Dear All,

Many thanks for the helpful suggestions. These seem to fall into several classes:

1) Model is just "too simple". While this is, on one level, completely right, it is perhaps less comforting than it might appear. To follow my own example, I had failed to think about it from the "grass point of view". Obviously the "grass" has an evolutionary incentive not to get eaten to death and may therefore "evolve" seeds hidden deep in the ground and so on. However, this "requires" that the grass is capable of "out evolving" what feeds on it (or else the system collapses from the grass end instead). This doesn't seem to be guaranteed. (And, incidentally, the feeders kill themselves before they wipe out the grass so in one sense there _isn't_ selection pressure on the grass to do this.) This argument also raises a practical problem for ABM. What do we have to assume about any evolutionary system so that it won't automatically break itself? (Does our minimal model really have to include the possibility of predator evolution just to remain stable? Specifically for my model should I just assume that the feeders _can't_ totally denude the grass by handwaving that the grass should "somehow" have evolved its way out of this?)

2) Model is just the Tragedy of the Commons again. Yes but the implications seem to be different. Arguably the TOTC is about the "perils" of selfish rationality which (one presumes) is not an "unavoidable" kind of behaviour and may be moderated by social convention and so on. But here evolution is not an "adopted" mode of behaviour (for which there might be alternatives) but the "baseline mechanism" of the system. Although evolution has no "normative value" you seldom hear about people worrying that evolution might "go wrong" as you do about selfishness/rationality "going wrong". Do we really need social behaviour to "rescue" evolution sometimes? Lucky we came along really ...

3) This is not a "problem" outcome because biological and social systems do sometimes blow themselves up. Which is the best evidenced example of this? In some cases (Easter Island?) it seems to be just one conjecture among many.

4) In another NL model (of cows with varying greediness) what matters is the speed of the cows. If they can get away quickly then society as whole pays for their greed (and everyone suffers). If they are "stuck" on the patches they have denuded then they suffer and greed is selected out. This is again a way of "fixing" the model but seems to put sensible models (and the "effectiveness" of evolution - not in a normative sense but in a sense of not "collapsing") back on a rather contingent footing.

Incidentally the aim of the model is a platform where biological evolution, cultural evolution and social evolution/adaptation can all be shown at work so I don't want the biological evolution to be terribly complicated unless it has to be.

All the best,

Edmund

PS All my agents do is eat, move and breed so about the only other thing I _could_ incorporate into my biological evolution is some sort of "movement gene" (or energetic trade off regarding moving and eating). Hmmmm.

-- 
  Edmund Chattoe-Brown
  [log in to unmask]

########################################################################

To unsubscribe from the SIMSOC list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=SIMSOC&A=1