Skip to content

Research Papers On Ethological Behaviour Of Animals


“Animals should be housed with a goal of maximizing species-specific behaviors and minimizing stress-induced behaviors” ( NRC 1996 , p 22)--a laudable goal, but can it be achieved? The answer is perhaps, but doing so will necessitate addressing some difficult questions. How do we maximize behaviors in an environment that is so different from the one in which the animal evolved its species-typical behaviors? Should the animal be allowed to perform all of its species-typical behaviors or only certain ones? If the latter, how do we choose which ones? How can we recognize and minimize stress-induced behaviors?

Although people have long been fascinated by the behavior of animals, the formal discipline of animal behavior--ethology--is actually relatively new, dating to the work of Konrad Lorenz in Austria in the 1930s. Application of ethological principles and methods to the study of animal welfare is an even newer endeavor, of course, and one that has generated a great deal of stimulating discussion and controversy during its short history. In this paper, I provide an overview of the development of behavioral approaches to the study of animal welfare. I then discuss some reasons that behaviors are important to animals and describe how an understanding of behavior can be useful when designing housing environments for laboratory animals.

Behavioral Approaches to the Study of Welfare


The report on intensive farming practices authored by the Brambell committee (1965 ) was probably the first published document to emphasize the importance of behavior in assessing animal welfare. This committee was established by the British government after the public outcry following publication of Ruth Harrison's expos? of what she referred to as “factory farming” methods in Animal Machines (1964). After hearing testimony and reviewing farming practices in Europe, the members of the committee wrote ( Brambell 1965 , p 10):

The scientific evidence bearing on the sensations and sufferings of animals is derived from anatomy and physiology on the one hand and from ethology, the science of human behavior, on the other … we have been impressed by the evidence to be derived from the study of the behavior of the animal. We consider that this is a field of scientific research in relation to animal husbandry which has not attracted the attention which it deserves and that opportunities should be sought to encourage its development.

They further concluded that animals had behavioral needs that could not be satisfied in barren, restrictive environments, and that not providing for those needs was likely to cause suffering, ideas that have proven to be very influential in shaping ethological research on animal welfare.

In an appendix to the Brambell report, the eminent zoologist W. H. Thorpe argued that welfare is promoted when animals are able to perform the activities that most closely resemble the behavioral repertoire of their free-ranging con-specifics. However, this idea soon fell into disfavor among many scientists studying animal welfare ( Dawkins 1980 ). Fraser (1989 ), for example, discussed the limitations of using the normal behavioral repertoire as the baseline for ensuring welfare with reference to 3 normal behaviors in swine: the distress calls given by piglets when they are separated from their mothers; nest building by sows before parturition; and wallowing, which is a thermoregulatory behavior shown only under hot conditions. Fraser pointed out that a pig's natural behavioral repertoire consists of things that the pig really does not want to do (such as give distress calls), wants to do (such as build a nest), and wants to do but only when conditions require it (such as wallow).

The 3 behaviors described above obviously have different implications for welfare. Providing a pregnant sow kept in a temperature-controlled environment with a wallow would do little to improve her welfare, whereas giving her nest-building material might improve her welfare a great deal. Placing piglets in a situation in which they give distress calls, however, would actually reduce their welfare. Thus, to assess the welfare significance of particular behaviors, it is important to have an understanding of what causes the behavior to occur in the first place.

Motivation and Welfare

Many factors can motivate the performance of behaviors, and a variety of models have been proposed in an attempt to explain how motivational systems work ( Jensen and Toates 1993 ; Toates 1986 ). One approach to studying motivation has been the attempt to determine the relative importance of internal and external factors in causing particular behaviors (for example, Hughes 1988 ). Some behaviors are classified as being motivated primarily by factors external to the animal, exemplified by thermoregulatory behaviors like wallowing in pigs and antipredator behaviors in prey species. Other behaviors, like food searching when hungry, appear to be largely internally motivated. Yet other behaviors are elicited by complex interplay between both internal and external factors. Mating behavior, for example, is motivated by hormonal state, which may in turn depend on seasonal and other environmental factors as well as the accessibility and readiness of appropriate sexual partners. Cues from one partner to the other during courtship can further influence the hormonal states and behavior of the courting pair.

The study of motivation has been very important in applied ethology because of its link to the Brambell committee's ideas about behavioral needs. Behavioral needs are generally conceptualized as those behaviors that the animal must perform regardless of environmental circumstances, that is, primarily internally motivated behaviors that may occur even in the absence of appropriate external stimulation, although sometimes in an aberrant form. Hens kept in cages without litter material, for example, will still perform dustbathing behaviors, although the movements are somewhat abnormal and the dustbathing episode is short compared with a normal dustbathing episode ( Vestergaard 1980 ). Behaviors of this type are called vacuum activities (because they occur “in a vacuum,” so to speak). Another form these behaviors can take is to become stereotyped. Many oral stereotypies, for instance, have been shown to be associated with the lack of opportunity to perform particular components of feeding behavior, including foraging ( Bayne and others 1991 ; Redbo and Norblad 1997 ; Rushen and others 1993 ).

