like math are integrated together so that axioms and propositions can create a whole formal system. Math is explicit in its assumptions. As you move away from math things become squishier, and I believe in areas like politics, philosophy or art, what we mean outpaces our ability to capture the nuance in language. This is compounded by the problem that everyone is interpreting what others say and formulating answers based on what they believe someone is trying to say. Words and their meanings do not always map clearly. Removed from the concrete world of social relations, which is information rich and providing us with constant input to correct for inaccuarcies in our mental models, this sort of conversation eventually devolves into semantics and the quest to communicate precisely becomes more important than the quest to communicate anything of substance.

As I noted above, individuals who are autistic lack a fully fleshed out capacity for intensionality. Some have noted that these individuals, who in their high functioning form are often labelled as having Asperger's Syndrome, often do very well in analytic/abstract disciplines. The argument often goes that in these fields the need for precise explication means that they are not disadvantaged, but perhaps their tendency to never assume something about the mental state of others, that is, what they "really mean," is one of their advantages.

Posted by razib at 04:09 PM