This implementation of simulated annealing is a very cool use of JavaScript and Canvas. My Mona Lisa is now 94.18% fit and very recognizable (see right).
The image under construction is labelled “Evolving”, and the original idea calls this “putting evolution to the test”. However, this isn’t actually evolution – at least, not as in the biological mechanism which (it is proposed) is responsible for the existence of humans. This mechanism indeed has some of the same features: a starting point, random changes, and a method of selecting some of those changes to be perpetuated. What it does not have, however, is a goal.
In the algorithm, you can compare where you’ve got to now with where you “ought” to be (100% – the real Mona Lisa). But in biological evolution, there is no “ought” – there is just “is”. The evolutionary journey has no signposts and no terminus. Evolution didn’t start with a picture of homo sapiens and, every time there was a mutation in an amoeba, ask “is this any closer?” This is, at best, a simulation of theistic, directed evolution (an unworkable compromise in my view, but that’s a matter for another time). Those who want to avoid what Steven Jay Gould famously called a “divine foot in the door” need to work a bit harder. ;-)
The questions the evolutionary mechanism might have (indirectly) asked are things like “Does this mutation make the animal survive better – reproduce faster or more often, get by on less food or be able to find more of it, etc.?” Things like this are the proposed selection algorithms. But if they are, we’ve ended up in a surprising place. Why is the result of millions of years of such pressures a dominant species which reproduces on average 2.4 times per pair in a 70-year life span, producing children which take an incredibly long time to become independent, has an odd tendency to monogamy, and can’t go without water for more than three days?
Because those are the characteristics that lead to dominance? Monogamy produces more stable societies, etc etc.
Because that’s not how it worked :) The “end result” of billions of years of evolution has given the world a great many species, all of them well-suited to their environment. H. sapiens sapiens appears to stand out because we’re weak, slow, exposed to the elements, and a score of other sub-optimal things.
But that’s because the selection pressure — the tendency of traits to make a creature more or less likely to live long enough to reproduce — tended toward big brains in humanity’s branch of the family tree. Big, strong humans like H. sapiens neanderthalensis weren’t really selected for or against by the world at large — but small, skinny guys could throw their spears as well as the Neanderthals could have (if they’d thought of it) and in the savannah that helped them along.
After a few generations of competition, human females decided they preferred “Brad Pitt in Fight Club” to “Arnold Schwarzenegger in Conan the Barbarian” and the rest is history :)
To look at other things you mentioned, the 2.4 kids deal is a result of society and its formation, not physiology. Heck, take a quick look at the teenaged girls at the mall next time you’re out shopping. Judging by the… plumage… that’s on display, humans are still wired to be ready to reproduce after 13 or 14 years, every year or so. The “idea” is that you’d be done cranking out babies by 25, a few of them would live to adulthood, and you’d croak somewhere around 30.
But we went and invented society and societal norms, and technologies that let us live longer. Now that we can expect to make our mid-70s, popping out podlings before the college years is no longer necessary, and in an effort to protect young ladies from predatory older men, the first years of fertility have been legislatively placed off-limits with (in some cases) exceptions for their peers.
Nature “wants” baby factories. Modern humans want careers, cars, vacations and the occasional kid. These giant brains we’ve evolved over the millennia are able to override the call of the wild.
It’s just one way the evolution thing worked out. Sharks haven’t changed in tens of thousands of years because there’s no room for improvement. Humans haven’t changed in tens of thousands of years because when we need to improve we invent a machine to do it for us.
Jason: you note that there are laws to “protect young ladies from predatory older men”. How, on the basis of this account of human origins, do you morally condemn these “predatory” older men who want to reproduce with young ladies? (Or do you actually disapprove of all age-related sex laws?) Surely that would just be the evolutionary mechanism at work?
Gerv
Gerv: In this context, I am not approving or disapproving; I’m simply stating that they exist. My use of the word “predatory” was meant to capture the mindset of those who first came up with the laws in the early 20th Century, not necessarily reflect my own views. Poor phrasing on my part — I needed a set of scare quotes.
