Search

Learning is a taxonomically widespread process by which animals change their behavioural responses to stimuli as a result of experience. In this way, it plays a crucial role in the development of individual behaviour and underpins substantial phenotypic variation within populations. Nevertheless, the impact of learning in social contexts on evolutionary change is not well understood. Here, we develop game theoretical models of competition for resources in small groups (e.g. producer-scrounger and hawk-dove games) in which actions are controlled by reinforcement learning and show that biases in the subjective valuation of different actions readily evolve. Moreover, in many cases, the convergence stable levels of bias exist at fitness minima and therefore lead to disruptive selection on learning rules and, potentially, to the evolution of genetic polymorphisms. Thus, we show how reinforcement learning in social contexts can be a driver of evolutionary diversification. In addition, we consider the evolution of ability in our games, showing that learning can also drive disruptive selection on the ability to perform a task.

Empiricists often struggle to apply game theory models to real-life cases of animal cooperation. One reason is that many examples of cooperation occur in stable groups, where individuals form social bonds that influence exchanges of help in ways that are not well described by previous models, including the extent of reciprocity and how relationships are initiated. We present a game theory model exploring the conditions under which social bonds between group members promote cooperation. In the model, bonds build up from exchanges of help in a similar way as the strength of association increases in learning, as in the Rescorla-Wagner rule. The bonds in turn affect partner choice and influence helping amounts. The model has a mechanism of reciprocity for bonded pairs, which can evolve toward either loose or strict reciprocation. Several aspects of the model are inspired by observations of food sharing in vampire bats. We find that small social neighborhoods are required for the evolutionary stability of helping, either as small group sizes, or if bonded members of larger groups can form temporary (daily) smaller groupings. The costs of helping need to be fairly low, while the benefits can be substantial. The form of reciprocity that evolves is neither immediate nor very strict. Individuals in need request help based on bond strength, but there is also an evolved preference for initiating bonds with new group members. In contrast, if different groups come into temporary contact, the evolved tendency is to avoid forming bonds between groups., Competing Interests: Competing interests statement:The authors declare no competing interest.

Cognitive flexibility can enhance the ability to adjust to changing environments. Here, we use learning simulations to investigate the possible advantages of flexible learning in volatile (changing) environments. We compare two established learning mechanisms, one with constant learning rates and one with rates that adjust to volatility. We study an ecologically relevant case of volatility, based on observations of developing cleaner fish Labroides dimidiatus that experience a transition from a simpler to a more complex foraging environment. There are other similar transitions in nature, such as migrating to a new and different habitat. We also examine two traditional approaches to volatile environments in experimental psychology and behavioral ecology: reversal learning, and learning set formation (consisting of a sequence of different discrimination tasks). These provide experimental measures of cognitive flexibility. Concerning transitions to a complex world, we show that both constant and flexible learning rates perform well, losing only a small proportion of available rewards in the period after a transition, but flexible rates perform better than constant rates. For reversal learning, flexible rates improve the performance with each successive reversal because of increasing learning rates, but this does not happen for constant rates. For learning set formation, we find no improvement in performance with successive shifts to new stimuli to discriminate for either flexible or constant learning rates. Flexible learning rates might thus explain increasing performance in reversal learning but not in learning set formation, and this can shed light on the nature of cognitive flexibility in a given system., Competing Interests: The authors declare no conflict of interest., (© The Author(s) 2023. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology.)

Game theory in biology gained prominence 50 years ago, when Maynard Smith & Price formulated the concept of an evolutionarily stable strategy (ESS). Their aim was to explain why conflicts between animals of the same species usually are of a 'limited war' type, not causing serious injury. They emphasized that game theory is an alternative to previous ideas about group selection, which were used by ethologists to explain limited aggression. Subsequently, the ESS concept was applied to many phenomena with frequency dependence in the evolutionary success of strategies, including sex allocation, alternative mating types, contest behaviour and signalling, cooperation, and parental care. Both the analyses of signalling and cooperation were inspired by similar problems in economics and attracted much attention in biology. Here we give a perspective on which of the ambitions in the field have been achieved, with a focus on contest behaviour and cooperation. We evaluate whether the game-theoretical study of the evolution of cooperation has measured up to expectations in explaining the behaviour of non-human animals. We also point to potentially fruitful directions for the field, and emphasize the importance of incorporating realistic behavioural mechanisms into models. This article is part of the theme issue 'Half a century of evolutionary games: a synthesis of theory, application and future directions'.

Social hierarchies are often found in group-living animals. The hierarchy position can influence reproductive success (RS), with a skew towards high-ranking individuals. The amount of aggression in social dominance varies greatly, both between species and between males and females within species. Using game theory we study this variation by taking into account the degree to which reproductive competition in a social group is mainly local to the group, emphasizing within-group relative RS, or global to a larger population, emphasizing an individual's absolute RS. Our model is similar to recent approaches in that reinforcement learning is used as a behavioural mechanism allowing social-hierarchy formation. We test two hypotheses. The first is that local competition should favour the evolution of mating or foraging interference, and thus of reproductive skew. Second, decreases in reproductive output caused by an individual's accumulated fighting damage, such as reduced parenting ability, will favour less intense aggression but should have little influence on reproductive skew. From individual-based simulations of the evolution of social dominance and interference, we find support for both hypotheses. We discuss to what extent our results can explain observed sex differences in reproductive skew and social dominance behaviour.

Interactions in social groups can promote behavioural specialization. One way this can happen is when individuals engage in activities with two behavioural options and learn which option to choose. We analyse interactions in groups where individuals learn from playing games with two actions and negatively frequency-dependent payoffs, such as producer-scrounger, caller-satellite, or hawk-dove games. Group members are placed in social networks, characterized by the group size and the number of neighbours to interact with, ranging from just a few neighbours to interactions between all group members. The networks we analyse include ring lattices and the much-studied small-world networks. By implementing two basic reinforcement-learning approaches, action-value learning and actor-critic learning, in different games, we find that individuals often show behavioural specialization. Specialization develops more rapidly when there are few neighbours in a network and when learning rates are high. There can be learned specialization also with many neighbours, but we show that, for action-value learning, behavioural consistency over time is higher with a smaller number of neighbours. We conclude that frequency-dependent competition for resources is a main driver of specialization. We discuss our theoretical results in relation to experimental and field observations of behavioural specialization in social situations.

