Saturday, April 14, 2007

Meta Evolution

For quite a while I have been wondering whether evolution itself evolves, and how it might evolve.

Evolution is a way for lineages of organisms to adapt to their environments and search for better solutions to the problems which they encounter.

[Plants have found the golden ratio for example, which mathematically maximises the amount of sunlight they are exposed to. And needless to say most animals are mathematical and engineering feats of excellence in their own rights]

I also wonder whether organisms search for these solutions differently, and whether slowly but surely their evolutionary searches improve.

I read recently (Climbing Mount Improbable, Richard Dawkins) that there are areas of evolutionary search which are restricted by organisms. I.e. parts of search space are excluded a priori by the organism, because those parts would probably lead to unfruitful findings. E.g. mammals more often than not, do not evolve asymmetrical features, rather they evolve in a symmetrical fashion.

That idea is fascinating. Organisms contain information which tells them where it's best not to bother looking for good solutions to their problems, thereby focusing on areas where more fruitful solutions can be found.

I wonder how these evolution restrictions are found? And whether there is any fundamental rhyme or reason behind them?

4 comments:

Nadia said...

this is so interesting. it has also got me thinking about using those terms in other situations, like the evolution of movements and being equipped to understand what will work and what won't.

Unknown said...

thx for dropping by nadia.

it is interesting.

on reflection i think organisms don't have to contain info which tells them where best to "search" - but the information is already contained in the population itself.

so athe chances that human offspring would have 3 legs are low, because 3 legged humans are rare in the population and therefore possible mates.

and the reason why 3 legged humans are rare, because any attempts to go in that direction were not that beneficial. so future searches do rely upon historical information to some extent.

get me? hope so...

Anonymous said...

Interesting article but the title suggested the evolution of the evolutionnary process as a general process, not how organisms "apply" it. Anyway I wonder how organism acquire this memory of events (DNA ? culture ? ...? ). In the exemple you give about excluded solution space this is interesting but also dangerous to a certain extent as the environment can change "back" (identical key factors) to a situation that appeared before but was then rejected and consequently creating an excluded solution space.
T=Time, S=Solution space
T1 -> S1 S2 S3, S1 doesnt work and is excluded, S2 work and is explored in S2'
T2 -> -S1 S2 S2' S3
...
Tn -> -S1 -S2 S2' S2" S3 S4 -S5
Tn+1 -> with key conditions getting close to T1, ... well ok it should work actually ;)

Could you please post some references to paper discussing evolution "strategy" ?

Unknown said...

"Interesting article but the title suggested the evolution of the evolutionnary process as a general process, not how organisms "apply" it."

I agree, but i realised that evolution is the organisms application of it -- you can't detach evolution from organisms, can u? (btw, i am no expert, just my fanciful thinking)

"Anyway I wonder how organism acquire this memory of events (DNA ? culture ? ...? )."

I would say their genes, and perhaps memes if u want to go in that direction.

"In the exemple you give about excluded solution space this is interesting but also dangerous to a certain extent as the environment can change "back" (identical key factors) to a situation that appeared before but was then rejected and consequently creating an excluded solution space."

Agreed, and that's what happens with high selection pressure. A good argument against racist super race ideologies.

"Could you please post some references to paper discussing evolution "strategy" ?"

Well Axelrod, has done some nice research - he's the only one that comes to mind - as i said, am not an expert :)