Re[2]: COMP/NEURO: Images Extracted from Cat Brain

Matt Gingell (mjg223@is7.nyu.edu)
Mon, 11 Oct 1999 00:40:12 -0400

> I would like your interpretation of the evident universality of some form
> of the "Law of Contradiction", "The Law of the Excluded Middle" the
> processes of "induction and axiomatic inference", interpretive prototypes
> like causality or conditionality, or evident perceptual stereotypes like
> "time" "quantity" and "space/size/distance". Not to mention the perva-
> sive use of hierarchical categorical structures.

If two individuals working separately on the same problem in mathematics discovered isomorphic solutions, would that be surprising? Would it imply that the mechanisms they used to deduce the proof are universal to all humans, or would it just mean that it was a natural and correct solution that any thinker might come to? What if a computer discovered the same proof by brute force?

I view concepts like causality, duration, weight, etc. in the same way. They don't have to be hard coded - they are learned because they are so useful under the universal physical constraints imposed upon all human beings. In natural language, for instance, I don't think the fact every human language has an analogy to verbs and nouns proves that either has a universal correlate in the mind's language facility. It could simply be that we all live in a world that is well described by nouns and verbs and the each mind observes that fact and takes advantage of it.

To draw another analogy, what is to be inferred from the fact the the eye has evolved independently in insects and mammals? It's not that evolution is predisposed to producing eyes, rather an eye is such a fantastically useful survival tool that it is selected for in unrelated evolutionary threads.

In my view, are thoughts are similar because we each develop concepts to take advantage of some of the same cognitive niches. We can communicate because our formulations are constrained into some degree of functional isomorphism. The implementation though might be profoundly different.

I'm not implying that some concepts, like 'natural number,' etc., aren't implemented in hardware. But if so, I view that as a product of evolutionary happenstance and optimization.

-matt