GREENBERG (2021) Ethologically Informed Design and DEEP Ethology…

Greenberg, N.  (in press 2021)  Ethologically Informed Design and DEEP Ethology in Theory and Practice. In: Health and Welfare of Captive Reptiles, 2nd Edition. C. Warwick, P.C. Arena, G.M. Burghardt (Editors) Cham, Switzerland: Springer.

Abstract The quality of our relationships with other species, as predators, prey, companions, and subjects of curiosity or research, profits from our being ethologically informed. This chapter explores several dimensions of these relationships along with comments on the importance of an ethological attitude and ethologically informed design in pursuit of a better understanding of how best to behave as responsible stewards of other species. Design, in the sense of a coherent program that guides our practice, involves identifying and defining the traits that appear important to us, as well as the ways in which we manipulate, observe, measure, and interpret them. Design both guides and is guided by the questions or problems we wish to address. To be ethologically informed, a design implicitly acknowledges four key biological perspectives, identified in the earliest conceptual beginnings of ethology. Each perspective reflects different temporal and spatial orientations and levels of organisation, but all are profoundly involved in the causation of behaviour; they are: developmental, ecological, evolutionary, and physiological (DEEP). This integrative biology in concert with an ethological attitude, emphasising freedom from implicit bias, are valuable approaches to all forms of captive animal management as well as research design. Such an approach will reveal connections within and between our subjects and ourselves that are of both great intrinsic interest and generalisable utility in solving problems that we all share.

 

Keywords: Reptile, Ethology, Research Design, Experimental Design, Stress, Physiology, Welfare, Captive Care, Animal Models

 

PRELIMINARY DRAFT:

Chapter 12

Ethologically Informed Design and DEEP Ethology in Theory and Practice

Neil Greenberg

 

N. Greenberg

The University of Tennessee, Department of Ecology and Evolutionary Biology, 569 Dabney Hall, Knoxville, TN 37996-1610, USA

e-mail: ngreeneb@utk.edu                                                                          

neilgreenberg.utk.edu

 

Abstract The quality of our relationships with other species, as predators, prey, companions, and subjects of curiosity or research, profits from our being ethologically informed. This chapter explores several dimensions of these relationships along with comments on the importance of an ethological attitude and ethologically informed design in pursuit of a better understanding of how best to behave as responsible stewards of other species. Design, in the sense of a coherent program that guides our practice, involves identifying and defining the traits that appear important to us, as well as the ways in which we manipulate, observe, measure, and interpret them. Design both guides and is guided by the questions or problems we wish to address. To be ethologically informed, a design implicitly acknowledges four key biological perspectives, identified in the earliest conceptual beginnings of ethology. Each perspective reflects different temporal and spatial orientations and levels of organisation, but all are profoundly involved in the causation of behaviour; they are: developmental, ecological, evolutionary, and physiological (DEEP). This integrative biology in concert with an ethological attitude, emphasising freedom from implicit bias, are valuable approaches to all forms of captive animal management as well as research design. Such an approach will reveal connections within and between our subjects and ourselves that are of both great intrinsic interest and generalisable utility in solving problems that we all share.

 

Keywords: Reptile, Ethology, Research Design, Experimental Design, Stress, Physiology, Welfare, Captive Care, Animal Models

 

Introduction

 

We are all, of course, deeply concerned with care and welfare of animals we live and work with, as well as those we wish to preserve, nurture, and study as subjects in scientific research. Whether behaviour is manifest in nature or the home or the zoo or laboratory, the welfare of our subjects is more than an ethical imperative, but one that also affects both the positive qualities of our experience: the effectiveness of veterinary care and the validity of research, particularly when the subjects are  unfamiliar. The effect of studying unfamiliar subjects is two-fold: we are drawn to find shared qualities such as details of physiology and behaviour but also the very alien nature of these fellow creatures challenges us to enlarge our understanding of the boundaries and possibilities of life itself. 

 

Our concerns in this volume are mainly with health and welfare of animals subject to our control in captive circumstances ranging from private homes to zoos, from nature preserves to laboratories.  It is our belief that as a general rule, our concerns are best served by eliminating or mitigating the distortion attributable to our intrusion into their lives, but this necessarily involves the best possible understanding of the undisturbed animal. The view adopted here is that this is best accomplished by adopting an ethological attitude and implementation of an ethologically informed design (EID) — a  program to guide and support practice by means of the integrative perspectives of DEEP ethology. Hopefully, the result of this approach is an ethologically informed practice for any project that involves our influence on other species, from pets or companions, to subjects in education and conservation, veterinary science, and experimental research. In saying more about these ideas, I will build on and extend a previous version of this essay that emphasised ethologically informed design – an idealised concept that can inform real-world practice (Greenberg 1995).

 

As described below, these integrative perspectives can contribute significantly to our ability to predict behaviour attributable to circumstances ranging from the exigencies of nature to anthropogenic perturbations. In all cases, the quality of our stewardship will profit significantly from our knowledge of the ethology of the organisms in our care. Even in field research, our best approximation of a ‘natural experiment’ (one in which variables are in play spontaneously), will always have a measure of artificiality attributable to our presence. The best we can do is to mitigate confounding or misleading variables according to our awareness of them and the resources for alternatives.

 

This program challenges us to be mindful of the traits of interest and their place within the cascades of information that flow through multiple levels of organisation—from molecular to organismic and environmental—in animals that may be very different from ourselves. We must ask how similar or different these traits are from the countless reference traits we hold in mind or the recorded notes of our own reconnaissance and that of other scholars. We need to be mindful also that features distinguishing species—even individuals—are affected by our own experiences and perceptions, and can thus never be fully extricated from each other. 

 

As I hope will be clear, amongst our most urgent concerns as stewards, scholars, and scientists is to see phenomena for what they truly are and not in terms of anything else. This is a concern of scientific practice in general (Shah et al. 2017) and resonates with and will build on some of the tenets of the philosophical tradition of phenomenology, a school of thought that in practice emphasises the avoidance of bias and preconception, the primacy of perception, and an emphasis on the life-worlds of individuals. By life-world, I mean the aggregate of congenital traits and previous experience that the individual brings to its current circumstances. Concern with bias recalls the now familiar sins of anthropomorphism, speciesism, and human exceptionalism, but those are just the most convenient of a litany of convenient examples. The constant interplay of subjective and objective perceptions and of deductive and inductive reasoning are essential elements in the ensemble of cognitive processes by which we function as organisms and as individuals. Amongst these cognitive processes, certain constellations may first appear as competing alternative ways of thinking (such as ‘art’ and ‘science’ or ‘theory’ and ‘practice’).  However, their boundaries, are inevitably lost in the minutiae revealed by close observation and their inevitable intertwining when applied to actually being in the world. So, while specific constellations cannot be expunged, we can be mindful of their seeming conflict at levels that will enable us to avoid error or harm.

 

Hopefully these ideas frame an approach to an understanding of other organisms sufficient to our task of assuring their welfare in all our possible interactions with them. This task begins by describing behaviour—either for its intrinsic interest or as a dependent variable in our pursuit of clues for a possible explanation of its causes and consequences. In either event, our best efforts will profit from an ethological attitude (Table 1), including an approach that integrates multiple streams of biological information or DEEP ethology (Table 2). Further, the complexity of the behavioural patterns and processes that are integrated in any phenomenon of interest and to which we must attend as part of our efforts at understanding requires an ethologically informed design (EID) (Table 3), all to be discussed below.

 

Ethological Attitude

 

As a matter of everyday routine, most of us maintain a ‘natural attitude’, grounded in our subjective experiences of an objective world and even of ourselves. We act to navigate our world based on these experiences and the expectations they engender. Experiences that fail to meet any part of the test of expectation create a stressful dissonance that we avoid with all the cognitive resources available—both nonconscious as well as elaborately calculated.

 

The ethological attitude, most simply put, involves our best effort at bias-free observing, documenting, and analysing a unit of behaviour or a pattern of units by means of an integrative biological approach to its causes and consequences (DEEP ethology, described below). The integrative approach requires that we take every effort to interpret the causes and consequences of phenomena in both bottom-up and top-down terms. These resonate with analytical and integrative approaches that are often viewed as alternatives with a long history of antipathy, but that are, in fact, intertwined: the ethological attitude views specific acts of behaviour in the context of their participation in larger patterns as well as being the outcome of interactions at successive subordinate levels. The effort (if not success) in perceiving these perspectives simultaneously is often called Janusian, referring to Janus, a Roman god of doorways and dualities, discussed below.

 

Table 1. Key Characteristics of the Ethological Attitude. ABOUT HERE

 

The eschewal of the biases that subvert valid understanding is a crucial element of the ethological attitude. This sounds simple, but biases exist at many levels of organisation of which conscious attention is only the most obvious, but they can be particularly difficult to identify and cope with when they have roots deeply conserved in our evolutionary biology and ingrained throughout our social and cultural development. Biases may be implicit as well as explicit and not the least of these is human exceptionalism—the venerable and persistent view that there is a profound discontinuity between ourselves and the rest of the natural world. Of course, there are discontinuities between species—indeed between all categories—but the paths that lead to them are (or should be) subject to scientific scrutiny. However, this scrutiny must be bi-directional or we are handicapped by apparently conflicted alternatives. Awareness of these (if not their suppression) can be difficult and we are likely best served, as Kuhn (1959) would put it (speaking of divergent and convergent thinking), by cultivating the ability to tolerate tensions “that can become almost unbearable but are one of the prime prerequisites for the very best sort of scientific research.” The single mind looking in two directions, the ‘Janus face of science’ (Burghardt 2013) has been identified as contributing significantly to a wide range of creative activities (Koestler 1978; Rothenberg 1979), but crucially for us, the Janusian perspective of simultaneously considering causes and consequences—looking up and down the apparent chain of events at  appropriate levels of organisation—is a key element of the ethological attitude. In my own research, the Janusian perspective was first apparent in simultaneously considering the reciprocity of top-down and bottom-up neurological causes and social behavioural consequences of specific units of behaviour.

 

We can fairly say that the ethological attitude is also the basis for ethologically informed design and practice. It includes explicit strategies to be deployed whenever we undertake organised efforts to better understand, remediate, repair, or otherwise secure the wellness of animals we encounter and for which we have responsibility. Such an approach has the added advantage of minimising our effects on an animal’s experience when we intrude into their lives to conduct experimental research, or prepare or modify their environment when in captivity, or determine causes for a health problem. Indeed, distorted experimental findings in particular have a way of cascading through the community of researchers and subsequent experiments in a pernicious and wasteful way. 

