Wednesday 16 April 2014

The cognitive genome - the genome may be able to learn and direct evolution


Just passing on the fruits of a recent conversation with a knowledgeable scientist whom I respect and trust - about a major piece of work he is doing.

There is a very interesting and plausible new perspective emerging (with evidence from multiple places, and nobody yet having combined it) that the genome may be regarded as having 'cognitive' properties, due to the systematic inter-communication of genes - and that this is consistent with the ability to direct genetic change (mutations and other changes) in an adaptive direction.

This would be a major revision to 'the central dogma' of molecular biology (that information only flows from nucleic acids to proteins), and the standard description of natural selection as being based upon undirected (so-called 'random') mutations - the new idea is that (presumably as a consequence of natural selection) the genome functions  somewhat like a 'brain' that models the environment, and responds by changing itself in a directed (and 'purposive') fashion that would be expected to enhance reproductive success.

This would mean that natural selection would not have to wait for undirected ('random') mutations to generate adaptive changes to the organism; but instead (or as well) the mutation process would itself be manipulated to make adaptive changes much more probable. The genome would itself be able to accelerate and direct evolutionary change. 

Having heard the evidence, and knowing about systems theory as an explanatory model, I find this plausible as a possible biological mechanism. It may emerge as a better - more comprehensive - over-arching theory (paradigm) than the one we have now - or a significant supplement to it.

The question is - even if true - how important it is - and whether it has played a central role in evolution; or just applies to unusual and specific situations.



Bruce B. said...

I’ve always imagined theistic evolution as God using a (at least partially) stochastic process to do the work that fundamentalist Christianity imagines as a purely deterministic process. Lawrence Auster always hated this idea when I presented it to him.
The idea you describe here seems consistent with theistic evolution. God gave us a “second brain” to push the general direction of things.

Bruce Charlton said...

@BB - The biological explanation would either be that the cognitive genome itself evolved by natural selection - or else was some kind of self-organizing/ strange attractor/ chaotic phenomenon.

Biology - as a subject, and like all sciences - excludes theistic explanations.

It does not (or should not) assume that theistic explanations are untrue, it does not disprove theistic explanations - instead they are simply left-out as a prior (simplifying) principle.

Bruce B. said...

I don’t understand genetics at all but his idea is fascinating to me. I know there are artificial intelligence computer programs and “agent-based models” that work through probabilistic “weighting” of behaviors, algorithms, etc.

Anonymous said...

Bruce B:
Lawrence Auster always hated this idea when I presented it to him.

Hah, I had the same experience! I discussed it with him at VFR in this thread in 2003 (I am "Matt").

Bruce B. said...

Matt, you seemed to be making the same argument to him that I did. Here at work, we have written monte-carlo simulations that arrive at an answer that could have been obtained with a closed-form solution (i.e. in a deterministic manner). In our case, we used the stochastic approach to save time. The solution is approximate, that is, there’s what I’d call “residual randomness” in the answer. I assume this residual randomness has some connection to our free will. Someone at VFR described this idea much better than I can but I can’t find the comment right now.
LA just kept repeating to me the same thing he wrote to you that it couldn’t, by definition, be random. I liked his writing very much, but he was a very stubborn man.

Bruce B. said...

Matt, I reread part of that VFR thread. You and LA were talking past each other. He was describing (critically) randomness as the “basis” (his words) and you were describing randomness as an “input” to the model.
God’s algorithms could just as easily be thought of as the “basis” and the random number generator could be thought of as one (possibly of many) input to the model.
Of course, these things we are describing are imperfect analogies.
I have lots of questions about this topic. For example, I assume the creation of man was an act of God’s positive will, but maybe some of our variation/ individuality (the residual randomness) is God’s permissive will?

Anonymous said...

Bruce B:
He was describing (critically) randomness as the “basis” (his words) and you were describing randomness as an “input” to the model.

Right, which is all that randomness really is in the neodarwinian synthesis. Where the neodarwinian synthesis goes off the rails isn't in taking randomness as (one) input: it is in taking natural selection (whatever it means at any given moment in any given argument) to be the primary algorithm determining outputs.

By refusing to distinguish between randomness at an input and purposelessness as a basis he was basically buying into the same confusion as the New Atheists. I think I tried a few more times, just as unsuccessfully, to get him to see this. I miss the stubborn fellow.

I have lots of questions about this topic.

Some things are always going to be a bit mysterious, I think, at least on this side of the Valley of the Shadow: like how is it that we are ourselves a causative power (free will)? And most especially how is it that we are a causative power in the creation of other, new human beings, who themselves become causative powers like us?