Variation in stress responses has been investigated in relation to environmental factors, species ecology, life history and fitness. Moreover, mechanistic studies have unravelled molecular mechanisms of how acute and chronic stress responses cause physiological impacts ('damage'), and how this damage can be repaired. However, it is not yet understood how the fitness effects of damage and repair influence stress response evolution. Here we study the evolution of hormone levels as a function of stressor occurrence, damage and the efficiency of repair. We hypothesise that the evolution of stress responses depends on the fitness consequences of damage and the ability to repair that damage. To obtain some general insights, we model a simplified scenario in which an organism repeatedly encounters a stressor with a certain frequency and predictability (temporal autocorrelation). The organism can defend itself by mounting a stress response (elevated hormone level), but this causes damage that takes time to repair. We identify optimal strategies in this scenario and then investigate how those strategies respond to acute and chronic exposures to the stressor. We find that for higher repair rates, baseline and peak hormone levels are higher. This typically means that the organism experiences higher levels of damage, which it can afford because that damage is repaired more quickly, but for very high repair rates the damage does not build up. With increasing predictability of the stressor, stress responses are sustained for longer, because the animal expects the stressor to persist, and thus damage builds up. This can result in very high (and potentially fatal) levels of damage when organisms are exposed to chronic stressors to which they are not evolutionarily adapted. Overall, our results highlight that at least three factors need to be considered jointly to advance our understanding of how stress physiology has evolved: (i) temporal dynamics of stressor occurrence; (ii) relative mortality risk imposed by the stressor itself versus damage caused by the stress response; and (iii) the efficiency of repair mechanisms., (Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.)

Social hierarchies are often found in group-living animals and can be formed through pairwise aggressive interactions. The dominance rank can influence reproductive success (RS) with a skew towards high-ranking individuals. Using game theory, we investigate how the opportunity for differently ranked individuals to achieve RS influences the costs of hierarchy formation and the strength of winner and loser effects. In our model, individuals adjust their aggressive and submissive behaviour towards others through reinforcement learning. The learning is based on rewards and penalties, which depend on relative fighting ability. From individual-based simulations, we determine evolutionary equilibria of traits such as learning rates. We examine situations that differ in the extent of monopolisation of contested RS by dominants and in the proportion of total RS that is contested. The model implements two kinds of fighting costs: a decrease in effective fighting ability from damage (loss of condition) and a risk of mortality that increases with the total accumulated damage. Either of these costs can limit the amount of fighting. We find that individuals form stable dominance hierarchies, with a positive correlation between dominance position and fighting ability. The accumulated costs differ between dominance positions, with the highest costs paid by low or intermediately ranked individuals. Costs tend to be higher in high-skew situations. We identify a 'stay-in, opt-out' syndrome, comprising a range from weaker (stay-in) to stronger (opt-out) winner-loser effects. We interpret the opt-out phenotype to be favoured by selection on lower ranked individuals to opt out of contests over social dominance, because it is more pronounced when more of the total RS is uncontested. We discuss our results in relation to field and experimental observations and argue that there is a need for empirical investigation of the behaviour and reproductive success of lower ranked individuals., (© 2022 The Authors. Journal of Animal Ecology published by John Wiley & Sons Ltd on behalf of British Ecological Society.)

Most analyses of the origins of cultural evolution focus on when and where social learning prevails over individual learning, overlooking the fact that there are other developmental inputs that influence phenotypic fit to the selective environment. This raises the question of how the presence of other cue 'channels' affects the scope for social learning. Here, we present a model that considers the simultaneous evolution of (i) multiple forms of social learning (involving vertical or horizontal learning based on either prestige or conformity biases) within the broader context of other evolving inputs on phenotype determination, including (ii) heritable epigenetic factors, (iii) individual learning, (iv) environmental and cascading maternal effects, (v) conservative bet-hedging, and (vi) genetic cues. In fluctuating environments that are autocorrelated (and hence predictable), we find that social learning from members of the same generation (horizontal social learning) explains the large majority of phenotypic variation, whereas other cues are much less important. Moreover, social learning based on prestige biases typically prevails in positively autocorrelated environments, whereas conformity biases prevail in negatively autocorrelated environments. Only when environments are unpredictable or horizontal social learning is characterized by an intrinsically low information content, other cues such as conservative bet-hedging or vertical prestige biases prevail. This article is part of the theme issue 'Foundations of cultural evolution'.

AbstractGroups of social animals are often organized into dominance hierarchies that are formed through pairwise interactions. There is much experimental data on hierarchies, examining such things as winner, loser, and bystander effects, as well as the linearity and replicability of hierarchies, but there is a lack evolutionary analyses of these basic observations. Here I present a game theory model of hierarchy formation in which individuals adjust their aggressive behavior toward other group members through reinforcement learning. Individual traits such as the tendency to generalize learning between interactions with different individuals, the rate of learning, and the initial tendency to be aggressive are genetically determined and can be tuned by evolution. I find that evolution favors individuals with high social competence, making use of individual recognition, bystander observational learning, and, to a limited extent, generalizing learned behavior between opponents when adjusting their behavior toward other group members. The results are in qualitative agreement with experimental data, for instance, in finding weaker winner effects compared to loser effects.

We focus on learning during development in a group of individuals that play a competitive game with each other. The game has two actions and there is negative frequency dependence. We define the distribution of actions by group members to be an equilibrium configuration if no individual can improve its payoff by unilaterally changing its action. We show that at this equilibrium, one action is preferred in the sense that those taking the preferred action have a higher payoff than those taking the other, more prosocial, action. We explore the consequences of a simple 'unbiased' reinforcement learning rule during development, showing that groups reach an approximate equilibrium distribution, so that some achieve a higher payoff than others. Because there is learning, an individual's behaviour can influence the future behaviour of others. We show that, as a consequence, there is the potential for an individual to exploit others by influencing them to be the ones to take the non-preferred action. Using an evolutionary simulation, we show that population members can avoid being exploited by over-valuing rewards obtained from the preferred option during learning, an example of a bias that is 'rational'., Competing Interests: The authors have declared that no competing interests exist.