 

This attitude is also informed by comparable principles developed in phenomenology, the philosophy that undertakes to include subjective experience in the perception of phenomena.  Phenomenology as presented and developed as opposed to traditional philosophy in the last half-century has had considerable success, but has regrettably neglected relationships with non-human animals. Important exceptions to that neglect are driven in part by the self-created existential crises for ourselves and nature in general and may enable healthier relationships, catalyse more productive understandings, and hopefully mitigate the environmental difficulties in which we are embroiled; see, for example, Phenomenology And The Non-Human Animal: At The Limits Of Experience, where Painter and Lotz (2007) point out the harmful ideology of human exceptionalism and the need for mindful ethics in dealing with non-humans. The seminal philosopher of phenomenology, Edmund Husserl (1970), believed that the attempt to separate observations from contaminating ideology is crucial and involves considerable personal effort, the outcome of which would result in the adoption of the deceptively simple ‘phenomenological attitude’ (see Greenberg et al. 2019). Nothing less than this is required to invoke an ethological attitude, distinctive by its eschewal of bias—most conspicuously human exceptionalism—and acknowledging and suspending other more subtle, but comparably misguiding, assumptions about the natural world and its constituent processes, not  least of which is its conformation to traditional, often arbitrary ideals. Thus, in ethology as in phenomenology, observers or researchers avoid idealisations or generalisations about their subjects and rather emphasise real animals in their real worlds. Here, the important concept of epoché—with its rigorous eschewal of bias—is applied. The concept is central to phenomenology but is, in the view of the physicist Piet Hut (2001), common amongst creative scientists. A related core element of phenomenology is ‘bracketing’—a much more general  setting aside of the burdens of acquired hypotheses and theory, of bias and expectations, in order to enjoy a greater clarity of perception that will enable us to conceptualise major revisions of received theories.  Further, practitioners can better focus on securing remediation, recovery, and future welfare of specific subjects. 

 

In an earlier report (Greenberg 1995) I made much of the pernicious effects of neglecting independent natural variables associated with the life history of the animal subject in any situation involving their care and welfare, not least in scientific inquiry. However, this neglect is sometimes strategically calculated when an ethological inquiry evokes ‘investigative optimising’ as researchers seek to balance urgency of question, available resources, efficiency, and effectiveness in the conduct of their work and the analysis of their findings. For example, zoological facilities are particularly eager for captive reproduction of their animals, but the expense and effort needed to cope with the many—often very sensitive—elements of reproductive behaviour require balancing their mandate to educate the public with that of conservation. 

 

Questions asked at different levels such as the proximate (e.g. physiological) and ultimate (e.g. evolutionary) causes and consequences of a specific unit or pattern of behaviour are characteristically answered by different modes of analysis, but when question, process, and mode of analysis are not carefully matched, much sterile controversy may be generated (Sherman 1988).  This is often attributable to the conflating or misattributing of levels of organisation. The familiarity of lists of levels such as cells to tissues to organs can  obscure the fact that to specialists, functions at each level can often be subdivided (as particularly prominent in the brain, Freeman 1995; Goebel 2014). Within a more proximate domain, levels from gene transcription to cellular activity have been sketched out for the stress response at levels that precede levels subject to specific adaptive contexts (Kassahn et al. 2009). A multitude of natural variables are much too infrequently considered conjointly because research methods and historical traditions have served to isolate them. While isolation can be an important component of the analytic part of creative inquiry, it cannot be allowed to exclude the integrative part and dominate the discourse. I turn now to the integrative vision of biology applied to behaviour that inspired the earliest ethologists and that is still central to effective and productive animal care.

 

Deep Ethology

 

The expression of behaviour at any given moment occurs at the intersection of four key processes or factors, each reflecting a different level of organisation through which biologically significant information flows in the causation of behaviour: developmental, ecological, evolutionary, and physiological (DEEP) (K.H. Greenberg 2019). This task begins by describing behaviour—either for its intrinsic interest as a trait or as a variable in our pursuit of the causes and consequences of some phenomenon of interest. Descriptions done well become a shared vocabulary for research and its application in diverse domains. Descriptions of behavioural patterns and their relationships are often listed in ‘behaviour inventories’ and ‘ethograms’ (exemplified in Greenberg 1978)

 

Description in hand, ethologists consider causes and consequences of behaviour and advances most productively by conceptualising a unit or pattern of behaviour as emergent at the intersection of the four DEEP domains that correspond to the aims identified in the earliest conceptual beginnings of ethology by Tinbergen (1951) and continues to guide ethological thinking.

 

A fifth aim, termed ‘private experience’ has been ventured by Burghardt (1997). This was influenced by Jakob von Uexküll who embedded the ‘inner world of the subject’ into his functional circle that emphasise the animal’s perceptual and phenomenal worlds (Burghardt 2020). A means of accomplishing this aim is to apply a ‘critical anthropomorphism’,  in which our human derived hypotheses are informed by what we scientifically know about the species’ natural history, ecology, physiology, perceptual and cognitive abilities,  as well as the prior experiences of the individual animals themselves.

More recently, the idea of a ‘fifth aim’ was also advanced at the 50th International Congress of the International Society for Applied Ethology (ISAE), held in Edinburgh, Scotland in 2016 (Siegforda et al. 2018). This most recent effort seemed unaware of the more theoretically informed proposals a generation earlier (Burghardt 1991) and was inspired by advances in detecting and analysing the shared physiology as well as the apparently homologous emotions of many animals. This approach  reflects the belief that a sense of shared neurocognitive architecture (see Font et al. this volume) as well as outward appearances of emotion imparts greater confidence in such an interpretation (see de Waal 2011; de Waal and Preston 2017). 

 

While we might conceptualise every act or pattern of behaviour as occurring in terms of the four perspectives of DEEP ethology, we must also remain mindful that each perspective has its own tradition and culture. Each observes behaviour at a different level of organisation, including raw perception (Brewer 2015), and each asks different questions, different standards of evidence, and different judgements as to what constitutes a satisfying story. Thus, these disciplines—individually as well as in various combinations entail more-or-less cognitive bias. The following paragraphs summarise the character of each discipline as commonly conceptualised by biologists, its assumptions, keywords, and kinds of questions that might be addressed. This is the approach used in Greenberg et al. (2019) to help naturalise phenomenology for researchers employing qualitative research methods. These sibling disciplines are perhaps familiar to most biologists, but I will try to emphasise those aspects that are highly suited to make connections with other domains. They are identified with examples of questions they address in Table 2.

 

Table 2. Disciplines that Converge on an Integrative Biology of Behaviour. ABOUT HERE

 

Development

 

Development involves all the processes that unfold within an individual organism from conception to demise. Understanding the developmental trajectories of animals we would better understand is essential to any long-term relationship with them. These processes involve the cascades of information encoded in the genes inherited from the previous generation(s) as well as those attributable to individual experiences within one’s lifespan. Activation of specific genes and the programs of change they instantiate are more-or-less tolerant of environmental influence (the ‘open’ or ‘losed’ genetic programs of Mayr 1974). For example, programmed developmental changes in diet or temperament or environmental needs must be understood for any companion animal, conservation, or research program.  

 

Developmental experiences such as isolation, agonistic trauma, dietary deficiencies, or deprivation of key stimuli during sensitive periods can all have persevering effects (see Burghardt and Layne 1995; Burghardt and Layne-Colon this volume; Mendyk and Augustine this volume). At present, studies involving other taxa offer some insight into possible welfare issues in reptiles. For example, in the laboratory, a well-meaning ethologist saw to it that newly hatched birds in the nest were well fed, only to see their parents reject them. Later, it was learned that in nature, hatchlings that do not beg for food are often discarded by their parents, presumably because they are sick or dead (Eibl-Eibesfeldt 1970). Acoustic qualities of various habitats such as forest, edge, or grassland are known to affect bird song, critical in many species for successful reproduction (e.g. Morton 1975).  Such selection is likely also when attributable to anthropogenic, urban sounds (e.g. Luther and Baptista 2010). The effects of history on an animal need not, of course, be so dramatic or limited to a particular context. In a study of several vertebrate taxa including reptiles, it was found that handling animals each day negatively affected weights of reproductive organs and fat stores (Meier et al. 1973). The degree of the effect depended on the regularity and time of handling relative to photoperiod, and was presumably a result of transient mild stress. The likely effects of even mild stress on metabolism reminds us that it is reasonable to expect significant spontaneous variability in non-domesticated species. Animals in the wild are certainly subject to occasional stress, and behavioural and physiological responses to stressors are so intimately involved with each other (see Hennessy and Levine 1979; Moore and Jessop 2003; Martinez-Silvestre 2014; Gangloff and Greenberg this volume; Figure 1), that there may be many ways that such episodes may even be necessary for the proper maturation of coping mechanisms.

 

Animals in relationships with humans learn a diversity of behavioural patterns that reflect their experience and efforts to cope. The biological and circumstantial ‘constraints on learning’ (Shettlesworth 1972) underscores the diversity of species and circumstances: for example, reptiles vary in the sensitivity of specific sensory modalities and the salience of any stimulus may vary with the animal’s evolved competencies as well as transient motivational state (see Crowe-Riddell and Lillywhite this volume). Learning is always keyed to experience in context and will be commented further in the following paragraphs on ecology (see also Font et al. this volume).

 

Ecology

 

It has become a truism that the behavioural patterns seen in animals are profoundly affected by the context in which they occur. The environment includes the temporal and spatial physical and biotic contexts in which organisms must survive and thrive. It is also the source of all perceptions that organisms use to create their reality that is the basis for adaptive behaviour, their umwelt, von Uexküll’s (1909) invaluable term, that emphasies each species uniquely perceived construct of ‘nature for themselves according to their special needs’. In recent years the concept of ‘embodiment’ (especially developed in phenomenology) has brought to the foreground the extent to which mental processes go beyond mere sensing of an organism’s physiological state to an ‘intertwining’ of mental and physiological processes such that study of the causation of behaviour must look beyond the nervous system. This concept began with innovative ways of thinking about emotion in humans (e.g. James 1884), but has come to attract significant attention in phenomenology (e.g. Johnson 2007). This phenomenal way of thinking emphasises the unity of mind and body in fruitfully integrative ways such that when Janus faces in both analytical and integrative directions, the questions can be more informed and incisive.

