Long-term effects of early nutrition and environmental matching on developmental and personality traits in zebra finches

Developmental plasticity is a key feature of many organisms and individuals can benefit from early programming to optimize their phenotypes for the expected environmental conditions. However, environmental conditions may sometimes change unexpectedly. Mismatches between early and adult life, for example, can have important repercussions for adult phenotypes, potentially leading to better performance under matched than mismatched conditions as predicted by the predictive adaptive response hypothesis. We conducted a long-term experimental manipulation of dietary conditions in a population of zebra finches, Taeniopygia guttata. Broods were exposed to two early nutritional treatments until independence and we used a split-brood design to independently manipulate nutritional conditions after independence to create matched and mismatched nutritional environments in later life. Developmental trajectories of all individuals were followed for more than 5 years and we scored behavioural responses in trials in a special environment and while interacting with a special object three times during adult life. Overall, we found no evidence for early programming affecting morphology. Tarsus and wing length were exclusively influenced by the early nutrition. Body weight showed lasting effects of the early treatment and independent effects of nutritional condition during adulthood, but no effects of environmental matching or mismatching. Special-object trials showed effects of the adult nutritional treatment while environmental matching affected hopping activity in special environments. These behavioural responses showed substantial long-term individual stability over a 3-month period and were only marginally smaller when measured over a period of more than 4 years. Interestingly, survival of individuals from low-quality early nutritional condition was higher compared with high-quality early condition individuals, which became evident only after years of survival monitoring. Beyond the nutritional treatment itself, we found sizable brood identity effects that slowly but steadily declined with age, indicating a significant but decaying effect of natural variation in parental provisioning on adult phenotypes.