The ability to dominate conspecifics and thereby gain access to resources depends on a number of traits and skills. Experience of dominance relationships during development is a potential source of learning such skills. We here study the importance of social experience, aggressiveness and morphological traits for competitiveness in social interactions (contest success) in male domestic fowl ( Gallus gallus domesticus ). We let males grow up either as a single (dominant) male or as an intermediately ranked male in a group of males, and measured their success in duels against different opponents. We found that single-raised males had lower contest success than group-raised males, and that aggression and comb size correlated positively with contest success. This indicates that experience of dominance interactions with other males increases future success in duels. We similarly studied the consequences of growing up as a dominant or subordinate in a pair of males, finding no statistically significant effect of the dominance position on contest success. Finally, we found that males were consistent over time in contest success. We conclude that social experience increases contest success in male domestic fowl, but that certain behavioural and morphological characteristics have an equal or even stronger covariation with contest success., (© 2021 The Authors.)

All organisms have a stress response system to cope with environmental threats, yet its precise form varies hugely within and across individuals, populations, and species. While the physiological mechanisms are increasingly understood, how stress responses have evolved remains elusive. Here, we show that important insights can be gained from models that incorporate physiological mechanisms within an evolutionary optimality analysis (the 'evo-mecho' approach). Our approach reveals environmental predictability and physiological constraints as key factors shaping stress response evolution, generating testable predictions about variation across species and contexts. We call for an integrated research programme combining theory, experimental evolution, and comparative analysis to advance scientific understanding of how this core physiological system has evolved., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Learning is an adaptation that allows individuals to respond to environmental stimuli in ways that improve their reproductive outcomes. The degree of sophistication in learning mechanisms potentially explains variation in behavioral responses. Here, we present a model of learning that is inspired by documented intra- and interspecific variation in the performance of a simultaneous two-choice task, the biological market task. The task presents a problem that cleaner fish often face in nature: choosing between two client types, one that is willing to wait for inspection and one that may leave if ignored. The cleaner's choice hence influences the future availability of clients (i.e., it influences food availability). We show that learning the preference that maximizes food intake requires subjects to represent in their memory different combinations of pairs of client types rather than just individual client types. In addition, subjects need to account for future consequences of actions, either by estimating expected long-term reward or by experiencing a client leaving as a penalty (negative reward). Finally, learning is influenced by the absolute and relative abundance of client types. Thus, cognitive mechanisms and ecological conditions jointly explain intra- and interspecific variation in the ability to learn the adaptive response.

In social interactions, including cooperation and conflict, individuals can adjust their behaviour over the shorter term through learning within a generation, and natural selection can change behaviour over the longer term of many generations. Here we investigate the evolution of cognitive bias by individuals investing into a project that delivers joint benefits. For members of a group that learn how much to invest using the costs and benefits they experience in repeated interactions, we show that overestimation of the cost of investing can evolve. The bias causes individuals to invest less into the project. Our explanation is that learning responds to immediate rather than longer-term rewards. There are thus cognitive limitations in learning, which can be seen as bounded rationality. Over a time horizon of several rounds of interaction, individuals respond to each other's investments, for instance by partially compensating for another's shortfall. However, learning individuals fail to strategically take into account that social partners respond in this way. Learning instead converges to a one-shot Nash equilibrium of a game with perceived rewards as payoffs. Evolution of bias can then compensate for the cognitive limitations of learning.

The relationship between animal cognition and consistent among-individual behavioral differences (i.e., behavioral types, animal personality, or coping styles), has recently received increased research attention. Focus has mainly been on linking different behavioral types to performance in learning tasks. It has been suggested that behavioral differences could influence also how individuals use previously learnt information to generalize about new stimuli with similar properties. Nonetheless, this has rarely been empirically tested. Here, we therefore explore the possibility that individual variation in generalization is related to variation in behavioral types in red junglefowl chicks ( Gallus gallus ). We show that more behaviorally flexible chicks have a stronger preference for a novel stimulus that is intermediate between 2 learnt positive stimuli compared to more inflexible chicks. Thus, more flexible and inflexible chicks differ in how they generalize. Further, behavioral flexibility correlates with fearfulness, suggesting a coping style, supporting that variation in generalization is related to variation in behavioral types. How individuals generalize affects decision making and responses to novel situations or objects, and can thus have a broad influence on the life of an individual. Our results add to the growing body of evidence linking cognition to consistent behavioral differences., (© The Author(s) 2019. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology.)

Genetic polymorphism can contribute to local adaptation in heterogeneous habitats, for instance, as a single locus with alleles adapted to different habitats. Phenotypic plasticity can also contribute to trait variation across habitats, through developmental responses to habitat-specific cues. We show that the genetic architecture of genetically polymorphic and plasticity loci may influence the balance between local adaptation and phenotypic plasticity. These effects of genetic architecture are instances of ecological genetic conflict. A reduced effective migration rate for genes tightly linked to a genetic polymorphism provides an explanation for the effects, and they can occur both for a single trait and for a syndrome of coadapted traits. Using individual-based simulations and numerical analysis, we investigate how among-habitat genetic polymorphism and phenotypic plasticity depend on genetic architecture. We also study the evolution of genetic architecture itself, in the form of rates of recombination between genetically polymorphic loci and plasticity loci. Our main result is that for plasticity genes that are unlinked to loci with between-habitat genetic polymorphism, the slope of a reaction norm is steeper in comparison with the slope favored by plasticity genes that are tightly linked to genes for local adaptation.

The importance of receiver biases in shaping the evolution of many signalling systems is widely acknowledged. Here, we show that receiver bias can explain which traits evolve to become warning signals. For warning coloration, a generalization bias for a signalling trait can result from predators learning to discriminate unprofitable from profitable prey. However, because the colour patterns of prey are complex traits with multiple components, it is crucial to understand which of the many aspects of prey appearance evolve into signals. We provide experimental evidence that the more salient differences in prey traits give rise to greater generalization bias, corresponding to stronger selection towards trait exaggeration. Our results are based on experiments with domestic chickens as predators in a Skinner-box-like setting, and imply that the difference in appearance between profitable and unprofitable prey that is most rapidly learnt produces the greatest generalization bias. As a consequence, certain salient traits of unprofitable prey are selected towards exaggeration to even higher salience, driving the evolution of warning coloration. This general idea may also help to explain the evolution of many other striking signalling traits found in nature., (© 2018 The Author(s).)