A thorny question remains, however: Is the performance of an aberrant behavior, or not meeting a behavioral need, linked to “suffering” (hereafter referred to as distress) as the Brambell committee claimed? Duncan (1978a ) argued that independent verification was necessary to demonstrate that unusual or inappropriate behaviors indicated reduced welfare. One common approach to providing such verification has been to measure physiological (including immunological) parameters associated with stress and then to attempt to correlate those parameters with abnormal behaviors or responses to behavioral restriction. However, physiological measures also have many limitations, and it has generally proven difficult to interpret their significance as indicators of welfare ( Dawkins 1980 ; Mason and Mendl 1993 ; Rushen and de Passill? 1992 ). The answer to the question about suffering is therefore still elusive, as can be demonstrated by considering research on abnormal behaviors.

Abnormal Behaviors

In the laboratory setting, abnormal behaviors are often used as the benchmark of poor housing conditions and the need for environmental enrichment. Behaviors that cause injury to either the initiator (self-mutilation) or the recipient (cannibalism) clearly have a negative effect on the welfare of the individual sustaining the injury. However, no consistent relationship has been demonstrated between noninjurious abnormal behaviors, particularly stereotypies, and other measures of reduced welfare ( Lawrence and Rushen 1993 ; Mason 1991 ; Mench and Mason 1997 ).

Stereotypies are sometimes, but not always, linked to physiological changes indicative of stress; and in fact, the performance of stereotypies sometimes appears to be rewarding or to reduce stress. Although abnormal behaviors often appear to arise from frustrated motivation, they may also be caused by factors that are not directly related to a poor environment, such as pathology or neurological predisposition. Most puzzling is that established stereotypies may also persist even when animals are placed in enriched environments. Has the enriched environment failed to improve the animal's welfare? Mason (1991 ) suggests instead that stereotypies may be as much “scars” of past experience as they are indicators of current frustration or environmental inadequacy. Since many stereotypies can be prevented from forming by altering the animal's environmental conditions, there is widespread agreement that appropriate steps should be taken to accomplish this ( Duncan and others 1993 ).

Preference Testing and Demand Curves

Another way in which behavior has been used to provide information about welfare is in studies giving animals choices and opportunities to express preferences ( Dawkins 1980 ). In a pioneering study of animal preferences, Hughes and Black (1973 ) decided to test a recommendation made by the Brambell committee--that laying hens should be kept in cages with floors made of rectangular metal mesh, rather than fine-gauge “chicken wire,” because the latter sags and the hen's foot is not well adapted to grip it comfortably. Hughes and Black gave hens a choice of 4 different types of flooring including chicken wire, 2 types of rectangular mesh, and perforated metal. The hens obviously viewed the situation somewhat differently than did the Brambell Committee; they chose the chicken wire in preference to the other floors, and subsequent visual inspection showed that the chicken wire in fact provided the best support for their feet. Preference testing of this type has now been used widely in studies of animal welfare, including assessing preferences shown for cage height, light intensity, flooring, bedding material, and enrichment devices by rats ( Blom and others 1995 , 1996 ; Chmiel and Noonan 1996 ; Manser and others 1995 ; van de Weerd and others 1996 ), sleeping locations by mice ( Sherwin 1996 ), flooring and social companions by hamsters ( Arnold and Estep 1990 , 1994 ), and social companions by rabbits ( Held and others 1995 ).

Cautions have been raised about the use and interpretation of preference testing ( Dawkins 1980 ; Duncan 1978b ). Preferences can be shaped by many factors, including genetics, previous experience, and the choices offered and testing methods used. Short-term choices may not correspond well to long-term welfare, since animals (like humans) may choose things that are immediately rewarding but are not necessarily best for them in the long term. An awareness of these problems has led to a rapid evolution of preference testing toward asking more precise questions and taking more comprehensive measures ( Fraser and others 1993 ); and preference testing is now widely considered to be a useful tool, particularly for evaluating specific aspects of the environment like flooring, temperature, and lighting.

One enduring criticism leveled against preference testing, however, is that it fails to distinguish between important choices and not-so-important ones--between so-called “luxuries” and “necessities.” Dawkins (1990 ) developed a model that she suggested could address this concern by allowing the strength of the motivation underlying particular choices to be determined. Her “consumer demand” model is derived from economic theory and involves requiring animals to pay some kind of cost to acquire commodities. For instance, a hen might have to peck a key or push through a barrier to gain access to a nestbox or dustbath. The amount of key pecking or barrier pushing she is willing to do can be compared with the amount she is willing to do to obtain a commodity known to be important to her, such as food. Priority can then be given to providing opportunities for animals to perform those behaviors for which they show a strong demand (that is, needs).

Dawkins argues that an animal will work the most for things that ensure its fitness or are perceived to be important in ensuring its fitness. This is an important concept because it emphasizes that domesticated and purpose-bred animals may continue to behave in ways shaped by their evolutionary history, even if those behaviors are not necessarily important for (or are sometimes even detrimental to) their survival or fitness in captivity. In fact, most behavioral changes accompanying domestication are qualitative rather than quantitative, that is, they result in changes in the frequency, rather than the elimination, of behaviors from the species-typical repertoire ( Price 1984 ).

