1 IntroductionUnderstanding, quantifying, and predicting the ability of organisms to adapt to changing environments is at the core of eco-evolutionary research[1,2]. In the face of unprecedented environmental change, natural populations, especially those with limited mobility, can avoid extinction via phenotypic plasticity and/or adaptive evolution [3]. However, our understanding of the interplay between selection and plasticity in changing environments is surprisingly limited[4–8]. This limitation is not trivial, for plasticity can itself evolve[9], can be adaptive or nonadaptive[10], and has seemingly contradictory effects on adaptive evolution[11], on which we focus here. For decades, researchers have theorized whether plasticity facilitates or hinders adaptive evolution[9,12]; the evidence is contradictory and general patterns have not emerged [5,10,11,13,14].The primary conflicting hypotheses for whether plasticity facilitates or hinders adaptive evolution are:(H1) plasticity weakens directional selection by masking genotypic variation (Bogert Effect [15]), thus slowing the rate of genetic change[5,16–18] vs.(H2) plasticity facilitates evolution by allowing the population to persist under environmental change long enough for genetic change to occur[19–22] (Plasticity-First Hypothesis [21] orBaldwin Effect [19]).This debate remains unresolved, for even when theoretical predictions agree with empirical findings[5,10,11,13,14,23], we lack a general framework to ascertain the context-dependency of the prevalent mechanism. Here, we introduce a framework based on environmental change context, to outline clear null hypotheses for when and how plasticity interacts with directional evolution. We place the plasticity facilitates vs. hinders selection debate on two ends of a continuum, and specify the properties of environmental change–rate of mean change , variability , andtemporal autocorrelation –that influence how plasticity impacts adaptive evolution.The type of environmental change a population experiences can alter its likelihood of adaptation and, ultimately, persistence[24–27]. Studies of demographic[28], genetic[29], and evolutionary rescue[30], show that rate of mean change, variability, and temporal autocorrelation of a population’s selective environment impact population persistence[24,25,29,31–35]. However, because different types of environmental change can have contradictory effects on plasticity and evolution[34,36–38], elucidating these dynamics is not trivial. Consequently, there is an urgent need to place this discussion on the environmental stage in a generalizable way that will allow ecologists and evolutionary biologists to better contextualize, mechanistically understand, predict, and compare their findings.Moving optimum theory links environmental change to the resulting evolutionary responses. Three decades of research on this theory shows that, when a population is confronted with an environment that changes directionally, there is a critical rate of changethat must be matched by change in the mean phenotype of the population, such that the mean remains close to the theoretical phenotypic optimum . In this context, a phenotypic lag between the mean phenotype and the optimum phenotype typically emerges which, if too large, makes extinction certain [39–41]. Evolutionary (e.g. , selection, genetic variation) and ecological processes (e.g. , within-generation life history, plasticity and population dynamics) together influence the limit of how far a population can lag without going extinct. The contribution of plasticity to population persistence and adaptation is largely determined by this phenotypic lag: how much of the short- or long-term lag can be compensated for or even hindered by plasticity?We argue that hypotheses such as the Bogert Effect and the Plasticity-First Hypothesis / Baldwin Effect are not mutually exclusive. Rather, plasticity may facilitate or hinder adaptive evolution depending on the properties of environmental change. To assess the impact of plasticity on the ability of a population to evolutionarily track a changing environmental optimum, we specify the links among the type of environmental change, plasticity, and adaptive evolution by considering several fundamental processes. Thus, we utilize both theoretical and experimental studies to:Assess how three key components of environmental change (rate of mean change , variability , and temporal autocorrelation ) each alter the mechanisms behind phenotypic tracking of a moving optimum ([i] Genetic variation, heritability, and selection , and [ii] life history, plasticity andpopulation dynamics ).Introduce a unified framework of testable hypotheses detailing how those three components of environmental change can influence the relative benefit of plasticity to adaptive evolution.

1. Thermal performance curves (TPCs) are commonly used to forecast species’ responses to temperature change. Recent work has demonstrated that the breadth and shape of a consumer’s TPC change with resource densities, highlighting the potential for inaccurate forecasts if resource densities are not static. In particular, if resource densities decline, the optimal temperature and breadth of thermal performance also declines leading to an enhanced risk of warming, particularly among species that may incur additional costs of behavioral thermoregulation. 2. Here, we investigate the relationship between resource density and temperature (warming) on the persistence of a consumer population which exerts top-down control on its resource via trophic interaction. Trophic coupling generally reduces the potential for resource declines to exacerbate the negative effects of warming on consumers; when warming has negative effects on the consumer, resource densities tend to increase due to a reduction in top-down control. However, if resources are more sensitive to warming (e.g. due to an asymmetry amongst their thermal performance curves), the negative effects of warming on consumers can be exacerbated by declining resources. 3. Our work elucidates the importance of jointly considering temperature and resource limitation when utilizing assessing the thermal performance of species. We demonstrate how knowledge of the thermal performance of a resource population can be used to generate realized consumer thermal performance curves.

1) As temperatures rise across the globe, many species may approach or even surpass their physiological tolerance to withstand high temperatures. Thermal performance curves, which depict how vital rates vary with temperature, are often measured under ideal laboratory conditions and then used to determine the physiological or demographic limits of persistence. However, this approach fails to consider how interactions with other factors (e.g. resources, water availability) may buffer or magnify the effect of temperature change. Recent work has demonstrated that the breadth and shape of a consumer’s thermal performance curve change with resource densities, highlighting the potential for temperature interactions and leading to a potential ‘metabolic meltdown’ when resources decline during warming (Huey and Kingsolver 2019). 2) Here, we further develop the basis for the interaction between temperature and resource density on thermal performance, persistence, and population dynamics by analyzing consumer-resource dynamic models. We find that the coupling of consumer and resource dynamics relaxes the potential for metabolic meltdown because a reduction in top-down control of resources occurs as consumers approach the limits of their thermal niche. However, when both consumers and resources have vital rates that depend on temperature, asymmetry between their responses can generate the necessary conditions for metabolic meltdown. 3) Moreover, we define the concept of a ‘realized’ thermal performance curve that takes into account the dynamic interaction between consumers, resources and temperature, and we describe an important role for this concept moving forward. 4) Synthesis. A better understanding of the link between temperature change, species interactions, and persistence allows us to improve forecasts of community response to climate change. Our work elucidates the importance of thermal asymmetries between interacting species, and resource limitation as a key ingredient underlying realized thermal niches.