Early-developmental stress, repeatability, and canalization in a suite of physiological and behavioral traits in female zebra finches.

Adaptive developmental plasticity allows individuals experiencing poor environmental conditions in early life to adjust their life-history strategy in order to prioritize short-term fitness benefits and maximize reproductive output in challenging environments. Much research has been conducted to test whether such adoption of a "faster" life-history strategy is accompanied by concordant changes in behavior and physiology, with mixed results. As research in this field has focused on comparison of mean-level responses of treatment groups, few studies include repeated measures of response variables and the effect that developmental stress may have on repeatability per se. We investigated how early-developmental stress affects the mean expression of (and repeatability in) a variety of behavioral and physiological traits in female zebra finches. We predicted that: (1) individuals subjected to nutritional restriction in the nestling phase would have higher feeding and activity rates, with associated increases in hematocrit and basal metabolic rates (BMRs), (2) nutritional restriction in early life would alter adults' stress-induced corticosterone level, and (3) developmental stress would, respectively, influence the amount of among-individual and within-individual variation in behavioral and physiological traits, hence affecting the repeatability of these traits. In comparison to control females, stressed females did not differ in activity rate or stress-induced corticosterone level, but they did have higher levels of feeding, hematocrit, and BMR. Among-individual variance and repeatability were generally higher in stressed females than in controls. Finally, we found that developmental dietary restriction significantly reduced the amount of within-individual variance both in activity rate in the novel environment and in stress-induced corticosterone level. Our results not only confirm previous findings on the effect of early-developmental stress on BMR, but also extend its effect to feeding rate and hematocrit, suggesting that developmental plasticity in these traits is ontogenetically linked. Early-developmental stress may disable particular genetic canalizing processes, which would release cryptic genetic variation and explain why repeatability and among-individual variance were generally higher in the stressed groups than in controls. For activity rate in the novel environment and with stress-induced corticosterone level, however, early-developmental stress significantly reduced within-individual variance, which may be a consequence of increased canalization of these traits at the micro-environmental level.