A number of problems have been reported (including by Dawkins herself) concerning application of the consumer demand model (see Dawkins 1990 and commentaries; Houston 1997 ), although these problems are largely methodological in nature. A broader issue, however, relates to the underlying assumptions of the model and, indeed, of the whole concept of behavioral needs. Jensen and Toates (1993 ) conclude that it is almost impossible to separate behaviors into needs and nonneeds. Using a series of examples, they demonstrate that both internal and external stimuli contribute to the motivation of even seemingly simple behaviors and that different circumstances can cause a particular type of stimulus to assume more or less importance in causing the behavior. The powerful effects that an external stimulus can have, even on a behavior that we think of as largely internally motivated, is unfortunately well-known to most of us. For example, consider how we respond to a particularly tempting dessert even after a very filling meal! Accordingly, Jensen and Toates (1993 ) argue that behavioral needs will be situation-specific rather than generally applicable and that trying to devise a catalog of a particular species' behavioral needs is futile. Thus we come full circle: If we are not able to identify behavioral needs, then what is the relationship between species-typical behaviors and welfare?

Why is Behavior Important?

It is worth remembering that behavior is what animals do to interact with, respond to, and control their environment. Behavior is generally the animal's “first line of defense” in response to environmental change. As such, careful observations of behavior can provide us with a great deal of information about animals' requirements, preferences and dislikes, and internal states ( Mench and Mason 1997 ), provided that our interpretation of those observations is firmly grounded in a knowledge of species-typical behavior patterns.

An approach to management or housing design that focuses primarily on behavioral needs is too narrow ( Mench 1998b ) and does not adequately consider the beneficial effects a behavior can have on welfare even when that behavior might not be defined typically as a need. I suggest that we reconsider simple behavioral preferences, and indeed species-specific behaviors in general, and identify the consequences for an animal performing particular behaviors. The association between potential welfare benefits and an animal's performance of certain behaviors is further discussed below. This, however, is an area in which much additional research is needed, particularly for laboratory animals.

Maintaining Physical Health or Physiological Normality

The performance of certain behaviors can lead to improvements in physical health. The beneficial effects of exercise on humans are well known, and similar effects can be observed in animals given the opportunity to engage in species-typical patterns of locomotion. For example, dairy cows walked daily have fewer leg problems, including noninfectious leg and hoof disorders, as well as lower incidence of mastitis, bloat, and calving-related disorders, than cows kept in tie stalls ( Gustafson 1993 ). Captive birds whose movement is restricted tend to develop osteoporosis and osteoarthritis ( Fedducia 1991 ; Knowles and Broom 1990 ), and 5% of caged laying hens are found to have old, healed bone breaks at necropsy ( Gregory and Wilkins 1989 ). Providing the opportunity for perching behavior, however, can help to reduce bone breakage. Caged hens given raised perches have greater leg-bone and wing-bone strength ( Hughes and Appleby 1989 ), as well as better foot health, than hens from cages without perches.

Less obvious, perhaps, are the effects that performing behaviors can have on normal physiological functioning. In studying the welfare of early-weaned calves, de Passillé and others (1993 ) found that the suckling reflex was stimulated when the calves were fed even small quantities of milk and that the reflex persisted for 10 min after intake. If the calves were allowed to suck a nonnutritive (“dry”) teat after their milk meal, production of digestive hormones, including CCK and insulin, was increased. Suckling behavior itself, then, has beneficial effects on digestive physiology in calves.

Studies like these have important implications, not only for animal welfare but also for the outcome and interpretation of biomedical research projects in which animals are used as models. As Valzelli (1973 ) demonstrated, “a mouse is not a mouse is not a mouse.” Socially isolated mice differ from group-housed mice not only behaviorally, but also in their immunological responses, hormone levels, brain neurochemistry, learning ability, pain thresholds, and sensitivity to drugs.

Behavior can thus influence research findings, sometimes in unforeseen ways. Capitanio and Lerche (forthcoming ) carried out a retrospective analysis of the effects of different variables, including medical history, housing, demographics, and contents of the inoculum, on the survival of rhesus monkeys experimentally infected with simian immunodeficiency virus. After analyzing colony data from 4 regional primate research centers, they found that housing had a significant effect on the monkeys' ability to survive. Housing relocations and social separations both during the 30-day period after infection and the 90-day period before it were associated with decreased survival, as was grouping during the period after infection. Some of the differences in survivability were dramatic--for example, all of the monkeys who were relocated several times shortly before or after inoculation died within 600 days after infection, even though 40% of the monkeys that had been maintained in stable social groups were still alive.

Preventing or Reducing Illness, Fear, Stress, Pain, or Tension

Behaviors can also be important in reducing illness, pain, fear, stress, or tension. If possible, animals will remove themselves from a fear-producing stimulus by fleeing or seeking cover (or will sometimes attempt to escape detection or injury by becoming immobile). Sick animals show a number of behavioral changes, including anorexia, sleepiness, depression, and a reduction in grooming activity, which help to conserve energy and thus facilitate healing ( Hart 1988 ).