Batesian mimicry evolution involves an initial major mutation that produces a rough resemblance to the model, followed by smaller improving changes. To examine the learning psychology of this process, we applied established ideas about mimicry in Papilio polyxenes asterius of the model Battus philenor. We performed experiments with wild birds as predators and butterfly wings as semiartificial prey. Wings of hybrids of P. p. asterius and Papilio machaon were used to approximate the first mutant, with melanism as the hypothesized first mimetic trait. Based on previous results about learning psychology and imperfect mimicry, we predicted that: melanism should have high salience (i.e., being noticeable and prominent), meaning that predators readily discriminate a melanistic mutant from appearances similar to P. machaon; the difference between the first mutant and the model should have intermediate salience to allow further improvement of mimicry; and the final difference in appearance between P. p. asterius and B. philenor should have very low salience, causing improvement to level off. Our results supported both the traditional hypothesis and all our predictions about relative salience. We conclude that there is good agreement between long-held ideas about how Batesian mimicry evolves and recent insights from learning psychology about the role of salience in mimicry evolution., (© 2018 The Author(s). Evolution © 2018 The Society for the Study of Evolution.)

Abstract: Cognition is fundamental to animals' lives and an important source of phenotypic variation. Nevertheless, research on individual variation in animal cognition is still limited. Further, although individual cognitive abilities have been suggested to be linked to personality (i.e., consistent behavioral differences among individuals), few studies have linked performance across multiple cognitive tasks to personality traits. Thus, the interplays between cognition and personality are still unclear. We therefore investigated the relationships between an important aspect of cognition, learning, and personality, by exposing young and adult red junglefowl ( Gallus gallus ) to multiple learning tasks (discriminative, reversal, and spatial learning) and personality assays (novel arena, novel object, and tonic immobility). Learning speed was not correlated across learning tasks, and learning speed in discrimination and spatial learning tasks did not co-vary with personality. However, learning speed in reversal tasks was associated with individual variation in exploration, and in an age-dependent manner. More explorative chicks learned the reversal task faster than less explorative ones, while the opposite association was found for adult females (learning speed could not be assayed in adult males). In the same reversal tasks, we also observed a sex difference in learning speed of chicks, with females learning faster than males. Our results suggest that the relationship between cognition and personality is complex, as shown by its task- and age-dependence, and encourage further investigation of the causality and dynamics of this relationship., Significance Statement: In the ancestor of today's chickens, the red junglefowl, we explored how personality and cognition relate by exposing both chicks and adults to several learning tasks and personality assays. Our birds differed in personality and learning speed, while fast learners in one task did not necessarily learn fast in another (i.e., there were no overall "smarter" birds). Exploration correlated with learning speed in the more complex task of reversal learning: faster exploring chicks, but slower exploring adult females, learned faster, compared to less explorative birds. Other aspects of cognition and personality did not correlate. Our results suggest that cognition and personality are related, and that the relationship can differ depending on task and age of the animal., Competing Interests: The authors declare that they have no conflict of interest.All applicable international, national, or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were conducted. The study was conducted according to ethical requirements in Sweden and was approved by Linköping Ethical committee (ethical permit number 122-10).

Understanding and predicting phenology has become more important with ongoing climate change and has brought about great research efforts in the recent decades. The majority of studies examining spring phenology of insects have focussed on the effects of spring temperatures alone. Here we use citizen-collected observation data to show that winter cold duration, in addition to spring temperature, can affect the spring emergence of butterflies. Using spatial mixed models, we disentangle the effects of climate variables and reveal impacts of both spring and winter conditions for five butterfly species that overwinter as pupae across the UK, with data from 1976 to 2013 and one butterfly species in Sweden, with data from 2001 to 2013. Warmer springs lead to earlier emergence in all species and milder winters lead to statistically significant delays in three of the five investigated species. We also find that the delaying effect of winter warmth has become more pronounced in the last decade, during which time winter durations have become shorter. For one of the studied species, Anthocharis cardamines (orange tip butterfly), we also make use of parameters determined from previous experiments on pupal development to model the spring phenology. Using daily temperatures in the UK and Sweden, we show that recent variation in spring temperature corresponds to 10-15 day changes in emergence time over UK and Sweden, whereas variation in winter duration corresponds to 20 days variation in the south of the UK versus only 3 days in the south of Sweden. In summary, we show that short winters delay phenology. The effect is most prominent in areas with particularly mild winters, emphasising the importance of winter for the response of ectothermic animals to climate change. With climate change, these effects may become even stronger and apply also at higher latitudes., (© 2017 The Authors. Journal of Animal Ecology © 2017 British Ecological Society.)

A fundamental question in life-history evolution is how organisms cope with fluctuating environments, including variation between stressful and benign conditions. For short-lived organisms, environments commonly vary between generations. Using a novel experimental design, we exposed wild-derived Drosophila melanogaster to three different selection regimes: one where generations alternated between starvation and benign conditions, and starvation was always preceded by early exposure to cold; another where starvation and benign conditions alternated in the same way, but cold shock sometimes preceded starvation and sometimes benign conditions; and a third where conditions were always benign. Using six replicate populations per selection regime, we found that selected flies increased their starvation resistance, most strongly for the regime where cold and starvation were reliably combined, and this occurred without decreased fecundity or extended developmental time. The selected flies became stress resistant, displayed a pronounced increase in early life food intake and resource storage. In contrast to previous experiments selecting for increased starvation resistance in D. melanogaster , we did not find increased storage of lipids as the main response, but instead that, in particular for females, storage of carbohydrates was more pronounced. We argue that faster mobilization of carbohydrates is advantageous in fluctuating environments and conclude that the phenotype that evolved in our experiment corresponds to a compromise between the requirements of stressful and benign environments.

There are many inputs during development that influence an organism's fit to current or upcoming environments. These include genetic effects, transgenerational epigenetic influences, environmental cues and developmental noise, which are rarely investigated in the same formal framework. We study an analytically tractable evolutionary model, in which cues are integrated to determine mature phenotypes in fluctuating environments. Environmental cues received during development and by the mother as an adult act as detection-based (individually observed) cues. The mother's phenotype and a quantitative genetic effect act as selection-based cues (they correlate with environmental states after selection). We specify when such cues are complementary and tend to be used together, and when using the most informative cue will predominate. Thus, we extend recent analyses of the evolutionary implications of subsets of these effects by providing a general diagnosis of the conditions under which detection and selection-based influences on development are likely to evolve and coexist., (© 2016 John Wiley & Sons Ltd/CNRS.)

