The World Is a Free Lunch

One of the strangest Darwinian models to be put forth recently has to be a paper by James V. Stone (a psychologist at Sheffield U, UK), published in PLoS Computational Biology.¹  He basically says that evolution is a free lunch.  Brains and whole body types can emerge if an organism can learn parts of adaptive tasks, provided the rest of the task has enough built-in substeps that will fall into line.  Despite his whiz-bang mathematics, does he prove the point that you can get something from nothing?  He says so.
    Some lines from his paper qualify for Stupid Evolution Quote of the Week.  Here’s a “Well, duh” line in the Author’s Summary:

Some behaviours are purely innate (e.g., blinking), whereas other, “apparently innate,” behaviours require a degree of learning to refine them into a useful skill (e.g., nest building).  In terms of biological fitness, it matters how quickly such learning occurs, because time spent learning is time spent not eating, or time spent being eaten, both of which reduce fitness….

He continues with a synopsis of his thesis:

Using artificial neural networks as model organisms, it is proven that it is possible for an organism to be born with a set of “primed” connections which guarantee that learning part of a skill induces automatic learning of other skill components, an effect known as free-lunch learning (FLL).  Critically, this effect depends on the assumption that associations are stored as distributed representations.  Using a genetic algorithm, it is shown that primed organisms can evolve within 30 generations.  This has three important consequences.  First, primed organisms learn quickly, which increases their fitness.  Second, the presence of FLL effectively accelerates the rate of evolution, for both learned and innate skill components.  Third, FLL can accelerate the rate at which learned behaviours become innate.  These findings suggest that species may depend on the presence of distributed representations to ensure rapid evolution of adaptive behaviours.

Stone attempts to prove his thesis with various computer runs that measure “fitness” and penalize errors.  According to William Dembski’s book No Free Lunch,² however, introducing a fitness function sneaks information into the back door that cannot be assumed in any realistic Darwinian scheme, because the essence of Darwinism is that evolution be undirected and purposeless.  With that restraint, there is no free lunch: any evolutionary algorithm is indistinguishable from blind search.
    This difficulty seems to have been lost on Dr. Stone, however.  In his final paragraph he waved his FLL (free-lunch-learning) as a possible solution to one of Darwinism’s biggest problems, the Cambrian explosion:

It has been demonstrated that FLL accelerates the evolution of behaviours in neural network models.  Given that FLL appears to be a fundamental property of distributed representations, and given the reliance of neuronal systems on distributed representations, FLL-induced behaviours may constitute a significant component of apparently innate behaviours (e.g., nest-building).  Results presented here suggest that any organism that did not take advantage of such a fundamental and ubiquitous effect would be at a selective disadvantage.  Finally, if FLL accelerates evolution in the natural world, then it may have been involved in the Cambrian explosion, an explosion that began when brains (and therefore learning) first appeared.

An astonishing claim by any standard: you can get a brain as a free lunch.

¹James V. Stone, “Distributed Representations Accelerate Evolution of Adaptive Behaviours,” Public Library of Science: Computational Biology, Aug 3, 2007, doi:10.1371/journal.pcbi.0030147.
²See ARN for info on this book.

Don’t be fooled by the math.  This is absolutely insane.  To think that inventing a term like FLL will generate brains and whole new body plans in nature is wonderfully weird.  Apparently Dr. Stone needs to learn some physics, particularly thermodynamics and information theory.  Even Julie Andrews could help: “Nothing comes from nothing; nothing ever could.”
    No amount of math can make up for bad assumptions.  Anybody can rig a model or a computer program to produce any kind of output he wants.  Set the starting conditions, then force the virtual organisms to evolve the way you want them to by rewarding the efforts you (as an intelligent human) desire, and they will obey your every bidding.  GIGO.  Don’t try this out on trilobites.
    Good grief, people, how did it come to this?  This is as silly as pouring sea monkeys into a test tube, adding water, and saying you have created life.  Psychologist, heal thyself.  Go get a real job, Stone, before you embarrass your colleagues.  The ID people are already laughing out loud.  If, on the other hand, this was an ornate prank to show the gullibility of journal editors, we forgive and encourage you.