Similarly, animals use behavioral mechanisms to deal with short-term stressors like social interactions. Macaques from a stable social group subjected to a brief period of crowding, for example, avoid conflict by reducing their level of activity ( Aureli, and others 1995 ). Mutual grooming activity similarly reduces stress and tension among socially interacting primates ( Boccia and others 1989 ; Schino and others 1988 ) and is associated with the release of endogenous opioids ( Keveme and others 1989 ). Social companionship and the ability to engage in social behavior, even aggressive behavior, can also have stress-buffering effects. Rhesus monkeys exposed to novelty show attenuated cortisol responses and less fearful behavior if a familiar social companion is present ( Hennessy 1984 ), and rats that can interact aggressively with another rat during exposure to shock have lower stress levels ( Conner and others 1971 ).

All of the foregoing are examples of behaviors or behavioral changes that are of relatively short-term benefit. However, the performance of behaviors can also have longer-term consequences for the animal. Many studies have shown that providing an enriched or complex environment for the animal, particularly early in life, has far-reaching effects. In a study by Chamove (1989 ), for example, standard laboratory mouse cages were modified by dividing them into 5 to 9 compartments using plastic sheets. The mice showed a preference for complex cages over the standard cages, and cage partitioning led to increased burrowing activity. In addition, mice from the partitioned cages displayed evidence of reduced stress and fear responses, including greater weight gains after weaning, lower adrenal gland weights, and lower fear scores in 2 standard tests, the open field and emergence test. Thus, although the cage modification was relatively simple, the increased behavioral opportunities afforded the mice had striking effects not only on their behavior, but also on their physiology.

Providing Pleasure, Comfort, or Satisfaction

Most animal welfare research has focused on identifying and minimizing causes of suffering. However, behaviors that contribute to animals' pleasure, comfort, or satisfaction have received comparatively little research attention, although they are widely recognized as important components of human well-being. Through surveys, Meyers and Diener (1995 ) identified a number of elements that contribute to life satisfaction in humans. These include a sense of control, meaningful social relationships, challenge, and active engagement.

Studies with animals tend to confirm the importance of these factors. Animals able to exercise some behavioral control over their environment show attenuated responses to a stressor compared with animals having no control ( Weiss 1972 ). Although some effects of conspecific social interactions are described above, research on affiliative and social play behaviors is lacking and greatly needed for most laboratory animal species, particularly nonprimate species. Nevertheless, social relationships with human caretakers have been shown to have positive behavioral, physiological, and immunological effects for several species ( Davis and Balfour 1992 ; Hemsworth and others 1993 ).

Animal preferences provide information about the importance of engagement and challenge. Unless they are quite hungry, animals of many species prefer to work for food rather than eat freely available food, a phenomenon known as contrafreeloading ( Inglis and others 1997 ). Many animals also show a preference for exploring novel environments and objects, even when those environments or objects are not directly associated with needed resources ( Mench 1998a ). Both contrafreeloading and exploratory behavior have an information-gathering function that is likely to be adaptive under natural conditions, but even in captivity these behaviors are still preferred and can be assumed to have reward value for the animal. Lack of active engagement has been implicated as a primary cause of boredom, depression, and anxiety in animals ( Wemelsfelder 1990 ). Zoo researchers have been particularly innovative in designing environmental features that increase the engagement and control that animals have, even in otherwise restricted enclosures ( Markowitz 1990 ).

Behavior As an Indicator

Behavior has another important function: providing information to human caretakers about the welfare of the animal (for example, Manser 1992 ). As noted above, however, thorough observation and a sound knowledge of species-typical, and often individual-specific, behavioral patterns are required to interpret this information. Although behaviors are widely used as indicators of pain or illness in laboratory animals ( Hart 1988 ; Morton and Griffiths 1985 ), crude measures of behavior may fail to correlate well with other measures of pain or distress ( Conzemius and others 1997 ). Nevertheless, well-designed experimental studies can provide information about which behaviors are valid indicators of pain or distress, and even the degree of pain or distress experienced by the animal.

In social species, vocalizations and other social signals may well provide one of the more sensitive indicators, since a function of these signals is to enlist aid from other members of the social group. Weary and Fraser (1995

Emilie C. Snell-Rood

I outline how understanding the mechanism of behavioural plasticity is important for predicting how organisms will respond to rapidly changing and novel environments. I define two major forms of behavioural plasticity: developmental and activational. Developmental plasticity refers to the capacity of a genotype to adopt different developmental trajectories in different environments. Activational plasticity refers to differential activation of an underlying network in different environments such that an individual expresses various phenotypes throughout their lifetime. I suggest that the costs and benefits of these two forms of behavioural plasticity may differ: developmental plasticity is slow, but results in a wider range of more integrated responses. Furthermore, the neural costs associated with activational plasticity may be greater because large neural networks must be maintained past an initial sampling and learning phase. While the benefits of plasticity are realized in variable environments, I argue that fine-grained and coarse-grained variation may differentially select for activational and developmental plasticity, respectively. Because environmental variation experienced by an organism is largely determined by behaviour, developmental plasticity may still evolve in fine-grained environments if niche choice results in coarse-grained 'realized' variation. Behavioural plasticity should impact evolution in novel environments because it increases the chances of survival in these environments. Developmental behavioural plasticity may be particularly important for diversification in novel environments because it can impact not only survival, but also the development of signals and preferences important in mate choice. Future areas of research on behavioural plasticity and rapid environmental change include stress as a mechanism underlying rapid integrated responses and life history perspectives on predicting developmental versus evolutionary responses. © 2013 The Association for the Study of Animal Behaviour.