 

Ultimately, our understanding of animal welfare hinges on our understanding of animal needs, but how can these needs be known? Further, the perceptual world of an organism is necessarily experienced in a uniquely self-centered way that affects the meaning of experiences. Thus, even in identical environments, different individuals can perceive, integrate, and act in unique ways. The ethological attitude would include consideration of each animal’s individual umwelt, an effort that necessarily involves, as mentioned above, an acute awareness of one’s personal biases as well as the disposition to characterise one animal in terms of a conspecific that may have a different developmental history.

 

At every level of organisation, every distinguishable element of life—from the multiplicity of organelles within a cell through the outermost boundaries of an organism—is embraced—embedded—in protean concentric spheres of the matrix of the world. In our concern for the behaviour of animals we relate to, this idea echoes principles that have entered contemporary phenomenology as ‘embodied cognition’ and ‘socio-cultural embeddedness’ and builds on the understanding that all cognitive processes involve proprioceptive, interoceptive, and exteroceptive input (Lakoff and Johnson 1999; Johnson 2007), as well as the influence of their place in the biotic world of conspecifics, predators, and prey. Output, then, may be manifest as behaviour when the need to cope with change exceeds automatic physiological responses as, for example, when one changes behaviour as a result of experience – that is, learning.

 

The overlap between development and ecology is particularly evident in learning, and reptile learning provides some of the most striking examples of context-dependent behaviour. The well-known difficulty in training many reptile species has engendered a sense of their relative inability to learn (see Burghardt 1977). However, some lizards manifest a surprising capacity to learn in response to ecologically relevant stimuli (Brattstrom 1978; Font et al. this volume). They are, in Seligman’s (1970) term, ‘prepared’ to make certain associations much more easily than others. Suboski (1992) suggests that the assumption that learning in reptiles is ‘impoverished’ is attributable in part to an inappropriate model of the learning process that neglects the importance of action-specific releasing stimuli. As Burghardt (1977) pointed out, expressions of behavioural plasticity such as learning are subject to gross misrepresentations by researchers if considered apart from ethological variables such as physiology or ecology. For example, in a field study of the curly-tail lizard (Leiocephalus schreibersi), by Marcellini and Jenssen (1991), researchers observed that one-trial acquisition of novel predator (human) avoidance behaviour was manifested by 80% of the animals tested. Interestingly, long flight distances developed more rapidly among females than males, suggestive of possible endocrine corollaries.

 

Many reptiles habituate to human observers and appear to behave naturally, that is in a manner comparable to that of reptiles covertly observed (e.g. agamas [Amphibolurus sp.], wall lizards [Podarcis muralis], side-blotched lizards [Uta stansburiana], and some Anolis spp., see Sugerman, 1990), but some species appear resistant to habituation (Sugerman and Hacker 1980). In the presence of an observer, captive collared lizards (Crotaphytus collaris) will markedly reduce activity levels (Sugerman and Hacker 1980), but the obtrusiveness of the observer is important. Even in the relatively insensitive anoles (Sugerman 1990), an observer will increase the duration of tonic immobility (Edson and Gallup 1972; see Gallup 1974 for review), but probably in response to eye contact (Gallup 1973). Further, the duration of the immobility manifested by anole lizards (A. carolinensis) in response to apparent predation (human handling) is significantly diminished in subsequent trials (McNight et al. 1978; and see Hennig 1979).

 

While the ecological variables of predator-prey dynamics, geology, climate, and physical habitat are important in many obvious ways, the subtlety of some of their interactions can be surprising: for example, slight differences in temperature can elicit alternative defensive responses to prospective predators (humans) in the anole lizard (Anolis lineatopus) (Rand 1964). Approach distances (how near a potential predator may approach before triggering an escape response) are generally reduced at lower body temperatures (Rociia and Bergallo 1990). The adaptive rationale was that neuromuscular responses were impaired at lower temperatures. Two other anoles (A. cristatellus and A. stratulus) were observed to have significantly different approach distances in comparable thermal environments that appeared most likely correlated with the degree of crypsis they manifested: the more cryptic anole (A. stratulus) had a shorter approach distance (Heatwole 1968). In anoles (Anolis carolinensis), the presumably defensive strategy of immobility when a prospective predator approaches is of significantly longer duration when foliage is nearby than when it is absent, but only in the early days of captivity (Hennig 1979). The transfer of animals from the field to the home, zoo, or laboratory is likely to result in altered behaviour, but we cannot know without preliminary field work if the variations in behaviour are within the range spontaneously manifested by the animal in nature. This idea of a range of tolerance resonates with what in stress research is referred to as adaptive scope, the range in which a stressor is accommodated without the necessity to move to a higher level of organisation (Greenberg 2002, Gangloff and Greenberg this volume, and see below). 

 

Evolution

 

Evolutionary biology is concerned with the change in traits and organisms and societies across generations, from ancestors to the present moment, and forward to direct and indirect descendants. Traits are understood to have their present form because of their preservation through the processes of natural selection of variations that are found adaptive—that is, able to compensate for environmental forces (often called ‘selection pressure’) that affect their ability to meet needs. Adaptation is at the center of concern. There are several definitions of adaptation, and all are unified by the idea of compensation for change, either short-term changes at the developmental level (such as a stimulus or life experience), or long-term changes when natural selection preserves them in subsequent generations (such as coping with changing climate) (Rappaport 1971). In a recent report, anoles (Anolis sagrei) were observed in experimentally established colonies on small islands with and without predators (Lapiedra et al. 2018). The researchers found that both behaviour and morphology were rapidly changed simultaneously and independently, and that differences in survival between males and females was likely attributable to differences in habitat use between the sexes.

 

A poignant exercise of Janusian thinking evoked by the evolutionary perspective is the ‘reverse engineering’ or deconstruction of the path taken by ancestors while simultaneously speculating on the advantage the trait may convey to descendants. There is a certain resilience in many evolved traits such that organisms need not be perfect (‘deal’), but only ‘good enough’ to be better than competitors. This is another example of ethologists, like phenomenologists, prioritising real individuals over ideal ones. Efforts to understand the source of this resilience in both epigenetically evoked alternative paths and the adaptive scope between levels of organisation through which information moves, are in themselves illuminating exercises.

 

Physiology

 

Physiology, the fourth domain of DEEP, describes the proximate processes by means of which  internal and external changes are undertaken. This domain includes the sensory detection of environmental and internal information from the body, its integration in conjunction with everything else the organism experiences (or may have experienced in the past) as well as expectations for outcomes, and then the selection (or suppression) of appropriate actions. 

 

The processes within and between organ systems, including the many functionally specialised components of the nervous system, are dynamic and in continual pursuit of balance (homeostasis).  However, such a state (known in human biology as ‘organ cross-talk’) is subject to constant ecological and developmental change, and thus can at best be only temporarily attained within a range of tolerance such that animals may appear outwardly stable. Because these processes are tightly integrated with memory as well as anticipated outcomes, various levels of error detection are an integral component. Most of an organism’s actions are directly or indirectly necessary to provide life-essential factors. Whether congenital or acquired, many such actions may be collateral or incidental to their main function. The microevolutionary process of ritualisation characterises many adaptive behavioural traits as having been transformed by natural selection from wholly unrelated functions (Hinde 1970). Many of these actions originate in homeostatic reflexes of the sympathetic nervous system as ways to cope with stressors. Together with the structures they utilise, ritualised behaviour often contributes to confusion when we try to tease out causal relationships. Ritualisation begins because the processes evoked are available for natural selection in serving other, often unrelated functions. Interpretations of ritualised behaviour is also complicated by the fact that these traits evolved in environments very unlike anything we might recognise today. As a result, there is often a mismatch between past utility and contemporary function.  

 

Many of the physiological processes with which we are concerned as proximate influences on overt behaviour are elaborated expressions of the routine maintenance of the organism in the course of moment-to-moment circumstances and the ebb and flow of diurnal, seasonal, and developmental needs. Beyond routine, the physiological processes are often recognised as stress, that as commented earlier (and see Gangloff and Greenberg this volume), is evoked by a real or perceived challenge to our ability to meet our real or perceived needs. The physiology of stress and its capacity to balance or reconfigure the cognitive processes associated with motivation can be discussed in conjunction with a biological interpretation of Maslovian motivational needs that are based in homeostasis and have biological fitness as an ultimate outcome (see below).

 

Ethologically Informed Practice

 

Applying the ethological attitude to the four domains of DEEP ethology with respect to the behaviour of animals affected by our relationship with them provides the best vantage point to assure their welfare.  Ethologically informed design (EID) is an idealised application of ethological principles that must be adapted to specific projects, particularly where research findings can inform cascades of subsequent work (Greenberg 1995). In practice, the ethological attitude is brought to bear on a broad spectrum of concerns for stewardship, husbandry, welfare, and research. A thoughtful application of this attitude will alert us to dimensions of practice that are unsuspected or even counterintuitive. For example, learning that is manifest by animals with which we may interact is often subject to unsuspected and uniquely species-specific traits generally involving experiences of specific kinds of stimuli in specific contexts (Seligman’s ‘preparedness’, mentioned above); or specific qualities of stimuli (such as Gibson’s ‘affordances’ 1979), elements of the stimulus to which an animal is particularly responsive. 

 

The questions associated with each domain of DEEP ethology (see Table 2) provide the beginning of a framework for characterising the causes and consequences of animal behaviour. Treating this issue as a pre-clinical or pre-treatment checklist may inform the practitioner or researcher in assessing how prepared they are to provide for the corresponding biological needs of the animals for which they have taken responsibility (see also Jessop et al. this volume). In the network of causes and consequences in which a behavioural pattern is manifest, any change in structure or process will ripple through the system with variable effect. An assumption in ethologically informed design (EID, Table 3) is that the system as it exists in nature possesses structures and mechanisms that are capable of buffering or otherwise coping with change as may have affected the organism’s fitness in ancestral environments. In modern or artificial environments, on the other hand, coping can become dysfunctional and our observations misleading.

 

This buffering is an expression of ‘adaptive scope’ mentioned above, the amount of tolerance for stress within any of the many levels of organisation within an organism before another, ‘higher’ level—typically more accessible to observation–needs to be invoked (see Gangloff and Greenberg this volume). This tolerance often proceeds unknown to caregivers or researchers. The level at which an effect of stress may be apparent to the researcher may not be the level at which it would be best addressed as a responsible keeper or clinician.