There are many situations where relatives interact while at the same time there is genetic polymorphism in traits influencing survival and reproduction. Examples include cheater-cooperator polymorphism and polymorphic microbial pathogens. Environmental heterogeneity, favoring different traits in nearby habitats, with dispersal between them, is one general reason to expect polymorphism. Currently, there is no formal framework of social evolution that encompasses genetic polymorphism. We develop such a framework, thus integrating theories of social evolution into the evolutionary ecology of heterogeneous environments. We allow for adaptively maintained genetic polymorphism by applying the concept of genetic cues. We analyze a model of social evolution in a two-habitat situation with limited dispersal between habitats, in which the average relatedness at the time of helping and other benefits of helping can differ between habitats. An important result from the analysis is that alleles at a polymorphic locus play the role of genetic cues, in the sense that the presence of a cue allele contains statistical information for an organism about its current environment, including information about relatedness. We show that epistatic modifiers of the cue polymorphism can evolve to make optimal use of the information in the genetic cue, in analogy with a Bayesian decision maker. Another important result is that the genetic linkage between a cue locus and modifier loci influences the evolutionary interest of modifiers, with tighter linkage leading to greater divergence between social traits induced by different cue alleles, and this can be understood in terms of genetic conflict.

Mimicry occurs when one species gains protection from predators by resembling an unprofitable model species. The degree of mimic-model similarity is variable in nature and is closely related to the number of traits that the mimic shares with its model. Here, we experimentally test the hypothesis that the relative salience of traits, as perceived by a predator, is an important determinant of the degree of mimic-model similarity required for successful mimicry. We manipulated the relative salience of the traits of a two-trait artificial model prey, and subsequently tested the survival of mimics of the different traits. The unrewarded model prey had two colour traits, black and blue, and the rewarded prey had two combinations of green, brown and grey shades. Blue tits were used as predators. We found that the birds perceived the black and blue traits to be similarly salient in one treatment, and mimic-model similarity in both traits was then required for high mimic success. In a second treatment, the blue trait was the most salient trait, and mimic-model similarity in this trait alone achieved high success. Our results thus support the idea that similar salience of model traits can explain the occurrence of multi-trait mimicry., (© 2015 The Author(s).)

The effect of spring temperature on spring phenology is well understood in a wide range of taxa. However, studies on how winter conditions may affect spring phenology are underrepresented. Previous work on Anthocharis cardamines (orange tip butterfly) has shown population-specific reaction norms of spring development in relation to spring temperature and a speeding up of post-winter development with longer winter durations. In this experiment, we examined the effects of a greater and ecologically relevant range of winter durations on post-winter pupal development of A. cardamines of two populations from the United Kingdom and two from Sweden. By analyzing pupal weight loss and metabolic rate, we were able to separate the overall post-winter pupal development into diapause duration and post-diapause development. We found differences in the duration of cold needed to break diapause among populations, with the southern UK population requiring a shorter duration than the other populations. We also found that the overall post-winter pupal development time, following removal from winter cold, was negatively related to cold duration, through a combined effect of cold duration on diapause duration and on post-diapause development time. Longer cold durations also lead to higher population synchrony in hatching. For current winter durations in the field, the A. cardamines population of southern UK could have a reduced development rate and lower synchrony in emergence because of short winters. With future climate change, this might become an issue also for other populations. Differences in winter conditions in the field among these four populations are large enough to have driven local adaptation of characteristics controlling spring phenology in response to winter duration. The observed phenology of these populations depends on a combination of winter and spring temperatures; thus, both must be taken into account for accurate predictions of phenology.

The development of multicellular organisms involves a delicate interplay between genetic and environmental influences. It is often useful to think of developmental systems as integrating available sources of information about current conditions to produce organisms. Genes and inherited physiology provide cues, as does the state of the environment during development. The integration systems themselves are under genetic control and subject to Darwinian selection, so we expect them to evolve to produce organisms that fit well with current ecological (including social) conditions. We argue for the scientific value of this explicitly informational perspective by providing detailed examples of how it can elucidate taxonomically diverse phenomena. We also present a general framework for linking genetic and phenotypic variation from an informational perspective. This application of Darwinian logic at the organismal level can elucidate genetic influences on phenotypic variation in novel and counterintuitive ways., (Copyright © 2015 Elsevier Ltd. All rights reserved.)

An organism's phenotype can be influenced by maternal cues and directly perceived environmental cues, as well as by its genotype at polymorphic loci, which can be interpreted as a genetic cue. In fluctuating environments, natural selection favors organisms that efficiently integrate different sources of information about the likely success of phenotypic alternatives. In such situations, it can be beneficial to pass on maternal cues that offspring can respond to. A maternal cue could be based on environmental cues directly perceived by the mother but also partly on cues that were passed on by the grandmother. We have used a mathematical model to investigate how the passing of maternal cues and the integration of different sources of information evolve in response to qualitatively different kinds of temporal and spatial environmental fluctuations. The model shows that the passing of maternal cues and the transgenerational integration of sources of information readily evolve. Factors such as the degree of temporal autocorrelation, the predictive accuracy of different environmental cues, and the level of gene flow strongly influence the expression of adaptive maternal cues and the relative weights given to different sources of information. We outline the main features of the relation between the characteristics of environmental fluctuations and the adaptive systems of phenotype determination and compare these predictions with empirical studies on cue integration.

The temporal aspects of life cycle characteristics, such as diapause development, are under strong selection in seasonal environments. Fine-tuning of the life cycle may be particularly important to match the phenology of potential mates and resources as well as for optimizing abiotic conditions at eclosion. Here, we experimentally study the spring phenology of the orange tip butterfly, Anthocharis cardamines, by analysing post-winter pupal development in three populations along a latitudinal cline in each of Sweden and the United Kingdom. These countries differ substantially in their seasonal temperature profile. By repeatedly recording pupal weights, we established that post-winter development has two separate phases, with a more rapid weight loss in the second phase than in the first, likely corresponding to a ramping up of the rate of development. Variation in the duration of the first phase contributed more strongly than the second phase to the differences in phenology between the localities and sexes. We found that insects from Sweden had a faster overall rate of development than those from the United Kingdom, which is consistent with countergradient variation, as Sweden is colder during the spring than the United Kingdom. Similar trends were not observed at the within-country scale, however. A cogradient pattern was found within Sweden, with populations from the north developing more slowly, and there was no clear latitudinal trend within the United Kingdom. In all localities, males developed faster than females. Our results point to the importance of variation in the progression of post-winter development for spring phenology., (© 2014 European Society For Evolutionary Biology. Journal of Evolutionary Biology © 2014 European Society For Evolutionary Biology.)

