ARTICLE IN PRESS UNCORRECTED PROOF /pscriptc:/sco4/jobs2/ELSEVIER/bio/week.41/Pbio2103.001.001 Tue Oct 30 10:29:13 2001 Page T BioSystems 000 (2001) 000 – 000 Deep insight from simple models of evolution Hans-Paul Schwefel * Department of Computer Science, Uniersity of Dortmund, D-44221 Dortmund, Germany Received 9 May 2001; received in revised form 2 July 2001; accepted 3 July 2001 Abstract On one hand, people admire the often strikingly efficient results of organic evolution. On the other hand, however, they decry mutation and selection to be a rather prodigal, inefficient trial-and-error strategy. Taking into account the parallel information processing in a heterogeneous population and sexual propagation with recombination, as well as the genetic control of the reproduction accuracy, computer simulated evolution reveals a couple of interesting, sometimes surprising, properties of nature’s learning-by-doing algorithm. Surial of the fittest, Spencer’s re ´sume ´ of Darwin’s view, turns out to be a bad advice if taken literally. Individual death, forgetting, and even regression show up to be necessary ingredients of the life game. Whether the process should be named gradualistic or punctualistic, is a matter of the observer’s point of view. © 2001 Published by Elsevier Science Ireland Ltd. Keywords: Evolutionary algorithms; Hillclimbing; Survival of the fittest; Auto-adaptation; Internal model of the environment; Requisite variety; Groping in the dark www.elsevier.com/locate/biosystems 1. Introduction Evolution can be looked at from a large variety of positions. Beginning with the closest physico- analytic viewpoint, one might focus attention on the molecular and cellular processes. A more dis- tant point of view centers on the behavior of populations and species. Another difference emerges from whether one emphasizes the homeo- static aspect of some kind of adaptation to a given environment, which is more relevant in the short term, or the macroscopic view of a develop- ment to the more complex, sometimes called higher, in the long term. Not many dare to use the euphemistic word better here, and some even do not like at all the term progress in connection with organic evolution. But how does such an attitude match with the silent presupposition that homo sapiens is the crown of creation and imitating a single brain of that species could help in solving most difficult real-world problems? Despite the fact that pioneers like Conrad (see Fogel et al., 2001) or Wilson (1998) have already worked on much more sophisticated models of evolution, the following considerations are based on very simple mathematical and computational models. The instruments used here will be a macroscope and a time accelerator. Moreover, for methodological reasons, an optimistic point of view will be shared by comparing macro-evolu- tion with iterative optimization, or, more cau- * Tel.: +49-231-9700-951; fax: +49-231-9700-959. E-mail address: hps@udo.edu (H.-P. Schwefel). 0303-2647/01/$ - see front matter © 2001 Published by Elsevier Science Ireland Ltd. PII:S0303-2647(01)00186-1