 

TABLE 3.  Ethologically Informed Design   ABOUT HERE:

 

Validity

 

Considering the potential distortions imposed on a captive animal’s development, relations with its environment, expressions of biological fitness, and physiological processes such as stress and reproduction, an EID informs the procedures used in animal care. Such ethologically-informed practice will have more consistent and reliable outcomes across a broader spectrum of applications. At best it alerts the practitioner to the diversity of possible responses and instills respect for the uniqueness of species and even individuals. Validity is not the same as reliability in which outcomes can be replicated frequently, but do not necessarily speak to the real animal in its real world.

 

It is easy to speak of anthropogenic distortions imposed by the way that we care for captive animals, but in some respects the research environment is arguably the most challenging of circumstances. Here, formal concerns regarding the validity of information that we deploy or acquire are important and deserve comment. The deployment of our idealised design in practice requires an important distinction to be made between ‘internal’ and ‘external’ validity: the first is highly reliable in a narrow context (e.g. an individual animal in a specific context); and the second, somewhat generalisable to other individuals, species or taxa, or even the same individual in a different context. Because, research often relies on specific animal models, generalisability of the information it provides is a fundamental concern of science in everyday practice. Researchers in medicine and drug development have learned—sometimes painfully—that findings in a model species do not always map on to humans (Pound and Ritsket-Hoitinga 2018); they are at best a start.

 

The proximate and ultimate causes and consequences manifest in DEEP ethology are concerns that help keep the researcher and the practitioner at an appropriate level of organisation for the questions being asked or problems to be solved. The immediate or physiological causes and consequences of behaviour are often termed ‘proximate’ to distinguish them and their levels of organisation from evolutionary, or ‘ultimate’ causes—the presumably naturally selected adaptive background to the proximate cause.

 

The immediacy of concern is the area where these terms are relevant: mitigation of pain, restoration of health, or successful reproduction and maximising biological fitness. The welfare of animals utilised in research is typically characterised in terms of apparently dysfunctional behaviour or expressions of distress or pain either in the course of captive maintenance or through a scientific procedure (reviewed by Morton et al. 1990). For example, hierarchically organised  ‘pain scales’ have been suggested (Orlans 1990), but as Bateson (1991a) points out, both the evolution as well as the functions of a subjective sense of pain is obscure. To assess animal well-being, both physiological and behavioural measures have been suggested and an enlightening debate has been engendered about their respective complexities and relative merits (summarised in part by Barnett and Hemsworth 1990). Novak and Suomi (1988) identified four different approaches to defining psychological well-being: physical health, behavioural repertoire, reaction to stress, and responsiveness to environmental events. Similarly, the importance of integrating local resources and the history and behaviour of specific organisms in the development of a maintenance or treatment plan cannot be negected (e.g. Rose and O’Brian 2020)  

  

However, some of these approaches, as many authors point out, lead to contrary conclusions about how best to manage captive animals. Even within a specific approach such as reaction to stress. Within an individual or small population, the consequences of sustained stress changes from positive to negative as physiological stress levels increase (see the Yerkes-Dodson ‘inverted-U’ phenomenon (Gangloff and Greenberg this volume). In addition, the great variability in the expression of distress between taxa makes generalisations overly dependent on extrapolation from more familiar to less familiar models. For example, when a presumed stressful stimulus (a live snake) is presented to squirrel monkeys, they will appear highly agitated and apparently stressed, but there is not necessarily an elevation in the stress hormone, cortisol, over what would result from mere novelty (Vogt et al. 1981). The emotional aftermath of a presumably stressful situation suggests that there may be a discontinuity between what we may perceive as a reflexive response and its affective component (Prinz 2003).

 

Behavioural and physiological perspectives have suggested various ‘objective’ criteria for welfare as well as some sterile controversy, but the ethological appreciation for the complexities of adaptive change may help resolve differences. For example, well-meaning efforts to maximise an animal’s welfare by minimising an ‘objective’ measure of stress such as elevated circulating levels of adrenal hormone may deprive animals of opportunities to exercise and develop responses needed to deal with the normal variability in their environments. While not all needs are of comparable importance to welfare (Greenberg 1992), we could argue that the  ‘ultimate biological need’ of any animal is the realisation of its maximum biological potential —its inclusive fitness, even where this involves a non-reproductive state. Long-term expressions of stress in animals that may appear healthy and stable may become manifest in phenomena such as retarded development and impaired reproduction (Tokarz and Summers 2011), and behavioural dysfunctions (see Christian 1980), as well as more subtle phenomena such as accelerated aging and impaired immunocompetence, (see Johnson et al. 1992; Gangloff and Greenberg this volume).

 

Laboratory and Field

 

For animal welfare in any context, not least the exquisitely sensitive context of experimentation, the scrutiny of presumably relevant variables while all other possible influences on dependent variables are held constant can lead to heroic efforts for control, if not uniformity, of all potential variables. However, the complexity of an animal’s interdependence with its natural habitat may dictate that meaningful study is possible only in nature or if necessary in a simulated habitat (Warwick and Steedman this volume)—even if this trade-off entails the sacrifice of fine control of potentially relevant variables. An extreme example of faulty design was identified by Warwick (1990) who observed that restrictive environments that do not allow sufficient exercise may lead thereby to an array of behavioural and physical problems.

 

It has long been appreciated that many debilitating dysfunctions in zoological parks are attributable to an incomplete accommodation for the needs of captive animals (for example, Hediger 1950, 1955). The concerns of zoological parks have undergone a marked transformation; modern zoos manifest much more interest in subtleties of welfare and behaviour, particularly when they impact the reproductive effectiveness of endangered species (see Eisner 1991). Special scrutiny has been afforded some of these problems in recent years with respect to environmental ‘enrichment.’  For example, Mahoney (1992) espouses a systematic investigation of the sensory capacities of captive animals and encourages accommodation of specific needs and aversions (see also Mendyk and Augustine this volume; Warwick and Steedman this volume).

 

In many contexts we, as stewards of welfare of the animals under our influence, are satisfied when practices apparently meet basic biological needs, including reproduction. In the more narrowly ‘problem-focused’ context of the clinic and the research laboratory, we must be able to contrast the specific problem with normalcy revealed by field studies. Ethologists typically do this by means of an inventory of units of behaviour and an ethogram of how these units may be organised in patterns of behaviour (Greenberg 1978). If sufficiently complete, these approaches will show the full range of traits that have evolved to cope with, even rare, challenges (Wingfield et al. 1998).

 

Frequently, even subtle aspects of an artificial setting may affect behaviour in ways that underscore the importance of previously unnoticed or under-appreciated environmental influences on behaviour in nature. For example, wavelength and intensity of ambient light may affect agonistic behaviour of some lizards (Moehn 1974; Mancera and Phillips this volume), environmental scale may affect timing of activity patterns of captive anoles (DeLong et al. 1986),  interaction of light and temperature affects reproductive activity in male anoles (Licht 1967),  humidity may affect reproductive activity in female anoles (Summers 1988; and see Stamps 1976), and mild stressors may affect some but not all of forms of tongue-flicking manifested by lizards — presumably modifying their awareness of certain environmental cues, but also expressing non-specific arousal (Greenberg 1985).

 

Misleading observations in any context may occur due to unknown changes introduced into an animal’s environment that may evoke a differential representation of behaviour in an individual. Such observer effects are familiar to herpetologists who have found that target animals were affected by stimuli ranging from observer-associated clothing (e.g. Putnam et al. 2017) to chemical or auditory stimuli or apparatus. These factors are often aversive but (rarely) might be attractive, presumably evoking curiosity (Rand et al. 1975). Frequently, only a subset of a population, such as social subordinates or a sex or age class, may be amenable to observation, and then perhaps only in specific parts of their environments such as an ecotonal boundary or within proximity to a refuge (see Zani et al. 2009). 

 

Insight into behavioural flexibility in the field can inform caregivers in captive situations. For example, different patterns of natural selection for risk-taking behaviour were observed in anoles (Anolis sagrei) in populations coping with the presence of natural predators (Lapiedra et al. 2018). With respect to differential responses to human intrusion, adult iguanas (Iguana sp.) may change their home ranges or sleeping areas in apparent response to perturbations such as observing at night with a spotlight or the activity of constructing a blind 80m away during the day; hatchlings may not disperse, but appear to increase the height of their sleeping perches where tall vegetation was available (Rodda et al. 1988). An artefact of observation of Anolis lizards in a naturalistic captive environment using video surveillance apparently evoked a mechanied version of the observer effect by reducing the amount of spontaneous activity recorded. When the dark round eye-shape of the lens was disguised by attaching an irregularly shaped camouflage painted piece of foam, activity levels were restored to levels seen in field observations (personal observation).

 

When it is necessary to sustain animals in captivity, we may minimise the distortions of circumstance by using field observations as a reference point: to the extent possible, we can quantify temporal and spatial aspects of key life-history variables such as activity patterns, feeding habits, social activities, or the proportions of various daily activities. Once documented in the field, one may then observe the spontaneous expression of the same variables in an empirically-derived simulation of the natural environment in the laboratory of zoo.

           

Observation and Context

 

Our knowledge of nature is grounded in observation. However, our senses are notoriously vulnerable to congenital as well as acquired biases and not everything we might wish to know is comparably amenable to observation, requiring us to resort to artificial (and thus more-or-less trustworthy) extensions of our senses. The ‘logic of the lamppost’ describes the disposition to look for answers where the illumination for one’s search is brightest. Suffering from either insufficient light or inadequate vision, many possible answers remain hidden in the deep shadows.  Unfortunately, the experiments of nature —spontaneously occurring combinations of variables that we suspect may illuminate a particular problem—are rare. In research we must, as J.S. Mill (1882) put it, resort to ‘artificial’ experiments of our own devising in order to add to nature’s experiments. Because the essence of an experiment is the selection and selective control of relevant variables, it is fair to say of science, as Langer (1957) famously said of art, that all useful representations are abstract. The premises of experimental observations—indeed all observations of other species with which we would productively interact–leads us to minimise sources of variation such as age, experience, and context and thus to emphasise reliability in (for example) animal models for research, from medical models to husbandry, productivity, and even companionship. Therefore, we must turn our attention briefly to models utilised for herpetological investigations

 

Animal Models 

 

Reptiles have been identified as useful animal models in research with implications for human well-being (e.g. Greenberg et al. 1989; Greenberg 1992; Lovern et al. 2004). Applications have been identified in development of behaviour (Burghardt 1978), brain research (Greenberg 2003; Nomura et al. 2013;), stress physiology (see Gangloff and Greenberg this volume), genetics (Tollis et al. 2014), and reproductive biology (Crews 1980; Lovern et al. 2004; Wade 2012;), among many other specific research domains. Sanger and Kircher (2017) identified the anole lizard (Anolis sp.) as a uniquely valuable model and reviewed the scholarly resources for integrative studies of ecology, evolution, development, and genetics.   

