Count data can be analyzed using generalized linear mixed models when observations are correlated in ways that require random effects. However, count data are often zero-inflated, containing more zeros than would be expected from the typical error distributions. We present a new package, glmmTMB, and compare it to other R packages that fit zero-inflated mixed models. The glmmTMB package fits many types of GLMMs and extensions, including models with continuously distributed responses, but here we focus on count responses. glmmTMB is faster than glmmADMB, MCMCglmm, and brms, and more flexible than INLA and mgcv for zero-inflated modeling. One unique feature of glmmTMB (among packages that fit zero-inflated mixed models) is its ability to estimate the Conway-Maxwell-Poisson distribution parameterized by the mean. Overall, its most appealing features for new users may be the combination of speed, flexibility, and its interface's similarity to lme4.

Organisms interact with their environments in various ways. We present a conceptual framework that distinguishes three mechanisms of organism–environment interaction. We call these NC3 mechanisms: niche construction, in which individuals make changes to the environment; niche choice, in which individuals select an environment; and niche conformance, in which individuals adjust their phenotypes in response to the environment. Each of these individual-level mechanisms affects an individual's phenotype–environment match, its fitness, and its individualized niche, defined in terms of the environmental conditions under which the individual can survive and reproduce. Our framework identifies how individuals alter the selective regimes that they and other organisms experience. It also places clear emphasis on individual differences and construes niche construction and other processes as evolved mechanisms. The NC3 mechanism framework therefore helps to integrate population-level and individual-level research.

Summary Biologists are increasingly interested in decomposing trait dynamics into underlying processes, such as evolution, plasticity and demography. Four important frameworks that allow for such a decomposition are the quantitative genetic animal model (AM), the ‘Geber’ method (GM), the age‐structured Price equation (APE) and the integral projection model (IPM). However, as these frameworks have largely been developed independently, they differ in the assumptions they make, the data they require, as well as their outcomes and interpretation. Here, we evaluate how each framework decomposes trait dynamics into underlying processes. To do so, we apply them to simulated data for a hypothetical animal population. Individual body size was affected by, among others, genes, maternal effects and food intake. We simulated scenarios with and without selection on body size and with high and low heritability. The APE and IPM provided similar results, as did the AM and GM, with important differences between the former and the latter. All frameworks detected positive contributions of selection in the high but not in the low selection scenarios. However, only the AM and GM distinguished between the high and low heritability scenarios. Furthermore, the AM and GM revealed a high contribution of plasticity. The APE and IPM attributed most of the change in body size to ontogenetic growth and inheritance, where the latter captures the combined effects of plasticity, maternal effects and heritability. We show how these apparent discrepancies are mostly due to differences in aims and definitions. For example, the APE and IPM capture selection, whereas the AM and GM focus on the response to selection. Furthermore, the frameworks differ in the processes that are ascribed to plasticity and in how they take into account demography. We conclude that no single framework provides the ‘true’ contributions of evolution, plasticity and demography. Instead, different research questions require different frameworks. A thorough understanding of the different definitions of their components is necessary for selecting the most appropriate framework for the question at hand and for making biologically meaningful inferences. This work thus supports both future analysis and the careful interpretation of existing work.

1.Large-scale fluctuations in abundance are a common feature of small mammal populations and have been the subject of extensive research. These demographic fluctuations are often associated with concurrent changes in the average body mass of individuals, sometimes referred to as the "Chitty effect". Despite the long-standing recognition of this phenomenon, an empirical investigation of the underlying coupled dynamics of body mass and population growth has been lacking. 2.Using long-term life-history data combined with a trait-based demographic approach, we examined the relationship between body mass and demography in a small mammal population that exhibits non-cyclic, large-scale fluctuations in abundance. We used data from the male segment of a 25-year study of the monogamous prairie vole, Microtus ochrogaster, in Illinois, USA. Specifically, we investigated how trait-demography relationships and trait distributions changed between different phases of population fluctuations, and the consequences of these changes for both trait and population dynamics. 3.We observed phase-specific changes in male adult body mass distribution in this population of prairie voles. Our analyses revealed that these changes were driven by variation in ontogenetic growth, rather than selection acting on the trait. The resulting changes in body mass influenced most life-history processes, and these effects varied among phases of population fluctuation. However, these changes did not propagate to affect the population growth rate due to the small effect of body mass on vital rates, compared to the overall differences in vital rates between phases. The increase phase of the fluctuations was initiated by enhanced survival, particularly of juveniles, and fecundity whereas the decline phase was driven by an overall reduction in fecundity, survival and maturation rates. 4.Our study provides empirical support, as well as a potential mechanism, underlying the observed trait changes accompanying population fluctuations. Body size dynamics and population fluctuations resulted from different life-history processes. Therefore, we conclude that body size dynamics in our population do not drive the observed population dynamics. This more in-depth understanding of different components of small mammal population fluctuations will help us to better identify the mechanistic drivers of this interesting phenomenon. Accepted ArticleThis article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Accepted ArticleThis article is protected by copyright. All rights reserved. 2. Using long-term life-history data combined with a trait-based demographic approach, we examined the relationship between body mass and demography in a small mammal population that exhibits non-cyclic, large-scale fluctuations in abundance.We used data from the male segment of a 25-year study of ...