Andrew Sih

A key issue in animal behaviour is the need to understand variation in behavioural responses to human-induced rapid environmental change (HIREC) such as habitat loss, exotic species, pollution, human harvesting and climate change. Why do some individuals show maladaptive behaviours, while others show adaptive responses to evolutionarily novel situations? At present, we lack a unified conceptual framework for generating predictions and guiding empirical and theoretical work on this critical question. Drawing from the concept of ecological traps, I suggest that a conceptual framework for explaining this variation should include four main points: (1) behavioural responses (adaptive or not) are the result of cue-response systems, or behavioural 'rules of thumb'; (2) limited or imprecise, unreliable information often underlies suboptimal behaviour; (3) the organism's behavioural flexibility affects its response to novel situations; and (4) evolution (and development) in past environments has shaped cue-response systems, responses to imperfect information and degree of behavioural flexibility to be adaptive in past environments, but not necessarily in novel environments. The degree of match/mismatch between past environments and novel environments altered by HIREC is thus a key to explaining adaptive versus maladaptive behaviours. I suggest several existing frameworks that address these four points, and are thus potentially useful for explaining behavioural responses to HIREC: signal detection theory, adaptive plasticity theory, extended reaction norms and cost-benefit theory on variation in learning. I further discuss more complex aspects of reality that would be useful to add to these existing frameworks. © 2013 The Association for the Study of Animal Behaviour.

Daniel Sol | Oriol Lapiedra | Cesar González-Lagos

While human-induced rapid environmental changes are putting many organisms at risk of extinction, others are doing better than ever. This raises the question of why organisms differ in their tolerance to environmental alterations. Here, we ask whether and how behavioural adjustments assist animals in dealing with the urbanization process, one of the primary causes of biodiversity loss and biotic homogenization. Based on a literature review, we present both theoretical and empirical arguments to show that behavioural adjustments to urban habitats are widespread and that they may potentially be important in facilitating resource use, avoiding disturbances and enhancing communication. While a growing number of studies report behavioural differences between urban and nonurban animals, very few studies directly address the underlying mechanisms. In some cases, the changes in behaviour occur very rapidly and involve learning, and hence can be attributed to behavioural plasticity. In other cases, however, it cannot be ruled out that behavioural differences between urban and nonurban animals result from natural selection or nonrandom sorting of individuals by behavioural traits that affect dispersal, habitat selection or establishment. Because the urbanization process is expected to continue to threaten biodiversity in the near future, there is some urgency to improve our understanding of the mechanisms through which behaviour helps animals to cope with such environmental alterations. © 2013 The Association for the Study of Animal Behaviour.

Niels J. Dingemanse | Max Wolf

Behavioural traits are characterized by their labile expression: behavioural responses can, in principle, be up- and down-regulated in response to moment-to-moment changes in environmental conditions. Evidence is accumulating that individuals from the same population differ in the degree and extent of this form of phenotypic plasticity. We here discuss how such between-individual differences in behavioural plasticity can result from additive and interactive effects of genetic make-up and past environmental conditions, and under which conditions natural selection might favour this form of between-individual variation. We highlight how spatial or temporal variation in the environment, in combination with competition among individuals, can promote adaptive individual differences in plasticity; and we detail how differences in plasticity can emerge as a result of selection pressures induced by social interactions or as a response to between-individual differences in state. We further discuss both ecological and evolutionary consequences of individual differences in plasticity. We outline, for example, how individual differences in plasticity can have knock-on effects on the rate of evolution; and how such differences can enhance the stability and persistence of populations. © 2013 The Association for the Study of Animal Behaviour.

Hans Slabbekoorn

Urbanization leads to homogenization of avian communities through local extinction of rare bird species and increasing numbers of the same common urban bird species over large geographical areas. Successful city birds often persist through some sort of behavioural plasticity that helps them survive and reproduce close to humans, in built-up areas, with all the typical urban feasts and hazards. In this review, I address whether behavioural plasticity of the acoustic phenotype can be an additional factor in explaining which species end up as urban survivors. Anthropogenic noise has been shown to negatively affect avian distribution and reproduction, especially for species that rely on relatively low-frequency songs for mediating territorial conflicts and attracting partners for mating. Spectral differences between songs of city and forest populations of the same species and correlations between individual song frequency use and local noise levels suggest that many successful city species shift song frequency upward under noisy urban conditions. Experimental evidence has confirmed the ability of several species to show rapid spectral adjustments as well as perceptual benefits of singing at higher frequency in noisy habitats. However, empirical evidence of fitness benefits for birds showing the ability and tendency of noise-dependent spectral adjustment is still lacking. Furthermore, depending on the species and the underlying mechanism for spectral change, there may also be fitness costs through a compromise on signal function. These two aspects are only two of many remaining avenues for future studies. The acoustic phenotype of urban birds provides a great model system to study fundamental processes such as causes and consequences of environmentally induced signal changes, 'cultural assimilation', and the relationship between phenotypic and genotypic evolution. Furthermore, the current and expected rate of urbanization remains high at a global scale, which will lead to further spread in time and space of artificially elevated noise levels. This should guarantee the continued interest of scientists, politicians and conservationists for many years ahead. © 2013 The Association for the Study of Animal Behaviour.