 

Animal models—like all models—are selective representations of phenomena of interest; they often obtain their status as a model because of circumstances of convenience. While this approach enables some of the most integrative and comprehensive overviews, it may also limit external validity. Acccordingly, although valuable reference points for the practitioner, they must be used with caution. Concurrently, animal models have led us into new areas such as emotion and cognition that may further diminish confidence in human exceptionalism (Siegforda et al. 2018; see also Font et al. this volume). Recent research, in Toadvine’s (2007) view  suggests that “the cognitive gap between humans and other animals is much narrower than has formerly been supposed, with the growing consensus that our differences are a matter of degree rather than kind”.

 

This factor, and growing prestige for the idea of continuity of traits, has instilled in many a deep sense of respect for the lives of research subjects (see Preston and de Waal 2002; de Waal 2019 for reviews). At the same time, uniqueness, indeed the exceptionalism of other species, is revealed by a deep familiarity with them and arguably, every species may be found to be exceptional. Although, we cannot know the inner life of any organism that manifests behaviour resembling our own, we can no longer assume that they do not have comparable experiences. In the spirit of Hobbes (1651/1982), when we observe animals in a context that would for us evoke joy or grief, we are inclined to “… thereby read and know what are their thoughts and passions … upon the like occasions.”

 

Therefore, once our eschewal of bias allows us to observe behaviour that resembles grief or joy or jealousy or love, we can employ our Janusian perspective and look for correspondences of causation with the most intensively studied organism—ourselves. We can ask about the critical stimuli evoking these affect-laden patterns. For example, in recent decades, the public imagination has been stimulated by so called, ‘odd couples’, (Nature 2012), extraordinary relationships between taxa that are rarely if ever in direct contact with each other. (e.g. Romm 2015).  Relationships other than those seen spontaneously in nature are often inadvertent experiments that can illuminate cognitive competencies that humans share with other species that structure social relationships. I turn now to several themes—’case studies’—in  reptile research that in aggregate, inform our understanding of important variables in  reptile management.

 

Case Studies

 

There are several domains of reptile research that can inform practitioners by virtue of the reciprocal illumination that is provided by the integration of findings analysed at different levels. Such case studies represent points of articulation between commonly isolated disciplinary domains (Bateson 1991b), and various species. These examples represent aggregates of good faith attempts to understand organisms from multiple perspectives by means of implicit application of the ethological perspective, including DEEP ethology, and by employing ethologically informed design when possible. Such overviews tell the best story they can with the best—most valid—information available and although patched together using information derived from various species, they provide a template against which new instances of application can be assessed; in other words, these case studies may be valuable in general, but often unreliable in their particulars.  

 

In the following few paragraphs I wish to briefly outline a few such areas in which theoretically fertile overviews are growing rapidly. The ethological attitude in action reveals, for example, dismissing the fallacy of reptiles for their ‘lack of complex social behaviour, emotions, cognition, and phenomena such as play, social learning, and deception’ (Doody et al. in press; Doody this volume; Gillingham and Clark this volume; Font et al. this volume). While evincing enthusiasm for integrative science, we should not infer disdain for analysis. Indeed, the relative intensity and focus provided by analysis offers solid grounding for integrative views that are brought to bear on understanding life at the level of organisation such as we, as organisms and communities, find ourselves. Such analyses also have the further virtue of illustrating how analysis and synthesis are mutually constitutive processes (see Beaney 2014).  

     

Thermoregulation

 

Thermoregulation is amongst the most studied of the physiological variables affecting reptile behaviour, and is valuable for us to consider because it occurs across all levels of organisation and in reptiles and is readily expressed in behaviour. Since the work of Cowles and Bogert (1944 and see Bogert 1959), few researchers would neglect the need for reptiles to regulate their body temperatures by behavioural selection of microhabitats of varying thermal qualities (see Heath 1965; Arena and Warwick this volume; Crowe-Riddell and Lillywhite this volume; Gillingham and Clark this volume; Lillywhite this volume). Because the means by which a ‘preferred body temperature’ is attained in a specific reptile or at a specific developmental time is very variable (e.g. ‘shivering’, or positioning to maximise absorption of solar radiation) and extensively studied in many contexts, the practitioner must extrapolate from well-studied examples that resemble the animal of interest at the moment before making assumptions about its welfare. Between its central importance to physiological functioning and its amenability to documentation, investigations of the various thermal strategies of reptiles provide many points of articulation between different levels of organisation and analysis. Ethologically integrative views of reptilian thermoregulation have focused on interlocking variables ranging from morphology (Pough 1980) to mental capacities (Avery 1976; Regal 1980) and have helped dispel the ‘endothermocentric’ prejudice that birds and mammals represet the evolutionary benchmarks for success (Greenberg 1976; Pough 1980). 

 

The richness of research into behavioural thermoregulation also provides us with examples of how compromised validity at one level can ramify through a cascade of subsequent research. For example, in adult male anoles (Anolis carolinensis), an appropriate body temperature is important to the responses of lizard adrenal to ACTH (Licht and Bradshaw 1969), and thus the entire ensemble of chronic stress effects on physiology and behaviour. The attainment of a specific body temperature is known to affect the production and action of androgens (Pearson et al. 1976) and spermatogenesis in Anolis carolinensis (Licht 1971), as well as the hormonal state and reproductive status of other species (Hutchison et al. 1966; Licht 1971; Garrick 1974; Schwarzkopf and Shine 1991; Daut and Andrews 1993). Other physiological variables that involve thermoregulation and are also likely to affect behaviour include diet, digestive state, and dehydration (Larson 1961; Regal 1966, 1980; Harwood 1979), and even response to infection and disease (see Regal 1980: Warwick 1991). Thermal influences on muscle physiology (Licht 1964a,b) might underlie the selection of specific defensive strategies at different temperatures (Rand 1964 on Anolis, and see Hertz et al. 1982 on Agama and for a brief review), although ‘escape burst speed’ and stamina documented in six genera of lizard by Bennett (1980) appear relatively independent of ambient temperature. 

 

Progressive environmental effects on thermoregulation are most familiar in the developmental phenomena of acclimation and acclimatisation. These processes, indicative of an organism’s attempt to maintain homeostasis by compensating for an environmental change, exemplifies the complexity of generalising laboratory research to observations in the field. When, in the laboratory, all conditions are held constant except for a single variable of interest (such as ambient temperature), the animal’s compensating adjustments are termed acclimation. In the more complex situation in the field (such as seasonal change), adaptive adjustments are termed acclimatisation (Prosser 1986). However, in either event, different species can be expected to compensate at different rates  (Art and Claussen 1982) and in different ways. Evolutionary background is brought to the fore when species such as the periodically winter-active anole (A. carolinensis) apparently uses partial acclimation to cold as an overwintering strategy while the sympatric race-runner lizard (Cnemidophorus sexlineatus), hibernates rather than acclimates (Ragland et al. 1981). In the laboratory, following acclimation to an elevated temperature, the anole (Anolis carolinensis) appears to be more heat-tolerant than those maintained at lower temperatures, but the temperature animals will spontaneously select (their ‘mean preferred temperature’) is unaffected (Licht 1968).

 

Animals studied after acclimation in the laboratory are often assumed to be comparable to those acclimatised to the more complex environmental stimuli in the field, but this assumption may be unwarranted. Gatten et al. (1988), for example, observed oxygen consumption and lactate concentration at rest and during induced exercise at 20 C in two populations of green anoles, (Anolis carolinensis): one acclimatised to various seasonal changes in the field and one acclimated to warm or cold conditions in the laboratory. In this study, resting oxygen consumption was unaffected by acclimation to warm or cold laboratory conditions, but did vary seasonally in the field animals. However, oxygen consumption during exercise showed significant differences between warm and cold acclimated lizards in the laboratory but was not affected by seasonal changes. Turtles are known to manifest an increase in their tolerance to abrupt temperature changes in response to increasing photoperiod and largely independent of seasonal temperature acclimation (Hutchison and Kosh 1965).

 

Stress

 
Welfare of an animal may seem necessarily compromised by stress, but this is often an unwarranted assumption. Without a test for circulating hormone levels (itself usually a stressor and often a poor indicator [Moore and Jessop 2003; Martinez-Silvestre 2014; Gangloff and Greenberg this volume]) the practitioner will have to rely on manifest behaviour—or (very fortunately) in some species, body colour (see about Anolis carolinensis, below). The stress response is an ensemble of coordinated autonomic and endocrine activities that in varying proportion are essential parts of other systems (see Axelrod and Reisine 1984; Johnson et al. 1992; Gangloff and Greenberg this volume), that serves both somatic and neurobehavioural (psychoactive) functions such as thermoregulation, aggression, and reproduction (Greenberg and Wingfield 1987), and can represent seasonal or developmental changes such as sexual maturation, social subordination, and reproductive competence (reviewed by Cooper and Greenberg 1992).  
 
The stress response is energetically demanding and if a limiting constituent such as one of the key enzymes or hormones is exhausted, could lead to death. Nevertheless, many species readily endure or even pursue highly stressful experiences when certain rewards (e.g. the higher need of successful reproduction) are at stake. For example, one of the key indications of stress is an increased level of circulating adrenal corticosteroid, but at least in some taxa, this hormone can manifest bimodal effects, stimulating feeding, exploration, or other activities at low levels while decreasing both feeding and activity when elevated (reviewed by Leshner 1978). This phenomenon is effectively described by the U-shaped ‘stress-response curve’ illustrating the adaptive response to increasing stress hormones up to a point, and then decreasing with additional stimulation (Chrousos and Gold 1992), what Zelena (2015) calls “The Janus Face of Stress on Reproduction”. This situation appears manifest in anoles (Anolis carolinensis) in which body colour associated with extreme stress (green with a dark eyespot) is seen briefly during male-male aggression, but is also evident in wounded or aged animals as they decline and approach their demise. In the field, zoo, or clinic such signs are often observed although cause and effect are unusually difficult to discern and the phenomenon is by far most studied in humans (for example, Borjigin et al. 2013).  

