The creative power of randomness.
Even after explaining this, I realize it's a bit too difficult to grasp. In another thread, someone said:
"That random mutations(chance) could drive this is unlikely.There is neither foresight or memory."
The genetic code serves as the memory, of course. But foresight isn't required. And randomness does drive the process. But randomness is not alone.
The evolutionary algorithm in the theory of evolution is an algorithm that's found elsewhere. Scientists use it to design new circuits and new products, for example. It's also at play in information theory, heavily used in computer programming. This book would be interesting:
http://www.amazon.com/Evolutionary-Algo ... B005LQBV4C
To understand how randomness drives the process, I'll lay out a game I once thought up if I were to become a science teacher(I would still like to be). The game would mimic evolution by starting with a species pool of 20 lizards. The lizards would be handmade wooden replicas. It would last most of the school year, as a pet project. The lizards would all start with a single solid color, each one different, and would be in a copse of trees within sight of the student body. The challenge is to pick the first lizard you see, acting as a predator. Pick the easiest one you're able to identify. If there are 20 or more students, then the last lizard to be picked is the parent for the next generation. If there are 20 or less, then the unpicked lizards are the parent population.
I'd guess the students would quickly pick the red/orange/purple/blue lizards, but miss the green and brown lizards - or at least pick them last.
The 'picking of lizards' is the non-random selection process that mimics the environment. Predators are a part of the environment, after all. Once the parent lizards are brought back to class, we produce offspring with similar colors. Perhaps add some detail, but not too much since it must be similar(nearly solid color). If old enough, the students could do this by painting themselves. Or there are computer programs that generate small random changes - you could print the generated paper and glue it onto the wooden lizards.
I'd expect the first generation offspring to be mainly greens and browns with a bit of texture. After the second generation, the ones with more camoflauged patterns(therefore better hidden) would be selected. The mutation parameter would be that you could only use colors 'nearby' on the color wheel. Perhaps some would have a little blue, and some would have a little yellow. That parameter aside, it should be randomized as much as possible. By the fourth and fifth generations, mixtures of browns and greens in different hues would dominate.
After 10 or so attempts, we compare the original population set with the 10th generation set.
Even though this would teach how randomly generated offspring drive the creative process of an evolutionary algorithm, I think creationists would get the wrong idea. "But it's kids doing the picking... therefore the lizards are designed." I think the small, random computer-generated changes might be better in that aspect, so it's not confusing. The randomness is simply that - randomness, whether by random human choice or random algorithmic generation.
What appears to lead to creativity is the selection process. If giraffe's appear to be so well suited to reaching high places that it must be
purposeful, think again. The environment as a selection parameter is
all that's needed. Shortnecks die or are unhealthy(they can't reach the plentiful food source up high), and longnecks are left. Rinse and repeat until necks are very long.
If you follow the link above to the book on evolutionary algorithms, you'll see the power of algorithms in creating new things. A blind mathematical set of steps is all that's needed to create anything and everything under the sun. No intelligence needed. Once the mechanism of this algorithm clicks in your head, and you understand it, the elegance speaks to it's truth. It makes sense on such a basic level. I hope teachers around the globe are able to lead students to that 'aha' moment regarding evolutionary algorithms.
EDIT - my experiment was a bit confusing. I clarified a few things.