Laure Cauchard | Neeltje J. Boogert | Louis Lefebvre | Frédérique Dubois | Blandine Doligez

Although interindividual variation in problem-solving ability is well documented, its relation to variation in fitness in the wild remains unclear. We investigated the relationship between performance on a problem-solving task and measures of reproductive success in a wild population of great tits, Parus major. We presented breeding pairs during the nestling provisioning period with a novel string-pulling task requiring the parents to remove an obstacle with their leg that temporarily blocked access to their nestbox. We found that nests where at least one parent solved the task had higher nestling survival until fledging than nests where both parents were nonsolvers. Furthermore, clutch size, hatching success and fledgling number were positively correlated with speed in solving the task. Our study suggests that natural selection may directly act on interindividual variation in problem-solving performance. In light of these results, the mechanisms maintaining between-individual variation in problem-solving performance in natural populations need further investigation. © 2012.

Matthew A. Wale | Stephen D. Simpson | Andrew N. Radford

Acoustic noise has the potential to cause stress, to distract and to mask important sounds, and thus to affect behaviour. Human activities have added considerable noise to both terrestrial and aquatic habitats, and there is growing evidence that anthropogenic noise affects communication and movement patterns in a variety of species. However, there has been relatively little work considering the effect on behaviours that are fundamental to survival, and thus have direct fitness consequences. We conducted a series of controlled tank-based experiments to consider how playback of ship noise, the most common source of underwater noise, affects foraging and antipredator behaviour in the shore crab, Carcinus maenas. Ship noise playback was more likely than ambient-noise playback to disrupt feeding, although crabs experiencing the two sound treatments did not differ in their likelihood of, or speed at, finding a food source in the first place. While crabs exposed to ship noise playback were just as likely as ambient-noise controls to detect and respond to a simulated predatory attack, they were slower to retreat to shelter. Ship noise playback also resulted in crabs that had been turned on their backs righting themselves faster than those experiencing ambient-noise playback; remaining immobile may reduce the likelihood of further predatory attention. Our findings therefore suggest that anthropogenic noise has the potential to increase the risks of starvation and predation, and showcases that the behaviour of invertebrates, and not just vertebrates, is susceptible to the impact of this pervasive global pollutant. © 2013 The Association for the Study of Animal Behaviour.

Andreas P. Modlmeier | Carl N. Keiser | Jason V. Watters | Andy Sih | Jonathan N. Pruitt

The concept of keystone individuals offers a unifying framework to study the evolution and persistence of individuals that have a disproportionately large, irreplaceable effect on group dynamics. Although the literature is teeming with examples of these individuals, disparate terminologies have impeded a major synthesis of this topic across fields. To allow a strict classification of potential keystone individuals, we offer herein some general terminology, outline practical methodological approaches to distinguish between keystone individuals and generic individuals that only occupy a keystone role, and propose ways to measure the effect of keystones on group dynamics. In particular, we suggest that keystone individuals should be classified as 'fixed' or 'episodic' according to the duration of time over which they impact their group. We then venture into the existing literature to identify distinctive keystone roles that generic and/or keystone individuals can occupy in a group (e.g. dominant individual, leader or superspreader), and describe traits that can give rise to keystone individuals. To highlight the ecological implications, we briefly review some of the effects that keystone individuals can have on their group and how this could affect other levels of organization such as populations and communities. In looking at their diverse evolutionary origins, we discuss key mechanisms that could explain the presence of keystone individuals. These mechanisms include traditional Darwinian selection on keystone-conferring genotypes, experience and state- or context-dependent effects. We close our review by discussing various opportunities for empirical and theoretical advancement and outline concepts that will aid future studies on keystone individuals. © 2013 The Association for the Study of Animal Behaviour.

Simona Kralj-Fišer | Wiebke Schuett

Research on animal personality variation has been burgeoning in the last 20 years but surprisingly few studies have investigated personalities in invertebrate species although they make up 98% of all animal species. Such lack of invertebrate studies might be due to a traditional belief that invertebrates are just 'minirobots'. Lately, studies highlighting personality differences in a range of invertebrate species have challenged this idea. However, the number of invertebrate species investigated still contrasts markedly with the effort that has been made studying vertebrates, which represent only a single subphylum. We describe how investigating proximate, evolutionary and ecological correlates of personality variation in invertebrates may broaden our understanding of personality variation in general. In our opinion, personality studies on invertebrates are much needed, because invertebrates exhibit a range of aspects in their life histories, social and sexual behaviours that are extremely rare or absent in most studied vertebrates, but that offer new avenues for personality research. Examples are complete metamorphosis, male emasculation during copulation, asexual reproduction, eusociality and parasitism. Further invertebrate personality studies could enable a comparative approach to unravel how past selective forces have driven the evolution of personality differences. Finally, we point out the advantages of studying personality variation in many invertebrate species, such as easier access to relevant data on proximate and ultimate factors, arising from easy maintenance, fast life cycles and short generation times. © 2014 The Association for the Study of Animal Behaviour.