 

A fundamental preliminary to all other biological needs is physiological homeostasis: the maintenance of an internal state that is stable within the range of variation an animal can experience before energetically demanding compensating responses are triggered. Within this range, sometimes termed ‘adaptive scope,’ animals can compensate with little or no energetic expenditure; beyond this range, a comprehensive physiological stress response is engaged that involves an ensemble of coordinated physiological and behavioural responses (Crews 1980; Crews and Garrick 1980; Greenberg 1990; Guillette et al. 1995). Clearly, failure to provide for one stress-sensitive variable can affect others because the stress response ramifies throughout the organism. In the absence of explicit information about the capacity of an organism to cope with a specific variable, we are obliged to limit potential challenges to a range within which it is known to be able to cope. Examples would be the altered yet stable endocrine profile of subordinate anoles (Anolis carolinensis), (Greenberg and Lumsden 1990), or lizards subjected to frequent handling (Meier et al. 1973).  

 

In practice, both short- and long-term changes in the physical appearance of many species are prominent phenomena that inform the practitioner regarding welfare. Much of appearance, such as weight and muscle tone can be an expression of sustained stress possibly indicating an underlying pathology. However, rapid, relatively brief, or seasonal episodes are also important in considering animal management and treatment, and a steward might be easily misled if natural stress dynamics were taken as indicative of disorder.

 

Body colouration of the green anole (Anolis carolinensis), is, for the wellness-concerned steward, an outward expression of usually hidden physiological phenomena. In this anole, only circulating hormones affect chromatophores: green is the basic colour of an unperturbed animal, whereas light brown indicates successful coping with modest disturbances, green complemented with a dark eye spot can represent a surge of stress hormones, and a blotchy green and brown colour is manifest in animals in extreme distress.  Only a few animals are known to so readily provide a visualisation of their inner state, but observing the phenomenon in real time provides a penetrating sense of the labile dynamics of acute and chronic stress, including its expression in social interactions, as it likely occurs in many other species. 

 

Social Dominance 

 

Whether in nature or captivity reptilian sociality may be relevant to welfare and dependent on ecological or artificial resources (see Brattstrom 1974; Doody this volume; Gillingham this volume). Many reptiles are seen in aggregations when resources are sufficient such that competition for them would waste energy, but in many species, competitive and social dominance relationships can change as resources change. Males of the black iguana (Ctenosaura similis) will compete for exclusive territories unless a sudden abundance of prey occurs at which times they appear mutually tolerant enough to aggregate (Evans 1951). Blue spiny lizards (Sceloporus cyanogenys) will compete for first daylight for warmth and for hibernacula in the evenings (Greenberg 1978). The behaviour of anole (Anolis carolinensis) males maintained in vivaria can be affected by each other’s presence, particularly after a loser of a competitive interaction is unable to escape the presence of the winner, something unlikely to occur in nature (Jenssen et al. 1995). 

 

In some species a continuum of degrees of territoriality can be seen in captivity and in the field (Hunsaker and Burrage 1969). At least a few species of lizard also shift their social organisation from territorial to social dominance hierarchy in response to environmental characteristics such as available space (Evans 1951 for black iguanas [Ctenosaura pectinate]; Norris 1953 for desert iguanas [Dipsosaurus dorsalis]). Field data on social dominance in the well-studied anole (Anolis carolinensis) (Jenssen et al. 1995) indicates seasonal dependence, but in the laboratory simulating breeding season conditions, dominance relationships are established that are rapidity suggestive of a conservative behavioural pattern (Greenberg et al. 1984). Social dominance, in the conventional sense of one animal manifesting a priority of access to a limited resource over another, is seen in winners of aggressive interactions between males cohabiting a single enclosure. In this species, winners continue to perch at the highest site available and court females, expressing little more than occasional interest in the cohabiting loser. However, the loser changes markedly: he becomes darker in body colour, selects lower perch sites, is less active, and does not court —he has become a social subordinate. Such pairs often share food and water and maintain stable relationships for extended periods, suggesting a pattern well fixed in their behavioural repertoire. Laboratory observations, no matter how consistent, can do no more than suggest ecological hypotheses about the possible advantages accruing from the changes in subordinates: depending on the species, adopting protective microhabitats, lower posture and activity levels, colours that could be cryptic or resemble non-provocative individuals such as juveniles or females.

 

The significance of understanding social relationships in husbandry settings is that the interacting dynamics of stress and reproductive physiology are both cause and consequence. Endocrine dynamics of stress interact with those of reproduction (courtship, mating, parental behaviour; see, for example, Greenberg and Wingfield 1987; Tokarz and Summers 2011). Natural elevation of testosterone, for example, can lead to aggressive interactions after which circulating levels are briefly elevated and then sustained in winners but reduced in losers of anoles (Anolis carolinensis) (Greenberg and Crews 1990). In at least captive settings, the behavioural consequences of reduced androgen in subordinates may be protective in that they will avoid the potentially harmful responses of their unavoidable dominant companion (Greenberg et al. 1995). Association of the stress response with social dominance (as indicated by the model anole lizard [Anolis carolinensis], Greenberg 2002; Summers 2002) is now a familiar idea. However, there is a diversity of relationships between the central nervous system, autonomic, and endocrine causes and consequences of social dominance or subordination (Greenberg 1983) that prevent easy generalisation.

 

Can we say anything as yet about the physiological substrate of altered social behaviour? While the ecological aspects of the phenomenon in the laboratory are as yet elusive, the control and consistency of the laboratory facilitate investigations of the neurobiology and behavioural endocrinology. Development, ecology and physiology clearly converge in investigations of forebrain sites apparently responsible for integration of stimuli leading to the expression of specific social displays (reviewed in Greenberg 1983, 1990), and brainstem nuclei controlling a key effector of the dewlap displays (Font et al. 1986; Font 1991). 

 

Body colour, mentioned above in discussing stress, is in many species a nexus of interest, manifesting both short-term and long-term physiological phenomena associated with autonomic reflexes and stress at levels of organisation from homeostasis to communication (Cooper and Greenberg 1992).  For example, the darker body colour of anoles (Anolis carolinensis) when losing a fight can provide a key to endocrine variables associated with social subordination and submissiveness. This phenomenon, well known in the laboratory (Greenberg and Noble 1944) is also seen in the field (Medvin 1990), and presents provocative possibilities because the hormones that affect the chromatophores are also associated with the physiological stress response and, at least in other taxa, have appear to act to suppress aggressiveness and facilitate the expression of social submissiveness (Leshner 1978). In the limited space in which captive reptiles are typically maintained where winners and losers of staged dominance contests remain in each other view, changes indicate increased melanotropin (animals remain dark) and reduced testosterone (animals will not court females), but when the dominant is removed, the formerly suppressed individual recovers green colour and motivation to court over a period of up to three days (Greenberg unpublished observations); interestingly, these defeated males (with presumably low testosterone) are at first aggressive to females introduced to test courtship, but over three days (with presumably higher testosterone) court in normal fashion.

 

Reproduction

 

Beyond mere individual survival, successful reproduction is the highest expression of the biological potential of individual animals—at least in evolutionary terms of direct and even indirect fitness.  While reproduction may be of great concern to stewards of animal welfare in zoos and conservation parks, many species can tolerate otherwise dangerous levels of stress to reproduce and to protect their progeny (see for example, MacLeod et al. 2019; Gangloff and Greenberg this volume). Because of the conservatism and complexity of reproductive processes, reproductive success is often regarded as an indication of well-being, and the environment in which this occurs may be regarded as at least adequate. However, physiological stress is a relatively extreme expression of an animal’s unmet needs. In reproductive behaviour, environmental influences can be subtle and counterintuitive (for example, see Whittier and Crews 1987).  Further, specific kinds of stressors can evoke physiological and behavioural change important in synchronising or activating important elements of reproduction.

 

Failure of captive or managed animals to produce successful progeny can, therefore, be regarded as a failure of adequate ethological insight. For example, while adult Anolis spp. can be successfully maintained and will reproduce in the laboratory (Greenberg and Hake 1990), rearing of progeny is notoriously difficult. Methods of facilitating reproduction in captive reptiles have thus rightly emphasised the convergence of external climatic, physical, and social stimuli on the internal processes leading to courtship, copulation, pregnancy, and parental behavioural patterns such as nest-site preparation, oviposition or parturition, and in some cases even brooding (see Carpenter 1980; Crews and Garrick 1980).  

 

Coordinated field and laboratory studies are effective ways of isolating the often subtly interdependent aspects of the complex reproductive process in ways important to practice. In particular, reptiles that are subject to arbitrary maintenance regimens and for whom different experiences, such as predator or prey or even illumination levels can frustrate efforts to provide for their welfare and ensure prospects for successful reproduction. For example, while some opportunities for thermoregulatory basking would certainly be provided by most ethologically-informed reptile keepers, only recently have clues been forthcoming about the means by which the physiology of brooding females of various viviparous species affects thermal preferences in order to accelerate embryogenesis and decrease gestation time. The altered basking patterns of gravid females of several viviparous species was reviewed by Shine (1980), who noted that while there are several reproductive advantages, there are also distinct costs. In the laboratory and in the field, gravid female viviparous skinks (Eulamprus tympanum), will attain body temperatures comparable to those of males and non-gravid females, but will increase the duration of daily basking time. Also, in the laboratory it was learned that the increased basking time results in accelerated gestation (Schwarzkopf and Shine 1991).

 

Mitigation of Distress

 

While we must anticipate that captivity distorts behaviour, the nature of that distortion may not be obvious if there are no ethological reference studies. The capacity of many species to compensate for inadequate resources to meet biological needs can obscure diminished competences until they are manifest in a conspicuous way such as failure to thrive or reproduce or expression of other stress-related signs, common in zoos and other captive breeding contexts. The laboratory context in which efforts are taken to minimise potentially confounding independent variables is particularly vulnerable to unknown distortions. However, beyond the practical compromises that may distort findings, research designs often utilise procedures that knowingly cause ‘unavoidable’ distress. The most extreme procedures involve surgery and sacrifice. Practitioner stewards must, like the reptile itself in nature, seek an optimal compromise. Such a circumstance is considered below.

 

Two reviews of the use of hypothermia and freezing as means of anesthesia and euthanasia for amphibians and reptiles (Shine et al. 2015; Lillywhite et al. 2017) underscore that our intuition might fail us when we judge other species by our own experience. For example, Langkilde and Shine (2006) present data indicating that hypothermia could be less stressful than a commonly recommended alternative. These authors also observed in their review that ‘stress’ and ‘distress’ are commonly conflated and that many authors use emotionally biased language in their narratives.

