Evolutionary Influences of Plastic Behavioral Responses Upon Environmental Challenges in an Adaptive Radiation

Abstract

At the end of the 19th century, the suggestion was made by several scientists, including J. M. Baldwin, that behavioral responses to environmental change could both rescue populations from extinction (Baldwin Effect) and influence the course of subsequent evolution. Here we provide the historical and theoretical background for this argument and offer evidence of the importance of these ideas for understanding how animals (and other organisms that exhibit behavior) will respond to the rapid environmental changes caused by human activity. We offer examples from long-term research on the evolution of behavioral and other phenotypes in the adaptive radiation of the threespine stickleback fish (Gasterosteus aculeatus), a radiation in which it is possible to infer ancestral patterns of behavioral plasticity relative to the post-glacial freshwater radiation in northwestern North America, and to use patterns of parallelism and contemporary evolution to understand adaptive causes of responses to environmental modification. Our work offers insights into the complexity of cognitive responses to environmental change, and into the importance of examining multiple aspects of the phenotype simultaneously, if we are to understand how behavioral shifts contribute to the persistence of populations and to subsequent evolution. We conclude by discussing the origins of apparent novelties induced by environmental shifts, and the importance of accounting for geographic variation within species if we are to accurately anticipate the effects of anthropogenic environmental modification on the persistence and evolution of animals.

Fig. 1

Genetic accommodation represented in norms of reaction. (A) Expression of a plastic trait of an ancestral population (dark squares) in its original environment (E(A)1) and in a novel environment (E(N)2). (B) A shift in responsiveness to phenotype following evolution in the new environment (open squares). (C) A shift in the pattern of plasticity in the derived population (open squares) indicated by differences in ancestral and derived slopes of the reaction norm.

Fig. 2

Ancestral plasticity can facilitate the persistence of populations and their subsequent evolution. (A) Reaction norms of plastic (solid line) and non-plastic (dashed line) populations. Initially two adaptive peaks exist in the environment (heavy lines in panels B–D) and distributions of phenotypes in both populations match one peak (B). In panel C, the environment changes, eliminating that adaptive peak. Only the plastic population persists because it has shifted phenotype to match the second, remaining adaptive peak. In panel D, the change in environment shifts the phenotype of the plastic population to match the second adaptive peak, but leaves both peaks intact. Both populations persist but they diverge evolutionarily as the plastic phenotype subsequently evolves and matches the novel peak. (After Price et al. [2003]).

Fig. 3

The adaptive radiation of the threespine stickleback. The central image represents the oceanic type, and peripheral images represent the freshwater derivatives that comprise the radiation. Dark boxes emphasize populations with the full anterior antipredator complex; dashed boxes are those in which it has been lost. From Bell and Foster (1994), with permission. The full anterior antipredator complex includes a pelvic girdle that supports the paired pelvic spines. The ascending processes of this girdle connect with anterior lateral plates that in turn support the first and second dorsal spines.

Fig. 4

Responses of wild-caught and laboratory-reared threespine stickleback to a simulated attack by a rainbow trout (Oncorhynchus mykiss). Latency to recovery was measured as the time from the attack to the point at which stickleback resumed feeding or exploratory behavior. Three replicate populations were nested within each of the four predation environments, and approximately 30 fish from each population, in each rearing condition (laboratory versus wild) were tested. Mixed model ANOVA on log-transformed data indicated significant differences due to rearing experience (laboratory versus wild: F_1,8 = 5.61, P = 0.045), predation environment (F_3,8 = 7.93, P = 0.009), and due to a population (within predation environment) × rearing interaction (F_8,650 = 6.40, P < 0.001). The significant overall effect of rearing indicates that learning is generally important in the development of antipredator behavior, and the significant population (within predation environment) × rearing interaction indicates possible inter-population variation in the degree to which learning is important. The significant predation environment effect indicates evolutionary (genetic) divergence in behavior. Bars are means ± SE. Figure adapted from Wund et al. (2015).

Fig. 5

Differentiation of phenotypes of benthic and limnetic lacustrine stickleback. Persistence and loss of features are inferred from knowledge of oceanic (ancestral) phenotypes. See text for further discussion (redrawn from Foster [1999]).

Fig. 6

Proportions of courtships incorporating the zig-zag dance and dorsal pricking in seven populations of threespine stickleback. Hatched bars indicate limnetic populations, gray bars oceanic fish, and black bars benthic fish. Bars connected by horizontal lines do not differ significantly (P > 0.05; post-hoc contrasts). Data represent at least 16 males per site, and only one courtship per male (after Foster et al. [1998]).

Fig. 7

The proportion of courtships by males from four Alaskan threespine stickleback populations that incorporated or failed to incorporate the zig-zag dance under field and laboratory environments. Only one courtship bout per male is included. Cannibalistic groups are present in the field in benthic and oceanic populations. Although historically absent from the limnetic population at Lynne Lake, cannibalistic groups, and male diversionary responses, are now common, but the incidence of courtships incorporating the zig-zag dance in the field is unchanged (Chock 2008). Figure adapted from Shaw et al. (2007).