My understanding is that, due to humans in general surviving longer than we do “in the wild”, people saw a large imbalance of power between, say, a 40-year-old man and a 14-year-old girl, to the point that it was assumed the girl couldn’t make an informed decision regarding sex. Like I said, that’s our big brains short-circuiting the natural impulse. Good thing that gray Jell-O is good for coming up with other stuff, isn’t it? :)
So the point I was trying to make is that we’re still set up to crank out multiple offspring, it’s just that the other major thing we’ve acquired over the years is making us suppress the urge to do so.
(And as for the age laws, I can see the need for them, but it needs to be more of a sliding scale in my opinion. A girl who’s 17 and 364 days is only capable of consenting with someone two years her senior (as an example, it varies from state to state) but the next day she’s magically able to consent with anybody? There’s change needed here, I’m just not exactly sure what it would be. And of course the act of trying to get it made would bring the “think of the CHILDREN!” crowd out of the woodwork.)
No, there’s not really a goal to evolution in biology. It’s just random changes (mutation, cross-over, etc…) and heredity (features of the parents are passed to their children). These two things combined with selection pressure (creatures that are better suited to their environment are more likely to reproduce) results in evolution. In other words, over many generations evolution results in creatures that are better and better suited to their environment (assuming the environment isn’t changing too quickly).
How does this compare with the image evolution software? The software represents the genes as the points, colors, and alpha of polygons. The set of genes that is best suited at representing the Mona List is most likely to “reproduce” (spawn more candidate images). The child images have similar features as their parent, even if they’ve mutated. It’s not exactly the same as biological/sexual reproduction because it doesn’t use multiple parents and do things like cross over, but it still gets the job done. The software doesn’t have a goal to re-create the Mona Lisa, but the programmer might.
I like your question at the end, but I don’t find it as surprising. While we tend to notice our differences with other organisms, there are a lot of ways that we’re like every other animal (composed of cells), mammals, primates, etc…
Evolution works on what is best for reproduction of the species. it doesn’t care what _you_ consider ‘better’. ‘Good’ in evolution is whatever gets the job done. ‘Better’ is anything that gets the job done more effectively then what ‘good’ did.
Our big brains negated the many traits we had that where positive for our survival. big muscles, stronger jaws, vitamin c production etc. the problem is those OTHER things all had trade offs. if you produce vitamin c but your diet has a huge proportion of it you can have problems because you need to now process the excess out, our mammal ancestors lived on predominantly fruits, the two ways to go was produce a vitamin c regulating mechanisms or deactivate vitamin c production. Either way was ‘better’ then the way before. We unfortunately lost vitamin C production. thats the way it works but it was still ‘better’ then it was before. those who produced vitamin C where at a disadvantage, a slight one, but a real one none the less. those without it (or a regulating mechanism but that didn’t develop) would be at an advantage, ergo they would reproduce more often, even if only slightly.
Big muscles take more fuel to build and more energy to run, hence if you don’t NEED those big muscles those who have them are at a disadvantage to those who don’t have as big. a slight one, but still a real one. this is a self balancing system. strong males had an advantage…extra strong males had a disadvantage generally and an advantage in specific areas. we see these types of balancing and trade off systems all over the place. we can even predict the RATIOS of the advantages of these systems in the wild by seeing what ratios actually occur (ectomorphs, endomorphs, mesomorphs) etc etc. why do blo-flies live only one day then die? because it assist in reproduction for the young. why do mammals survive longer then there birth patterns? because we nurture our young etc etc. evolution doesn’t just explain things it explains ALL of it. every pip and iota of biology makes sense in relation to evolution. No one has ever presented evidence of something that goes against evolution, only ever a new wrinkle in evolution (like predator prey dynamics, parasite / symbioses production etc).
Heck want to see something amazing? look into walachia bacteria, the little parasite that is the cause of something like 90% of the speciation events in insects. Amazing little bugger, ‘evil’ also. It made insects that have only one gender, species that can’t reproduce if there cured from the bacteria etc etc. nasty little gene fiddler.