 

A perennial concern in both clinic and laboratory is cooling reptiles for experimental procedures.  This is identified and briefly reviewed by Gangloff and Greenberg (this volume) and Arena et al. (this volume) where, work is cited that supports the idea that small ectotherms probably do not experience pain when subjected to cooling anaesthesia or euthanasia (Shine et al. 2015; Keifer and Zheng 2017). Until such time as sufficient data is collected to develop such a design, Warwick et al. (2018) recommend we hold off, and primary scientific advisory bodies continue to recommend against induced hypothermia in most situations (AVMA 2013; OIE 2019). Surely animal size, life history, and experience are significant and given the great physical and physiological diversity of Reptilia, ethologically informed design is crucial to avoid or mitigate the stress of unavoidable procedures. In fact, evidence of having such a design should be clearly evident in any proposed appeal for research support.

 

Animal Welfare Conclusions

 

Concerns of researchers and practitioners for the welfare of their animal subjects have grown greatly in recent years. Sometimes we do this as a matter of compassion and responsible stewardship, other times out of a concern for validity  of  research findings. Integrative thinking as exemplified in DEEP ethology will help minimise the neglect of variables outside one’s traditional disciplinary experience.

 

Welfare is usually perceived as the circumstances that enable specific animals to meet their biological (including psychological) needs in specific contexts This includes sanctuaries, farms, zoos, households, or research laboratories. These efforts should be pursued even in light of bias attributable to the narrower needs of their human caretaker. What constitutes ‘welfare’ can be quite variable depending on context. Nevertheless, all animals have comparable basic needs such as those characterised by Maslow’s (1943) in his famous hierarchy of motivational needs. We can translate this original scheme into one emphasising biology: physiology (homeostasis, health), safety, sociality, individuality, and self-actualisation (reproduction, maximising direct or indirect biological fitness) (see Table 3).

 

Ethologically informed design considers natural history from multiple persectives: including DEEP ethology (Table 2), in concert with a hierarchical view of the animals needs (Table 3).  Taken together, such consideration of description and function will enable care-givers to provide effective husbandry. But, in the culture of animal welfare-concerned traditions, several other design models have emerged. These include, for example, the ‘five domains’ model (nutrition, environment, health, behaviour, affective experience; see Mellor and Beausoleil 2015). However, methods of assessment for affective experience are likely to be variable in different reptile taxa and are in need of closer, species-specific, study. This need was emphasised by Benn et al. (2019) who looked at and evaluated the utility of both resource-based and animal-based factors and applied a proposed inventory of criteria to a specific endangered skink. Consistent with the comparative perspective of ethologically-informed design, Benn et al. observed that environmental enrichment for captive reptiles might promote ‘positive welfare states’ in different ways for different species.

Before we undertake any intrusion into the life of another species we must be thoughtful about our motivation. Where we have justified such an intrusion, the ethological attitude (Table 1) will hopefully hopefully improve welfare. To this end, I suggest the use of flexible checklists in which the elements of DEEP ethology—the variables relevant to development, ecology, evolution, and physiology (Table 2). These are easily documented and organised in a format convenient to interspecies comparison in as much detail as possible for a given taxon. Continuing developments in data management and  access have the potential to make such an enterprise available. This approach resembles a kind of behaviour taxonomy—individualised for species—extended into the DEEP domains. For example, such lists could include, in the spirit of Warwick et al. (1992) species-typical behavioural patterns (see also Carpenter 1977; Greene 1988), and detailed inventories of social variables (McBride 1976). Animal care and welfare concerns could then be better grounded in what is known about a particular species or the next most closely related species, or even species known to occupy comparable ecological niches. I think we can be hopeful that current rapidly emerging computer resources can facilitate development of such a resource.

 

Given the accelerating pace of environmental degradation, we cannot move quickly enough to secure the welfare of animals subject to the impact of human activities. We must also be unrelenting in our efforts to facilitate the activities of researchers and ensure the utility of their labours. These closely related, often mutually dependent concerns can be united and significantly enhanced by efforts to approximate the ethological attitude (Table 1), and the ideal of an ethologically informed design (EID, Table 3). Taken together, these efforts may significantly enhance our aspirations for responsible research and conscientious animal care and husbandry.

 

Acknowledgements

 

I am deeply grateful for the sometimes spirited but always collegial and enriching comments and suggestions of Clifford Warwick, Gordon Burghardt, Phil Arena, Katherine H. Greenberg, and an unidentified reviewer, all of whom provided invaluable feedback.

 

References

 

Art GR, Claussen DL (1982)  The rate of thermal acclimation in the lizard, Anolis carolinensisCopeia 1982(1):189-191

 

Avery RA (1976)  Thermoregulation, metabolism and social behavior in Lacertidae.  In Morphology and Biology of Reptiles.  Edited by A d ’A Bellairs and C.B. Cox.  Academic Press, N.Y.  pp245-259

AVMA (2013) Guidelines for the euthanasia of animals. Available at: www.avma.org/KB/Policies/Documents/ euthanasia.pdf.

Axelrod J, Reisine TD (1984)  Stress hormones: their interaction and regulation.  Science 224:452-459.

 

Barnett JL, Hemsworth PH (1990) The validity of physiological and behavioural measures of animal welfare.  Applied Animal Behaviour Science 25:177-187.

 

Bateson P (1991a)  Assessment of pain in animals.  Animal Behaviour  42:827-839.

 

Bateson P (1991b)  Levels and processes.  In The Development and Integration of Behaviour.  P. Bateson, editor.  Cambridge University Press, N.Y.  pp. 3-16.

 

Benn, AL, McLelland DJ, Whittaker AL (2019) A review of welfare assessment methods in reptiles , and preliminary application of the welfare quality protocol to the pygmy blue-tongued skink, Tiliqua adelaidensis, using animal-based measures. Animals 9(27):  doi :10.3390/ani9010027 https://www.mdpi.com/2076-2615/9/1/27

 

Bennett AF (1980)  The thermal dependence of lizard behaviour.  Animal Behaviour 28:752-762.

Bestion E, Jacob S, Zinger L, Di Gesu L, White J, Cote J (2017) “Climate warming reduces gut microbiota diversity in a vertebrate ectotherm,” Nature Ecology and Evolution, doi:10.1038/s41559-017-0161, 2017. https://www.nature.com/articles/s41559-017-0161

 

Borjigin J, Lee U, Liu T, Pal DHuff SKlarr DSloboda JHernandez JWang MM, Mashour GA (2013) Surge of neurophysiological coherence and connectivity in the dying brain.  https://doi.org/10.1073/pnas.1308285110

Bogert CM (1959)  How reptiles regulate their body temperature.  Scientific American   105-120.

 

Brattstrom BH (1974) The Evolution of Reptilian Social Behavior  Amer. Zool., 14:35-49 .

 

Brattstrom BH (1978) Learning studies in lizards. In: , Behavior and Neurology of Lizards.  N. Greenberg and P.D. MacLean, eds.  Bethesda, MD:  NIMH,  pp. 173-181.

 

Burghardt GM (1977)  Learning processes in reptiles.   In:  Biology of the Reptilia, Volume 7, Ecology and Behaviour A,  C. Gans and D.W. Tinkle, editors.  Academic Press, N.Y.  pp. 555-681.

 

Burghardt GM (1978) Behavioral ontogeny in reptiles: whence, whither, and why. In The development of behavior: Comparative and evolutionary aspects. (G. M. Burghardt & M. Bekoff, eds.). Garland STPM press, New York, 1978, 149-174.

 

Burghardt, G.M. (1997) Amending Tinbergen: A fifth aim for ethology. In R.W. Mitchell; N. S. Thompson; H. L. Miles (eds.). Anthropomorphism, anecdotes and animals. Albany: State University of New York Press. pp. 254–276.

 

Burghardt GM (2013) The Janus-Faced Nature of Comparative Psychology – Strength or Weakness?  Evolutionary Psychology 11(3): 762-780. www.epjournal.net.

 

Burghardt GM (2020) Insights found in century-old writings on animal behaviour and some

cautions for today. Animal Behaviour 164(2020): 241-249

 

Burghardt GM, Layne DG (1995) Effects of early experience and rearing condition.  In: Warwick C, Frye FL, Murphy JB (eds) Health and Welfare of captive reptiles pp 165-185.  London, Chapman & Hall.

 

Carpenter CC (1977)  Variation and Evolution of stereotyped behavior in reptiles, Part I. A survey of stereotyped reptilian behavioral patterns.  In:  Biology of the Reptilia, Volume 7, Ecology and Behaviour A,  C. Gans and D.W. Tinkle, editors.  Academic Press, N.Y.  pp. 335-403, 459-507.

 

Carpenter CC (1980)  An ethological approach to reproductive success in reptiles.  In: Reproductive Biology and Diseases of Captive Reptiles, SSAR Contributions to Herpetology Number 1, J.B. Murphy and J.T. Collins, editors. Society for the Study of Amphibians and Reptiles.  Pp. 33-48.

 

Christian JJ (1980)  Endocrine factors in population regulation.  In Biosocial mechanisms of population regulation  M.N. Cohen, R.S. Malpass, and H.G. Klein, editors.  Yale, New Haven.  Pp. 55-115.

 

Chrousos GP & Gold PW (1992) The concepts of stress and stress system disorders: overview of physical and behavioral homeostasis. JAMA 267, 1244–1252 (1992).

 

Cooper WE Jr, Greenberg N (1992) Reptilian coloration and behavior.  In: Hormones, Brain, and Behavior, Vol. 18 of Biology of the Reptilia, C Gans and D. Crews, editors.  University of Chicago Press, pp 298-422.

 

Cowles RB, Bogert CM (1944) A preliminary study of the thermal requirements of desert reptiles. Bulletin of the American Museum of Natural History.  83(5):261-296.

 

Crews D (1980) Interrelationships among ecological, behavioral, and neuroendocrine processes in the reproductive cycle of Anolis carolinensis and other reptiles. Advances in Studies of Behavior. 11 (1980):1-74.

 

Crews D, Garrick LD (1980)  Methods of inducing reproduction in captive reptiles.  In: Reproductive Biology and Diseases of Captive Reptiles, SSAR Contributions to Herpetology Number 1, J.B. Murphy and J.T. Collins, editors.  Society for the Study of Amphibians and Reptiles.  Pp. 49-70.

 

Daut, EF Andrews RM (1993)  The effect of pregnancy on thermoregulatory behavior of the viviparous lizard Chalcides ocellatusJournal of Herpetology.  27(1):6-13.

 

DeLong AJ, Greenberg N, Keaney C (1986)  Temporal responses to environmental scale in the lizard, Anolis carolinensis (Reptilia, Lacertilia, Iguanidae).  Behavioral Systems. 13:339-352.  

