Guest post: Evolution is the control of development by ecology
Today's guest blogger: Willem Frankenhuis
Today’s post is written by Willem Frankenhuis. It’s the second in a series of guests posts. – Rob K.
Ted Kennedy stated: “Some men see things as they are and say why – I dream things that never were and say why not” … at the funeral of his brother Bobby. As evolutionary psychologists, we often observe phenotypes (e.g., behaviors) and attempt to explain why they have the design features they do. However, rarely do we imagine non-existing designs and ask, why not?
The goal of evolutionary psychology is to document human nature—the universal architecture of the human mind—and to explain its form in light of evolution. In realizing this goal, we often focus on the design features of mature minds, and in principle, there’s nothing wrong with this focus—however, it’s incomplete. All phenotypes are products of development, so natural selection alters phenotypes by modifying developmental systems—the array of processes that constructs phenotypes (Barrett, 2007; Frankenhuis & Panchanathan, 2011; Tooby et al., 2003). Therefore, we also need to focus on the design features of developmental systems.
Some biologists claim their field is undergoing a change of perspective from the “evolution of phenotypes” toward an “evolution of ontogenies” (Fusco, 2001). Others note this view has long been central to the field (de Jong & Crozier, 2003). In the early 1970s, Leigh van Valen remarked: “Evolution is the control of development by ecology.” Still, there is a difference between acknowledging a claim, and actively incorporating it into research. I would like to share a paper discussing how a focus on development can enrich the study of adaptation. The paper is titled “The developmental renaissance in adaptationism” (written by Mark Olson), and currently “in press” at Trends in Ecology and Evolution.
Olson’s article poses the question: When we observe only a subset of all imaginable phenotypes, how do we know whether this subset resulted from natural selection (i.e., the culling of available designs) or developmental constraints (i.e., certain designs are not developmentally possible), or both? Olson describes three ways of finding out: embryology, artificial selection, and the comparative method.
You may know the joke: “There is no natural selection—just species that Chuck Norris has allowed to live.”
Since Darwin and Mayr, we all know that species have no essences, and are better thought of as “clouds in phenotype space.” Still, we can study species-typical mechanisms, or the typical distribution of phenotypic variation across ecologies. Some embryologists, however, are explicitly interested in the full array of phenotypes produced in development, including in rare phenotypes, which are developmentally possible, but disfavored by selection. These phenotypes are rare, in all likelihood, because Chuck Norris did not allow most of them to live.
Also, if ethics permit it, embryologists can use artificial selection to drag individuals to regions of phenotype space that they do not normally occupy, and assess the fitness of these variants in the wild. If artificial selection can produce a phenotype, natural selection might, too, given the right conditions (e.g., long enough time, large populations, etc.). So, artificially generable phenotypes that are not observed in nature may have been culled by selection. Studies showing that artificially produced phenotypes actually attain lower fitness than existing designs constitute even better evidence for natural selection.
The comparative strategy documents the ways that multiple species fill phenotype space to see which parts are filled. On this view, points in phenotype space are species, not individuals, and empty spaces are considered suggestive of developmental constraints. Of course, empty spaces could result from natural selection, too, so we should be cautious in drawing conclusions. However, convincing examples exist. For instance, apparently, none of the over 1000 species of geophilomorph centipedes has an even number of segments (Minelli, 2009), a restriction that seems to result from an ontogenetic segmentation mechanism precluding even numbers—unless, in these animals (but not in others), Chuck Norris has a distaste for even numbers.
Olson spends less time discussing a fourth strategy for distinguishing adaptations from constraints: modeling. Adaptations can be identified when a trait accommodates a presumed function “with sufficient precision, economy, [and] efficiency” (Williams, 1966, p. 10). To understand why regions of phenotype space may be empty, we can build models exploring the design features of developmental systems that natural selection might favor, given particular ecologies (Frankenhuis & Panchanathan, 2011). When systems observed in nature match the model’s predictions, this supports adaptation. If they deviate, this could indicate a developmental constraint (or the model may be mistaken about what selection is maximizing). Optimization models may thus facilitate discovery of constraints by describing how a system ought to behave, if unconstrained, serving as a benchmark.
I’ll end with a quote by two of the leaders in our field, Martin Daly and Margo Wilson:
“Sociobiologists have not generally paid much attention to developmental processes. Ontogeny is often lumped together with immediate causation as “proximate explanation” and scorned as merely “descriptive” in contrast to the more “explanatory” analysis of “ultimate causation”. However, attention to developmental processes seems to us essential. A life history “strategy” is a developmental chronology, which can never be fully understood without attention to the causal processes in ontogeny. Moreover, any attempt to model “strategic” behavior … inevitably raises the question of the organism’s capacity to acquire, process, and retain experientially given information; thus, sociobiologists cannot ignore analyses of the learning process. Evolution may be portrayed as a phylogenetic series of adult phenotypes, but “phenotype” as it is usually understood is an impoverished conception of the organism, a static cross-section at one stage of the life history; evolution itself must be intelligible as a modification of developmental process over generations” (1983, p. 244).
References
Barrett, H. C. (2007). Development as the target of evolution: A computational approach to developmental systems. In S. Gangestad, & J. Simpson (Eds.), The evolution of mind: Fundamental questions and controversies (pp. 186–192). New York: Guilford.
Daly, M., & Wilson, M. (1983). Sex, evolution and behavior, 2nd ed. Belmont CA: Wadsworth.
De Jong, G., & Crozier, R. H. (2003). A flexible theory of evolution. Nature, 424, 16–17.
Frankenhuis, W. E., & Panchanathan, K. (2011). Balancing sampling and specialization: An adaptationist model of incremental development. Proceedings of the Royal Society B, 278, 3558–3565.
Fusco, G. (2001). How many processes are responsible for phenotypic evolution? Evolution and Development, 3, 279–286.
Minelli, A. (2009). Forms of becoming: The evolutionary biology of development. Princeton University Press
Olson, M. E. (in press). The developmental renaissance in adaptationism. Trends in Ecology and Evolution.
Tooby, J., Cosmides, L., & Barrett, H. C. (2003). The second law of thermodynamics is the first law of psychology – Evolutionary developmental psychology and the theory of tandem, coordinated inheritances: Comment on Lickliter and Honeycutt (2003). Psychological Bulletin, 129, 858–865.
Van Valen, L. (1973). Festschrift. Science, 180, 488.
Williams, G. C. (1966). Adaptation and natural selection. Princeton, NJ: Princeton University Press.
Copyright Willem Frankenhuis, All Rights Reserved
One Comment