July 25, 2014

One Evolutionary Trajectory, Many Processes

Two months ago, I wrote a post on how we might think more deeply about human biological variation. This post also involved a discussion about the recent Nicholas Wade book (A Troublesome Inheritance), which has engendered its own rancor on the internet [1]. In the second part of the post, I discussed some potential ways we can more effectively model human variation. This was decidedly exploratory, the intent of which was to follow-up on the discussion. In this post, I will discuss the need for taking cultural evolution and other factors into account when interpreting genetic variation.

To understand where I am coming from here, consider the difference between population genetics and behavioral ecology approaches. In population genetics, the concern is over observing the patterns of standing variation in a population. We discuss topics such as allele frequencies, admixture, and the mechanisms of genetic differentiation. But populations also behave, and in behavioral ecology topics such as sexual selection, foraging patterns, and strategic behaviors are also taken into account.

While there is indeed implicit overlap between these two fields in the literature, there is little direct theoretical synthesis in this direction. For example, if one takes species concepts into account [2], we can see the issue rear its head: one can apply a host of species concepts which explain both the behavioral and genealogical dynamics of a population, but a unified conceptual framework (e.g. one that is not contradictory) is elusive.

Yet even when only taking behavioral dynamics into account, there are a multitude of factors that make direct comparisons between populations difficult. Species differ in both their sociality and acquisition of culture. This differentiation is even more profound in terms of how culture has shaped a species' ability to adaptively radiate and persist over multiple generations. Humans are not only an intensely eusocial species, but also fall into the latter category of being shaped by culture as much as by environmental selection.

One might simply refer to this as "cultural selection", but a better approach is to model the process of genealogical and cultural (or social) evolution as nominally separate but interrelated processes. In "Playing the Long Game of Human Biological Variation", I advocated for the use of dual process models. Such models treat the same population as being subject to two or more distinct processes simultaneously. In a Synthetic Daisies post re-published at Humanity+ [3], I introduced a dual process Artificial Life-based model that integrates genealogical dynamics and biogeographic processes (specifically changes in geomorphology).

There are a good number of examples of dual process models in the literature which integrate cultural and biological evolution. A good starting point is the work of Richerson, Boyd, McElreath, and Henrich [4, 5], who use a dual inheritance model (DIT) with similar genetical and cultural inheritance mechanisms. While this does not distinguish between the mode transmission for genetical units (genealogy) and cultural units (social learning), it does allow for their dynamics to differ within a population.

This provides us with a conceptual expression of "nature" not being equivalent to "nurture", even though we end up in a place similar to the species concept example. But this does not necessarily solve a key issue; namely, that culture and genetics do not simply have the potential to follow divergent trajectories. Culture might also provide a coherent and context-dependent evolutionary constraint [6] which can influence "fast" human evolution [7]. Specifically, culture might influence genetic evolution indirectly through evolutionary constraints (EC) on admixture, migration, local environmental genetic polymorphisms, and demographic fluctuations [8].

One example of a dual process model (in this case, an example from niche construction). COURTESY: Niche Construction page, Semiotics Encyclopedia Online.

Notice that this is quite a bit different than claims of genetics influencing cultural evolution, or culture acting as a multiplier of genetic differences. In fact, the effect is not a feedback or other type of causal mechanism at all, but rather an incongruence [9]. Evolutionary incongruence (EI) occurs when the evolutionary trajectory of the genome and the cultural environment do not lead in the same direction [8].

For example, even though you might possess a genotype that makes you very unfit for a certain environment, possessing a cultural adaptation on top of this genotype might make you fit enough (or even very fit). EI and EC can also determine more general outcomes in a dual process model. In the case of humans, where culture enables humans to survive in environments beyond what is enabled by genes alone, EI is much more dominant than EC. You can still find genetic variants that result from adaptation to a specific local environment, but they are not the determining factor in survival. In a very different context, for example in the case of a solitary species, EC might dominate over EI.

In summary, accounting for variation within and between human groups might best be done using a sophisticated theoretical framework. This framework includes 1) the use of a dual process model that represents cultural and genetic evolutionary processes, and 2) the identification of how culture contributes to genetic variation, namely either through constraint (which enables feedback between genes and culture) or incongruence (where the variation contributed by genetic and cultural evolutionary processes point in different directions). In the case of eusocial species that possess culture (example: Homo sapiens), incoherence will be predominant, although constraint can drive forward local genetic adaptation when needed.

 Examples of eusocial (left) and solitary (right) species.

In a future post, I will explore another theme in the original "Long Game" blog post, namely the idea that panmixia might not be the best way to assess the absence of population subdivision. Instead of using a traditional population genetics model, using scale-free networks to represent the null hypothesis might give us a more profound theory and more realistic results. Look forward to it.

[1] I'm not particularly interested in ideological debates. But be aware that this post will be largely theoretical and perhaps a bit too speculative. That's the way theoretical advances are made!

[2] Wheeler, Q. and Meier, R.   Species Concepts and Phylogenetic Theory: a debate. Columbia University Press (2000).

[3] Alicea, B.   Artificial Life meets Geodynamics (EvoGeo). Humanity+ Magazine, December 7 (2012).

[4] Richerson, P.J. and Boyd, R.   Not By Genes Alone: How Culture Transformed Human Evolution. University of Chicago Press (2005).

[5] McElreath, R. and Henrich, J.   Dual inheritance theory: the evolution of human cultural capacities and cultural evolution. In "Oxford Handbook of Evolutionary Psychology", R. Dunbar and L. Barrett eds., Oxford University Press (2007).

[6] Boyd, R. and Richerson, P.   The cultural transmission of acquired variation: effects on genetic fitness. Journal of Theoretical Biology, 100, 567-596 (1983).

[7]  Hawks, J., Wang, E.T., Cochran, G.M., Harpending, H.C., and Moyzis, R.K.   Recent acceleration of human adaptive evolution. PNAS, 104(52), 20753–20758 (2007).

[8] NOTE: The terms and abbreviations for evolutionary constraint (EC) and evolutionary incongruence (EI) are of my own coinage.

[9] Laland K. and Brown, G.   Sense and Nonsense: Evolutionary Perspectives on Human Behavior. Oxford: Oxford University Press (2002).

1 comment:

  1. Eusocial refers to groups with a reproductive caste and a non-reproductive (worker) caste. Ants, termites, bees, and naked mole rats are eusocial, humans and apes are not. They are simply social.