John C. Wingfield

Coping with perturbations of the environment such as severe storms and other climatic extremes, habitat degradation, changes in predator numbers, invasive species and social disruption is one of the most essential physiological and behavioural processes. The palaeontological record shows that organisms have had to cope with environmental perturbations throughout the history of life on Earth. These ancient processes show highly conserved mechanisms, but also great flexibility in responses to social and physical environment challenges. Adrenocortical responses to perturbations can trigger a coping response called the emergency life history stage (EHLS). However, if the adaptive value of the ELHS declines because of trade-offs with other life history stages such as breeding, then the adrenocortical response to acute perturbations (stress) can be modulated. Mechanisms involve allostasis and reactive scope with three foci of regulation: hormone secretion, transport and response. It is now well known that modulation of the adrenocortical responses to perturbations occur through gene-environment interactions during development and throughout the life cycle. These modulations involve individual differences in gender, age, experience and condition as well as latitudinal, altitudinal and hemispheric variations. Dramatic consequences of human-induced rapid environmental change such as increasing frequency and intensity of environmental perturbations will likely have implications for continued adaptation to extreme events. Note that modulation of the stress response also involves three major processes: modulation of robustness (i.e. become more resistant to acute stress); modulation of responsiveness (i.e. modulate the actual response to stress for more flexibility); and modulation of resilience (i.e. how quickly and completely the recovery is after the perturbation has passed). Mechanisms underlying these modulations remain largely unexplored. © 2013 The Association for the Study of Animal Behaviour.

Lynne U. Sneddon | Robert W. Elwood | Shelley A. Adamo | Matthew C. Leach

© 2014 The Association for the Study of Animal Behaviour. The detection and assessment of pain in animals is crucial to improving their welfare in a variety of contexts in which humans are ethically or legally bound to do so. Thus clear standards to judge whether pain is likely to occur in any animal species is vital to inform whether to alleviate pain or to drive the refinement of procedures to reduce invasiveness, thereby minimizing pain. We define two key concepts that can be used to evaluate the potential for pain in both invertebrate and vertebrate taxa. First, responses to noxious, potentially painful events should affect neurobiology, physiology and behaviour in a different manner to innocuous stimuli and subsequent behaviour should be modified including avoidance learning and protective responses. Second, animals should show a change in motivational state after experiencing a painful event such that future behavioural decision making is altered and can be measured as a change in conditioned place preference, self-administration of analgesia, paying a cost to access analgesia or avoidance of painful stimuli and reduced performance in concurrent events. The extent to which vertebrate and selected invertebrate groups fulfil these criteria is discussed in light of the empirical evidence and where there are gaps in our knowledge we propose future studies are vital to improve our assessment of pain. This review highlights arguments regarding animal pain and defines criteria that demonstrate, beyond a reasonable doubt, whether animals of a given species experience pain.

Jess Isden | Carmen Panayi | Caroline Dingle | Joah Madden

Individuals exhibiting a high level of cognitive ability may also exhibit more elaborate traits and so gain higher levels of mating success. This suggests that selection may act on cognitive performance through mate choice. Studies investigating this relationship have tended to focus on single cognitive tasks, or tasks that are closely related to existing natural behaviours, and individuals are frequently tested in captive conditions. This can introduce test artefacts and may tell us more about selection on specific display behaviours that we imagine being particularly cognitively complex, rather than a general cognitive ability. We tested free-living male spotted bowerbirds, Ptilonorhynchus maculatus, that exhibit elaborate sexual displays which appear to be cognitively demanding. We describe a method for testing individuals in the wild, without the need for constraint or captivity. We looked for evidence of a general cognitive ability in males by assaying their performance in a series of novel tasks reflecting their natural bower-building behaviour (bower maintenance) or capturing more abstract measures of cognitive ability (colour and shape discrimination, reversal learning, spatial memory and motor skills). We related performance in these tasks to their mating success. An individual's performance in one task was a relatively poor predictor of performance in any other task. However, an individual's performance across tasks could be summarized by a principal component which explained a level of total variance above which has previously been accepted as evidence of a general cognitive ability. We found no relationships between an individual's overall performance, or performance in any single task, and mating success. Our results highlight the need for further investigation of whether selection on cognition in bowerbirds is exerted through mate choice. We offer this as an example of how classic cognitive tasks can be transferred to the wild, thus overcoming some limitations of captive cognitive testing. © 2013 The Association for the Study of Animal Behaviour.

Kerryn D. Carter | Jennifer M. Seddon | Celine H. Frère | John K. Carter | Anne W. Goldizen

Many species exhibit fission-fusion dynamics, yet the factors that influence the frequent changes in group size and membership in these species have not been widely studied. Social ties may be influenced by kinship but animals may also form preferred associations because of social attraction or may only associate because they have similar habitat preferences. We investigated the association patterns of 535 wild giraffes, Giraffa camelopardalis, in Etosha National Park, Namibia using behavioural and genetic data from individually identified giraffes. We collected 726 records of group composition over a 14-month period and calculated pairwise association indices, which were tested against a null model. We found that female-female pairs, but not male-male pairs, showed both preferred and avoided relationships. We tested whether females' relationships could be explained by the degree of relatedness between pairs and whether pairs overlapped spatially. Correlations between matrices of pairwise associations, spatial overlap and relatedness showed that female-female associations were strongly correlated with amounts of spatial overlap and pairs that exhibited preferred relationships were more closely related than would be expected by chance. However, only about one-quarter of the variation in observed associations could be explained by spatial overlap and relatedness and therefore much of this variation may have been related to individual social preferences. © 2012 The Association for the Study of Animal Behaviour.