Individuals in social species commonly form dominance relationships, where dominant individuals enjoy greater access to resources compared to subordinates. A range of factors such as sex, age, body size and prior experiences has to varying degrees been observed to affect the social status an individual obtains. Recent work on animal personality (i.e. consistent variation in behavioural responses of individuals) demonstrates that personality can co-vary with social status, suggesting that also behavioural variation can play an important role in establishment of status. We investigated whether personality could predict the outcome of duels between pairs of morphologically matched male domestic fowl (Gallus gallus domesticus), a species where individuals readily form social hierarchies. We found that males that more quickly explored a novel arena, or remained vigilant for a longer period following the playback of a warning call were more likely to obtain a dominant position. These traits were uncorrelated to each other and were also uncorrelated to aggression during the initial part of the dominance-determining duel. Our results indicate that several behavioural traits independently play a role in the establishment of social status, which in turn can have implications for the reproductive success of different personality types.

The theory of mimicry explains how a mimic species gains advantage by resembling a model species [1-3]. Selection for increased mimic-model similarity should then result in accurate mimicry, yet there are many surprising examples of poor mimicry in the natural world [4-8]. The existence of imperfect mimics remains a major unsolved conundrum. We propose and experimentally test a novel explanation of the phenomenon. We argue that predators perceive prey as having several traits, but that the traits differ in their importance for learning. When predators learn to discriminate prey, high-salience traits overshadow other traits, leaving them under little or no selection for similarity, and allow imperfect mimicry to succeed. We tested this idea experimentally, using blue tits as predators and artificial prey with three prominent traits: color, pattern, and shape. We found that otherwise imperfect color mimics were avoided about as much as perfect mimics, whereas pattern and shape mimics did not gain from their similarity to the model. All traits could separately be perceived and learned by the predators, but the color trait was learned at a higher rate, implying that it had higher salience. We conclude that difference in salience between components of prey appearance is of major importance in explaining imperfect mimicry., (Copyright © 2014 Elsevier Ltd. All rights reserved.)

A traditional question in community ecology is whether species' traits are distributed as more-or-less regularly spaced clusters. Interspecific competition has been suggested to play a role in such structuring of communities. The seminal theoretical work on limiting similarity and species packing, presented four decades ago by Robert MacArthur, Richard Levins and Robert May, has recently been extended. There is now a deeper understanding of how competitive interactions influence community structure, for instance, how the shape of competition kernels can determine the clustering of species' traits. Competition is typically weaker for greater phenotypic difference, and the shape of the dependence defines a competition kernel. The clustering tendencies of kernels interact with other effects, such as variation in resource availability along a niche axis, but the kernel shape can have a decisive influence on community structure. Here we review and further extend the recent developments and evaluate their importance., (© 2013 Published by Elsevier Ltd. All rights reserved.)

Stability of 'state' has been suggested as an underlying factor explaining behavioural stability and animal personality (i.e. variation among, and consistency within individuals in behavioural responses), but the possibility that stable social relationships represent such states remains unexplored. Here, we investigated the influence of social status on the expression and consistency of behaviours by experimentally changing social status between repeated personality assays. We used male domestic fowl (Gallus gallus domesticus), a social species that forms relatively stable dominance hierarchies, and showed that behavioural responses were strongly affected by social status, but also by individual characteristics. The level of vigilance, activity and exploration changed with social status, whereas boldness appeared as a stable individual property, independent of status. Furthermore, variation in vocalization predicted future social status, indicating that individual behaviours can both be a predictor and a consequence of social status, depending on the aspect in focus. Our results illustrate that social states contribute to both variation and stability in behavioural responses, and should therefore be taken into account when investigating and interpreting variation in personality.

In mutualistic interactions, partners obtain a net benefit, but there may also be costs associated with the provision of benefits for a partner. The question of whether aphids suffer such costs when attended by ants has been raised in previous work. Transgenerational effects, where offspring phenotypes are adjusted based on maternal influences, could be important in the mutualistic interaction between aphids and ants, in particular because aphids have telescoping generations where two offspring generations can be present in a mature aphid. We investigated the immediate and transgenerational influence of ant tending on aphid life history and reproduction by observing the interaction between the facultative myrmecophile Aphis fabae and the ant Lasius niger over 13 aphid generations in the laboratory. We found that the effect of ant tending changes dynamically over successive aphid generations after the start of tending. Initially, total aphid colony weight, aphid adult weight and aphid embryo size decreased compared with untended aphids, consistent with a cost of ant association, but these differences disappeared within four generations of interaction. We conclude that transgenerational effects are important in the aphid-ant interactions and that the costs for aphids of being tended by ants can vary over generations.

Species interacting in varied ecological conditions often evolve in different directions in different local populations. The butterflies of the cryptic Leptidea complex are sympatrically distributed in different combinations across their Eurasian range. Interestingly, the same species is a habitat generalist in some regions and a habitat specialist in others, where a sibling species has the habitat generalist role. Previous studies suggest that this geographically variable niche divergence is generated by local processes in different contact zones. By varying the absolute and relative densities of Leptidea sinapis and Leptidea juvernica in large outdoor cages, we show that female mating success is unaffected by conspecific density, but strongly negatively affected by the density of the other species. Whereas 80% of the females mated when a conspecific couple was alone in a cage, less than 10% mated when the single couple shared the cage with five pairs of the other species. The heterospecific courtships can thus affect the population fitness, and for the species in the local minority, the suitability of a habitat is likely to depend on the presence or absence of the locally interacting species. If the local relative abundance of the different species depends on the colonization order, priority effects might determine the ecological roles of interacting species in this system., (© 2013 The Authors. Journal of Evolutionary Biology © 2013 European Society For Evolutionary Biology.)

The difference in phenotypes of queens and workers is a hallmark of the highly eusocial insects. The caste dimorphism is often described as a switch-controlled polyphenism, in which environmental conditions decide an individual's caste. Using theoretical modeling and empirical data from honeybees, we show that there is no discrete larval developmental switch. Instead, a combination of larval developmental plasticity and nurse worker feeding behavior make up a colony-level social and physiological system that regulates development and produces the caste dimorphism. Discrete queen and worker phenotypes are the result of discrete feeding regimes imposed by nurses, whereas a range of experimental feeding regimes produces a continuous range of phenotypes. Worker ovariole numbers are reduced through feeding-regime-mediated reduction in juvenile hormone titers, involving reduced sugar in the larval food. Based on the mechanisms identified in our analysis, we propose a scenario of the evolutionary history of honeybee development and feeding regimes.

