Tuesday, June 1, 2010

Intelligence, Evolution, and a thought experiement.

Intelligence n. 1: capacity for learning, reasoning, understanding, and similar forms of mental activity; aptitude in grasping truths, relationships, facts, meanings, etc.

If you had a biosphere that lived on prediction where, rather than getting energy from the sun or from other organisms, the energy was gleaned by way of making accurate predictions about some unknown external world. Where there exists some way of confirming or denying these predictions (senses) and survival was predicated on accuracy of predictions. We would have, in our though experiment, a mind.

The capacity for learning new things for making a new species of predictions would be evolution. The capacity for understanding for being capable of making accurate predictions (this is largely what we mean by "understand", "I know why, how, and what it'll do in various situations") would be evolution. The aptitude for grasping things, for finding relationships, facts, meanings, and truths about the world would be, in this hypothetical biosphere, the process by which these organisms adapt to filling these new niches or better filling the niches than other species would be evolution.

* We can map the process of biological evolution on to a hypothetical world of predictions and find something that fits our given understanding of what we mean by intelligence.

* We can further understand that because of fundamental differences between this hypothetical and AI and the like, why such enterprises routinely fail miserably. It's a bit like designing squirrels by hand that could adapt to any situation. That even the best set of expert systems isn't really going to be remotely as powerful as this. And if we need something this powerful then AI in many of it's facets is an attempt to create a Creator God able to make new sets of predictions as they are needed (without really knowing predictions are needed).

*  We can also see that there are underlying evolutionary properties to the way brains grow and work. We can also see that prediction is what the brain actually does. That we constantly make predictions about the world and about what we should see next.

* This would also be able to do many of the amazing things the brain does like getting fooled by optical illusions to hallucinating away our blind spots. We don't predict there's nothing there... rather we predict what we think should be there.

If we take this thought experiment to its logical limits and to the potential complexity and usefulness of the world of nature we see around us thriving on various abstractions of sunlight. We should have little doubt that such a world, taken as a whole, should properly be seen as a mind by simply replacing light with enlightenment as the ultimate food source.

In the distant future I see open fields for far more important researches. Psychology will be based on a new foundation, that of the necessary acquirement of each mental power and capacity by gradation.
-- Charles Darwin, The Origin of Species, 1859, p. 449.


Tracy said...

I like this... it makes a lot sense.

possibly the chicken said...

I completely agree with this. I just didn't know how to put words to it. Although, I must admit I will have to re-read it a couple more times to really grasp... I have said before if it took intelligence to figure it out it must have taken intelligence to make it happen. hmmm but I don't really know if that makes sense or not.

btw... nice blog.

Tatarize said...

Hm. It makes sense, but it's not necessarily true. For example, it takes a lot to figure out fractals but it doesn't imply that there's some sort of fractal creator. There are a number of sort of fractal and chaotic processes that take a lot of work to produce but none of it is necessarily intellectual work, but it does take a heck of a lot of intelligent work to figure them out.