Jonathan N. Pruitt | Lena Grinsted | Virginia Settepani

Understanding how colony-level behaviour is determined is of evolutionary significance because colony-level traits can influence individual fitness and group success. Here we explore how the composition of individual behavioural types within colonies influences colony-level behaviour in the social spider Stegodyphus sarasinorum. First, we tested whether S.sarasinorum show stable individual differences in behaviour (i.e. individual personalities) and whether aspects of individuals' behaviour are correlated across contexts, in the form of a behavioural syndrome. As documented in many other animals, S.sarasinorum showed stable individual differences in behaviour that were repeatable across time, and correlated across contexts (i.e. aggressiveness, boldness). Second, we tested for and confirmed the presence of consistent intercolony variation (i.e. colony-level personalities) in collective foraging behaviour. Third, we generated artificially reconstituted colonies of known group size and personality composition to test for associations between colonies' personality composition and their collective foraging behaviour. In experimental colonies, we found that the average phenotypes of colony constituents were associated with colony-level behaviour, where colonies composed of smaller and bolder spiders were more responsive during foraging. However, the single best predictor of colony-level behaviour was the behavioural type of the single most extreme individual, where the boldness score of boldest individuals explained 66-69% of the variation in colony-level behaviour. Together, our results suggest that variation in the personality composition of social groups may be an important driver of variation in colony-level personality. © 2013 The Association for the Study of Animal Behaviour.

Damien R. Farine

Grouping is a very common outcome of selection that operates on individual animals. Largely considered to be driven by immediate benefits, such as avoiding predators, animal groups often consist of individuals that are phenotypically more similar than expected from the population distribution. This suggests that the distribution and fitness of phenotypes may be shaped by multiple levels of selection operating along different axes of behaviour. Thus, quantifying assortative mixing, or the measure of association between similar individuals in social networks, should be a key component of the biologist's toolbox. Yet, assortment is rarely tested in animal social networks. This may be driven by a lack of tools for robust estimation of assortment, given the reliance of current methods on binary networks. In this paper, I extend existing approaches that calculate the assortativity coefficient of both nominal classes and continuous traits to incorporate weighted associations. I have made these available through a new R package 'assortnet'. I use simulated networks to show that weighted assortment coefficients are more robust than those calculated on binary networks to added noise that could arise from random interactions or sampling errors. Finally, I demonstrate how these methods differ by applying them to two existing social networks estimated from wild populations, exploring assortment by species, sex and network degree. Given the parallel theoretical developments of the importance of local social structure on population processes, and increasing data on social networks being collected in free-living populations, understanding phenotypic assortment could yield significant insight into social evolution. © 2014 The Association for the Study of Animal Behaviour.

Marjolein De Rijk | Marcel Dicke | Erik H. Poelman

Parasitoid foraging decisions are often affected by community characteristics such as community diversity and complexity. As part of a complex habitat, the presence of unsuitable hosts may affect foraging behaviour of parasitoids. First, unsuitable herbivores may affect the localization of patches where hosts are present. Second, encounters with unsuitable herbivores in the food plant patch may affect parasitoid decisions during their searching behaviour in the patch. In this review, we outline the importance of the presence of unsuitable herbivores on the behavioural responses of parasitoids during both these foraging phases. Nonhosts feeding on a neighbouring plant or on the same plant individual the host is feeding from may affect odour-based searching by parasitoids in a way specific for the species combination studied. Feeding by specific host and nonhost-herbivore combinations may induce volatiles that are more, less or equally attractive compared to those from plants infested by the host only. Within the food patch, mixed presence of host and nonhost may reduce the number of hosts parasitized per time unit and reduce parasitoid foraging efficiency. Importantly, we show that a single nonhost species may have contrasting effects in terms of its effects on odour-based searching and patch residence decisions. We conclude that studying host searching behaviour at both phases of foraging is essential for our understanding of parasitoid foraging behaviour in natural and agricultural settings. We further speculate on the ecological context in which unsuitable herbivores affect either of the two phases of parasitoid foraging. © 2013 The Association for the Study of Animal Behaviour.

Lucy M. Aplin | Ben C. Sheldon | Julie Morand-Ferron

Blue tits are famous for the 'milk bottle' innovation, which emerged at numerous sites across Britain in the early 20th century. However, overall we still know little about the factors that foster or hinder the spread of innovations, or of the impact of individual differences in behaviour on social transmission. We used a two-action and control experimental design to study the diffusion of innovation in groups of wild-caught blue tits, and found strong evidence that individuals can use social learning to acquire novel foraging skills. We then measured six individual characteristics, including innovative problem solving, to investigate potential correlates of individual social-learning tendency. Consistent with a hypothesis of common mechanisms underlying both processes, we fou nd evidence for a relationship between social learning and innovativeness. In addition, we observed significant age- and sex-biased social learning, with juvenile females twice as likely to acquire the novel skill as other birds. Social learning was also more likely in subordinate males than dominant males. Our results identify individual variation and transmission biases that have potential implications for the diffusion of innovations in natural populations. © 2013 The Association for the Study of Animal Behaviour.