In Batesian mimicry, a harmless prey species imitates the warning coloration of an unpalatable model species. A traditional suggestion is that mimicry evolves in a two-step process, in which a large mutation first achieves approximate similarity to the model, after which smaller changes improve the likeness. However, it is not known which aspects of predator psychology cause the initial mutant to be perceived by predators as being similar to the model, leaving open the question of how the crucial first step of mimicry evolution occurs. Using theoretical evolutionary simulations and reconstruction of examples of mimicry evolution, we show that the evolution of Batesian mimicry can be initiated by a mutation that causes prey to acquire a trait that is used by predators as a feature to categorize potential prey as unsuitable. The theory that species gain entry to mimicry through feature saltation allows us to formulate scenarios of the sequence of events during mimicry evolution and to reconstruct an initial mimetic appearance for important examples of Batesian mimicry. Because feature-based categorization by predators entails a qualitative distinction between nonmimics and passable mimics, the theory can explain the occurrence of imperfect mimicry., (© 2011 The Author(s). Evolution© 2011 The Society for the Study of Evolution.)

Phenotypically plastic traits can be expressed as continuous reaction norms or as threshold traits, but little is known about the selective conditions that favor one over the other. We study this question using a model of prey defenses in which prey can induce any level of defense conditional on cues that are informative of local predator density. The model incorporates a trade-off between defense expression and fecundity and feedback between the defense level of prey and predator attack rates. Both continuous reaction norms and threshold traits can emerge as evolutionarily stable solutions of defense induction, and we show that the shape of the trade-off curve plays a key role in determining the outcome. Threshold traits are favored when selection is disruptive. Ecological conditions that favor defense dimorphisms in the absence of cues will favor threshold traits in the presence of slightly informative cues. We caution that continuous reaction norms and threshold traits may result in similar patterns of defense expression at the population level, and we discuss potential pitfalls of inferring reaction norm type from observational data.

For structured populations in equilibrium with everybody born equal, ln(R (0)) is a useful fitness proxy for evolutionarily steady strategy (ESS) and most adaptive dynamics calculations, with R (0) the average lifetime number of offspring in the clonal and haploid cases, and half the average lifetime number of offspring fathered or mothered for Mendelian diploids. When individuals have variable birth states, as is, for example, the case in spatial models, R (0) is itself an eigenvalue, which usually cannot be expressed explicitly in the trait vectors under consideration. In that case, Q(Y| X):=-det (I-L(Y| X)) can often be used as fitness proxy, with L the next-generation matrix for a potential mutant characterized by the trait vector Y in the (constant) environment engendered by a resident characterized by X. If the trait space is connected, global uninvadability can be determined from it. Moreover, it can be used in all the usual local calculations like the determination of evolutionarily singular trait vectors and their local invadability and attractivity. We conclude with three extended case studies demonstrating the usefulness of Q: the calculation of ESSs under haplo-diploid genetics (I), of evolutionarily steady genetic dimorphisms (ESDs) with a priori proportionality of macro- and micro-gametic outputs (an assumption that is generally made but the fulfilment of which is a priori highly exceptional) (II), and of ESDs without such proportionality (III). These case studies should also have some interest in their own right for the spelled out calculation recipes and their underlying modelling methodology.

A major question for the study of phenotypic evolution is whether intra- and interspecific diversity originates directly from genetic variation, or instead, as plastic responses to environmental influences initially, followed later by genetic change. In species with discrete alternative phenotypes, evolutionary sequences can be inferred from transitions between environmental and genetic phenotype control, and from losses of phenotypic alternatives. From the available evidence, sequences appear equally probable to start with genetic polymorphism as with polyphenism, with a possible dominance of one or the other for specific trait types. We argue in this review that to evaluate the prevalence of each route, an investigation of both genetic and environmental cues for phenotype determination in several related rather than in isolated species is required., (Copyright © 2011 Elsevier Ltd. All rights reserved.)

In applying game theory to problems in biology, differences between individuals are often ignored. In particular, when analysing the evolution of cooperation it is often implicitly assumed that ignoring variation will produce predictions that approximate the solution when differences are included. This need not be true. As we demonstrate, differences are not innocuous noise, but can fundamentally change the nature of a game. Even small amounts of variability can stabilize cooperation by, for example, maintaining the need to deal with cheaters. Differences promote the need to learn about others in an interaction, leading to contingent behaviour that can reduce conflict, and to negotiated outcomes that may or may not be more cooperative than unconditional actions. Once there are mechanisms such as mutation and environmental influences that maintain variation within populations, whether cooperation evolves may depend on the variation in the cooperativeness trait. Variation means that it may be worth taking a chance that a partner is cooperative by being cooperative. When there are markets, so that individuals can break off interactions to seek a better partner, variation promotes choosiness and hence penalizes those uncooperative individuals, who are rejected. Variation promotes the need to monitor the previous behaviour of others, and once this social sensitivity exists, the need to maintain a good reputation can promote cooperation.

Studies of the evolution of helping have traditionally used the explanatory frameworks of reciprocity and altruism towards relatives, but recently there has been an increasing interest in other kinds of explanations. We review the success or otherwise of work investigating alternative processes and mechanisms, most of which fall under the heading of cooperation for direct benefits. We evaluate to what extent concepts such as by-product benefits, pseudo-reciprocity, sanctions and partner choice, markets and the build-up of cross-species spatial trait correlations have contributed to the study of the evolution of cooperation. We conclude that these alternative ideas are successful and show potential to further increase our understanding of cooperation. We also bring up the origin and role of common interest in the evolution of cooperation, including the appearance of organisms. We note that there are still unresolved questions about the main processes contributing to the evolution of common interest. Commenting on the broader significance of the recent developments, we argue that they represent a justified balancing of the importance given to different major hypotheses for the evolution of cooperation. This balancing is beneficial because it widens considerably the range of phenomena addressed and, crucially, encourages empirical testing of important theoretical alternatives.

