Entrapment in Local Optimum
Evolution sometimes gets stuck on solutions that are locally but not globally optimal. A locally optimal solution is one where any small change would make the solution worse, even if some big changes might make it better.
Being trapped in a local optimum is especially likely to account for failure to evolve polygenic traits that are adaptive only once fully developed and incur a fitness penalty in their intermediary stages of evolution. In some cases, the evolution of such traits may require an improbable coincidence of several simultaneous mutations that might simply not have occurred among our finite number of ancestors. A crafty genetic engineer may be able to solve some of the problems that were intractable to blind evolution. A human engineer can think backward, starting with a goal in mind, working out what genetic modifications are necessary for its attainment.
The human appendix, a vestigial remnant of the cecum in other mammals, while having some limited immunological function (Fisher 2000), easily becomes infected. In the natural state, appendicitis is a life-threatening condition and is especially likely to occur at a young age. There is also evidence that surgical removal of the appendix reduces the risk of ulcerative colitis (Koutroubakis and Vlachonikolis 2000; Andersson et al. 2001). It appears that removal of the appendix would have increased fitness in the EEA. However, a smaller appendix increases the risk of appendicitis. Carriers of genes predisposing for small appendices have higher risks of appendicitis than noncarriers and, presumably, lower fitness (Nesse and Williams 1998). Therefore, unless evolution could find a way of doing away with the appendix entirely in one fell swoop, it might be unable to get rid of the organ; whence, it remains. An intervention that safely and conveniently removed it might be an enhancement, increasing both fitness and quality of life.
Another source of evolutionary lock-in is antagonistic pleiotropy, referring to a situation in which a gene affects multiple traits in both beneficial and harmful ways. If one trait is strongly fitness-increasing and the other mildly fitness-decreasing, the overall effect is positive selection for the gene (Leroi et al. 2005). One example is the e4 allele of apolipoprotein E. Having one or two copies of the allele increases the risk of Alzheimer disease in middle age but lowers the incidence of childhood diarrhea and may protect cognitive development (Oria et al. 2005). Antagonistic pleiotropy has also been discussed in relation to theories of aging. The local optimum here is to retain the genes in question, but the global optimum would be to eliminate the antagonistic pleiotropy by evolving genes that specifically produced the beneficial traits without detrimental effects on other traits. Over longer timescales, evolution usually gets around antagonistic pleiotropy, for example, by evolving modifier genes that counteract the negative effects (Hammerstein 1996), but such developments can take a long time, and in the meanwhile, a species remains trapped in a local optimum.
Yet another way in which evolution can get locked into a suboptimal state is exemplified by the phenomenon of heterozygote advantage. This refers to the common situation where individuals who are heterozygous for a particular gene
(i.e., have two different alleles of that gene) have an advantage over homozygote individuals (who have two identical copies of the gene). Heterozygote advantage is responsible for many cases of potentially harmful genes being maintained at a finite frequency in a population.
The classic example of heterozygote advantage is sickle-cell gene, where homozygote individuals suffer anemia while heterozygote individuals benefit from improved malaria resistance (Allison 1954; Cavalli-Sforza and Bodmer 1999). Heterozygotes have greater fitness than both types of homozygote (those lacking the sickle-cell allele and those having two copies of it). Balancing selection preserves the sickle-cell gene in populations (at a frequency that varies geographically with the prevalence of malaria). The “optimum” that evolution selects is one in which, by chance, some individuals will be born homozygous for the gene, resulting in sicklecell anemia, a potentially fatal blood disease. The “ideal optimum”—everybody being heterozygous for the gene—is unattainable by natural selection because of Mendelian inheritance, which gives each child born to heterozygote parents a 25% chance of being born homozygous for the sickle-cell allele.
Heterozygote advantage suggests an obvious enhancement opportunity. If possible, the variant allele could be removed and its gene product administered as medication. Alternatively, genetic screening could be used to guarantee heterozygosity, enabling us to reach the ideal optimum that eluded natural selection.
The phenomenon of heterozygote advantage points to potential enhancements beyond reducing susceptibility to diseases such as malaria and sickle-cell anemia. For instance, there is some indirect evidence that at least Type I Gaucher’s disease (and possibly other sphingolipid storage diseases) is linked to improved cognition, given the significantly higher proportion of sufferers in occupations correlated with high IQ (Cochran et al. 2006). This, and other circumstantial evidence, is used by the authors of the cited study to argue that heterozygote advantage can explain the high IQ test scores and the high prevalence of Type I Gaucher’s disease among Ashkenazi Jews. Should this prediction be borne out by finding an IQ advantage for heterozygote carriers of the diseases, it would suggest that screening to promote heterozygosity, or genetic interventions to induce it, would be viable forms of cognition enhancement that meet the EOC.
One other kind of evolutionary entrapment is worth noting here, that of an evolutionary stable strategy (ESS), “a strategy such that, if all the members of a population adopt it, no mutant strategy can invade” (Smith 1982). One way in which a species can become trapped in an ESS is through sexual selection. In order to be successful at wooing peahens, peacocks have to produce extravagant tails which serve to advertise the male’s genetic quality. Only healthy peacocks can afford to produce and carry top-notch tails. It is adaptive for peahens to prefer to mate with peacocks that sport an impressive tail, and given this fact, it is also adaptive for peacocks to invest heavily in their plumage. It is likely that the species would have been better off if it had evolved some less costly way for males to signal fitness. Yet no individual peacock or peahen is able to defect from the ESS without thereby removing themselves from the gene pool. If there had been a United Nations of the peafowl, through which the birds could have adopted a coordinated millennium plan to overcome their species’ vanity, the peacocks would surely soon be wearing a more casual outfit.
The concept of an ESS can be generalized to that of an evolutionary stable state. A population is said to be in an evolutionary stable state if its genetic composition is restored by selection after a disturbance, provided the disturbance is not too large (ibid). Such a population can be genetically monomorphic or polymorphic. Thus, while ESS refers to a specific strategy that is stable if everybody adopts it, an evolutionary stable state can encompass a set of strategies whose distribution is stable under small perturbations. It has been suggested that the human population has been in a stable state in the EEA with regard to sociopathy, which can be seen as a defector strategy which can prosper when it is rare but becomes maladaptive when it is more common (Mealey 1995).
Another way in which evolution can fail to produce solutions that are fitness- maximizing for organisms is intragenomic conflict, in which phenomena such as meiotic drive, transposons, homing endonuclease genes, B-chromosomes, and plasmids result from natural selection among lower-level units such as individual genes (Burt and Trivers 2006). In cases where we can identify intragenomic conflict as responsible for a suboptimal outcome, there is an opportunity for enhancement that can meet the EOC (provided we have the technological means to make the requisite interventions). Genes or traits that would not have evolved, or which would not have been stable against intragenomic competition, could be inserted, possibly, supported by interventions removing some of the competing genetic elements.