The two-step hypothesis of Müllerian mimicry evolution states that mimicry starts with a major mutational leap between adaptive peaks, followed by gradual fine-tuning. The hypothesis was suggested to solve the problem of apostatic selection producing a valley between adaptive peaks, and appears reasonable for a one-dimensional phenotype. Extending the hypothesis to the realistic scenario of multidimensional phenotypes controlled by multiple genetic loci can be problematic, because it is unlikely that major mutational leaps occur simultaneously in several traits. Here we consider the implications of predator psychology on the evolutionary process. According to feature theory, single prey traits may be used by predators as features to classify prey into discrete categories. A mutational leap in such a trait could initiate mimicry evolution. We conducted individual-based evolutionary simulations in which virtual predators both categorize prey according to features and generalize over total appearances. We found that an initial mutational leap toward feature similarity in one dimension facilitates mimicry evolution of multidimensional traits. We suggest that feature-based predator categorization together with predator generalization over total appearances solves the problem of applying the two-step hypothesis to complex phenotypes, and provides a basis for a theory of the evolution of mimicry rings.

Generalization is at the heart of many aspects of behavioral ecology; for foragers it can be seen as an essential feature of learning about potential prey, because natural populations of prey are unlikely to be perfectly homogenous. Aposematic signals are considered to aid predators in learning to avoid a class of defended prey. Predators do this by generalizing between the appearance of prey they have previously sampled and the appearance of prey they subsequently encounter. Mimicry arises when such generalization occurs between individuals of different species. Our aim here is to explore whether the specific shape of the generalization curve can be expected to be important for theoretical predictions relating to the evolution of aposematism and mimicry. We do this by a reanalysis and development of the models provided in two recent papers. We argue that the shape of the generalization curve, in combination with the nature of genetic and phenotypic variation in prey traits, can have evolutionary significance under certain delineated circumstances. We also demonstrate that the process of gradual evolution of Müllerian mimicry proposed by Fisher is particularly efficient in populations with a rich supply of standing genetic variation in mimetic traits.

Service providers may vary service quality depending on whether they work alone or provide the service simultaneously with a partner. The latter case resembles a prisoner's dilemma, in which one provider may try to reap the benefits of the interaction without providing the service. Here we present a game-theory model based on the marginal value theorem, which predicts that as long as the client determines the duration, and the providers cooperate towards mutual gain, service quality will increase in the pair situation. This prediction is consistent with field observations and with an experiment on cleaning mutualism, in which stable male-female pairs of the cleaner wrasse Labroides dimidiatus repeatedly inspect client fish jointly. Cleaners cooperate by eating ectoparasites off clients but actually prefer to cheat and eat client mucus. Because clients often leave in response to such cheating, the benefits of cheating can be gained by only one cleaner during a pair inspection. In both data sets, the increased service quality during pair inspection was mainly due to the smaller females behaving significantly more cooperatively than their larger male partners. In contrast, during solitary inspections, cleaning behaviour was very similar between the sexes. Our study highlights the importance of incorporating interactions between service providers to make more quantitative predictions about cooperation between species.

The usefulness of GST and similar measures of genetic differentiation has been questioned repeatedly because of their dependence on the amount of heterozygosity within populations, creating problems when comparing degrees of divergence at loci with different mutation rates. Although the effect of mutation on GST is expected to be small in the early phases of divergence, it is unclear for how long after separation from a common ancestral population that GST is largely unaffected by mutation and by the resulting effect on heterozygosity. We address this question through analysis of the recursion equations for gene identity under the infinite allele model of mutation, and derive conditions describing when the effect of mutation on GST can be ignored under mutation-migration-drift equilibrium conditions and during the preceding transition phase. An important result is that during the transition phase GST is not only affected by mutation, but also by the heterozygosity in the base population from which the subpopulations diverged. The effect of mutation on GST is significant from the very start of the divergence process when initial heterozygosity is low, whereas GST is only weakly affected by mutation in the early phases of differentiation when initial heterozygosity is high. Thus, differentiation following a severe bottleneck is strongly dependent on mutation. The standardized measure of differentiation, G'ST, suggested by Hedrick (2005), may be helpful when comparing amounts of divergence at loci with different mutation rates under steady-state conditions, provided that migration is very low. In many other situations the use of G'ST might be misleading, however, and its application should be exercised with caution.

We have analyzed the evolution of a quantitative trait in populations that are spatially extended along an environmental gradient, with gene flow between nearby locations. In the absence of competition, there is stabilizing selection toward a locally best-adapted trait that changes gradually along the gradient. According to traditional ideas, gradual spatial variation in environmental conditions is expected to lead to gradual variation in the evolved trait. A contrasting possibility is that the trait distribution instead breaks up into discrete clusters. Doebeli and Dieckmann (2003) argued that competition acting locally in trait space and geographical space can promote such clustering. We have investigated this possibility using deterministic population dynamics for asexual populations, analyzing our model numerically and through an analytical approximation. We examined how the evolution of clusters is affected by the shape of competition kernels, by the presence of Allee effects, and by the strength of gene flow along the gradient. For certain parameter ranges clustering was a robust outcome, and for other ranges there was no clustering. Our analysis shows that the shape of competition kernels is important for clustering: the sign structure of the Fourier transform of a competition kernel determines whether the kernel promotes clustering. Also, we found that Allee effects promote clustering, whereas gene flow can have a counteracting influence. In line with earlier findings, we could demonstrate that phenotypic clustering was favored by gradients of intermediate slope.

In recent years evidence has been accumulating that personalities are not only found in humans but also in a wide range of other animal species. Individuals differ consistently in their behavioural tendencies and the behaviour in one context is correlated with the behaviour in multiple other contexts. From an adaptive perspective, the evolution of animal personalities is still a mystery, because a more flexible structure of behaviour should provide a selective advantage. Accordingly, many researchers view personalities as resulting from constraints imposed by the architecture of behaviour (but see ref. 12). In contrast, we show here that animal personalities can be given an adaptive explanation. Our argument is based on the insight that the trade-off between current and future reproduction often results in polymorphic populations in which some individuals put more emphasis on future fitness returns than others. Life-history theory predicts that such differences in fitness expectations should result in systematic differences in risk-taking behaviour. Individuals with high future expectations (who have much to lose) should be more risk-averse than individuals with low expectations. This applies to all kinds of risky situations, so individuals should consistently differ in their behaviour. By means of an evolutionary model we demonstrate that this basic principle results in the evolution of animal personalities. It simultaneously explains the coexistence of behavioural types, the consistency of behaviour through time and the structure of behavioural correlations across contexts. Moreover, it explains the common finding that explorative behaviour and risk-related traits like boldness and aggressiveness are common characteristics of animal personalities.