De Waal FBM  (2011) What is an animal emotion   Ann N Y Acad Sci. 2011 Apr;1224:191-206.  Doi:10.1111/j.1749-6632.2010.05912.x. https://www.ncbi.nlm.nih.gov/pubmed/21486301

 

de Waal F,  Preston S. (2017)  Mammalian empathy: behavioural manifestations and neural basis. Nat Rev Neurosci 18, 498–509 doi:10.1038/nrn.2017.72

 

De Waal, FBM (2019) Mama’s Last Hug: Animal Emotions and What They Tell Us About Ourselves. W.W. Norton & Company.

 

Doody JS, Dinets V, Burghardt GM (in press) Secret Societies: The Social Life of Reptiles.  Johns Hopkins University Press.

 

Edson PH, Gallup GG Jr (1972) Tonic immobility as a fear response in lizards (Anolis carolinensis).  Psychon. Sci. 26:27-28.

 

Eisner R (1991) Scientists roam the habitat as zoos alter their mission.  The Scientist. 5(11): 1,10,21.

 

Eibl-Eibesfeldt, I (1970)   Ethology, the Biology of Behavior, second edition.  Holt, Rinehart and Winston, N.Y.  625 pp.

 

Evans LT (1951) Field study of the social behavior of the black lizard  Ctenosaura pectinata.  Amer. Mus.  Novitates 1493:1-26.

 

Font E, Greenberg N, Switzer RC III. (1986)  Brainstem origins of motoneurons controlling the hyoid contribution to the dewlap display of Anolis lizards.  Proc Soc Neuroscience 12(1):498 (abstract).

 

Font E (1991) Localization of brainstem motoneurons involved in dewlap extension in the lizard, Anolis equestrisBehav. Brain Res. 45:171-176.

 

Font E, Burghardt, E, Leal, M (in press 2019). Reptile Brains, Behavior, and Cognition: Multiple Misconceptions. In: Warwick C, Arena PC, Burghardt GM, Frye FL, Murphy JB (Eds) Health and welfare of captive reptiles, 2nd  ed. Cham, Switzerland: Springer

 

Freeman WJ (1995) Societies of brains: A study in the neuroscience of love and hate.  Freeman, W. J. (1995).  Hillsdale, NJ, Erlbaum Associates, Inc.  https://psycnet.apa.org/record/1995-97653-000

 

Gallup GG Jr (1973) Simulated predation and tonic immobility in Anolis carolinensisCopeia. 1973(3):623-624.

 

Gallup GG Jr (1974)  Animal hypnosis: Factual status of a fictional concept.  Psychology Bulletin. 81:836-853.

 

Garrick LD (1974) Reproductive influences on behavioral thermoregulation in the lizard, Sceloporus cyanogenys. Physiol. & Behav. 12:85-91.

 

Gatten RE Jr, Echternacht AC, Wilson MA (1988)  Acclimatization versus acclimation of activity metabolism in lizard.  Physiol. Zool 61(4):322-329.

 

Gibson JJ (1979) The Ecological Approach to Visual Perception, Boston, Houghton, Mifflin & Co. ISBN 0-89859-959-8.

 

Goebel R (2014) Understanding the Human Brain at Multiple Levels of Organization: Possibilities and Limitations of Ultra-High Field fMRI UCLA Advanced Neuroimaging Summer Program Los Angeles, USA, July 20, 2014. http://www.brainmapping.org/NITP/images/Summer2014Slides/20140721_UnderstandingHumanBrain_Goebel.pdf

 

Greenberg B, Noble GK (1944)  Social behavior of the American chameleon (Anolis carolinensis Voight).  Physiological Zoology  17(4):392-439.

 

Greenberg KH,  Sohn BK, Greenberg NB, Pollio HR, Thomas SP, Smith JT  (2019) The Phenomenological Heart of Teaching and Learning. Routledge. Particularly chapter 2.

 

Greenberg N (1976)  Thermoregulatory aspects of behavior in the blue spiny lizard, Sceloporus cyanogenys (Sauria, Iguanidae).  Behaviour  59:1-21.

 

Greenberg N (1977)  An ethogram of the blue spiny lizard, Sceloporus cyanogenys (Reptilia, Lacertilia, Iguanidae).  Journal of Herpetology 11(2):177‑ 195.

 

Greenberg N (1978) Ethological considerations in the experimental study of lizard behavior.  In: Behavior and Neurology of Lizards.  N. Greenberg and P.D. MacLean, eds.  Bethesda, MD:  NIMH,  pp. 203-226.

 

Greenberg N (1983) Central and autonomic aspects of aggression and dominance in reptiles. In: Ewert, J.-P., R.R. Capranica, and D.J. Ingle, editors, Advances in Vertebrate Neuroethology.  Plenum Press, N.Y.   pp. 1135-1143.

 

Greenberg N (1985)  Exploratory behavior and stress in the lizard, Anolis carolinensisZeitschrift fur Tierpsychologie 70:89-102.

 

Greenberg N (1990) Behavioral endocrinology of physiological stress in a lizard.  Journal of experimental Zoology.  Supplement. 4:170-173.

 

Greenberg N (1992)  The saurian psyche revisited: Lizards in research. In: The Care and Use of Amphibians, Reptiles, and Fish in Research, D.O. Schaeffer, K.M. Kleinow, L. Krulish, editors. Scientists Center for Animal Welfare, Bethesda, MD. Pp 75-91.

 

Greenberg N (1993) Tongue-flicking and exploratory behavior in Anolis carolinensis. Brain, Behavior, and Evolution  41:210-218.

 

Greenberg N (1995) Ethologically informed design in husbandry and research. In: Warwick C, Frye FL, Murphy JB (eds) Health and welfare of captive reptiles, 1st ed. Chapman & Hall, London, p 239-262.

 

Greenberg N (2002) Ethological aspects of stress in a model lizard, Anolis carolinensis. Integ Comp Biol 42 (3):526-540. doi:10.1093/icb/42.3.526

 

Greenberg N, Burghardt GM, Crews D, Font E, Jones RE, Vaughan G (1989) Reptile models for biomedical research. In: Woodhead AD, ed. Nonmammalian Animal Models for Biomedical Research. Boca Raton: CRC Press, Inc. p 289-308.

 

Greenberg N, Crews D (1983)  Physiological ethology  of aggression   in  amphibians and  reptiles.   In  B.  Svare  (ed.),  Hormones  and  aggressive behavior, pp.  469-506.   Plenum   Press, New  York.

 

Greenberg N, Crews D (1990)  Endocrine and behavioral responses to aggression and social dominance in the green anole lizard, Anolis carolinensis.  General and Comparative Endocrinology, 77:1-10.

 

Greenberg N, Crews D, Summers C,  Harris J (1995).  Adaptive responses to social subordination.  Proc. XXIV Internat. Ethology Conf. Honolulu, Hawaii, Aug 10-17.

 

Greenberg N, Hake L (1990) Hatching and neonatal behavior of the lizard, Anolis carolinensisJournal of Herpetology, 24(4):402-405.

 

Greenberg N, Lumsden S  (1990). Social dominance and reproductive activity in the lizard, Anolis carolinensis.  American Zoologist, 29(4):105A. (Abstract)

 

Greenberg N, Wingfield J (1987)  Stress and reproduction: Reciprocal relationships.  In: Reproductive Endocrinology of Fish, Amphibians, and Reptiles. D.O. Norris and R.E. Jones, eds.,  New York, NY:  Springer. Pp. 461-503.  DOI https://doi.org/10.1007/978-1-4613-1869-9_16  ISBN 978-1-4612-9042-1

 

Greenberg N, Chen T, Crews D (1984)  Social Status, gonadal state, and the adrenal stress response in the lizard, Anolis carolinensis. Hormones and Behavior. 18:1-11.

 

Greene HW (1988)  Antipredator mechanisms in reptiles.    In: Biology of the Reptilia, Volume 16.  C Gans and R.B. Huey, editors. John Wiley & Sons, New York pp1-152.

Guillette Jr., L.J., Cree, A. and Rooney, A.A. (1995) Biology of stress: interactions with reproduction,immunology and intermediary metabolism. In: Warwick C, Frye FL, Murphy JB (eds) Health and Welfare of captive reptiles, London, Chapman and  Hall, pp. 32-70

Harwood RH (1979) The effect of temperature on the digestive efficiency of three species of lizards, Cnemidophorus tigris, Gerrhonotus multicarinatus, and Sceloporus occidentalis, Comparative Biochemistry and Physiology. 63A:417-433

 

Heath JE (1965) Temperature regulation and diurnal activity in horned lizards.  University of California Publications in Zoology.  64(3):97-136. 

 

Heatwole H (1968)  Relationship of escape behavior and camouflage in anoline lizards. Copeia 1968(1):109-113.

 

Hediger H (1950) Wild Animals in Captivity.   Buttersworth, London.  Republished by Dover Publ., N.Y., 1968.

 

Hediger H (1955) The psychology and Behavior of Animals in Zoos and Circuses. Buttersworth, London.  Republished by Dover Publ., N.Y., 1968.

 

Hennessy JW, Levine S (1979).  Stress, arousal, and the pituitary-adrenal system: a psychoendocrine hypothesis.  Progress in Physiological Psychology 8:133-178.

 

Hennig, C.W.  (1979) The effects of physical environment, time in captivity, and defensive distance on tonic immobility, freezing, and flight behaviors in Anolis carolinensis. Animal Learning and Behavior. 7(1):106-110.

 

Hertz PE, Huey RB, Nevo E (1982) Fight versus flight: body temperature influences defensive responses of lizards.  Animal Behaviour 30:676-679.

 

Hinde RA. (1970) Animal Behaviour: A Synthesis of Ethology and Comparative Psychology, second edition. New York, McGraw Hill.

 

Hobbes, T. (1651/1982) Leviathan.  (Bobbs-Merrill, Indianapolis.  (The Introduction, p.24)

 

Hunsaker D, Burrage BR (1969) The significance of interspecific social dominance in  iguanid lizards.  American Midland Naturalist 81:500-511.

 

Hut P (2001). The role of Husserl’s epoché for science: A view from a physicist. Paper presented at the meeting of the Husserl Circle, Bloomington, IN. Retrieved from  http://archive.is/oxDc

 

Hutchison VH, Kosh RJ (1965) The effect of photoperiod on the critical thermal maxima of painted turtles (Chrysemys picta).  Herpetologica  20(4): 233-238.

 

Hutchison VH, Dowling HG, Vinegar A (1966)  Thermoregulation in a brooding female Indian python, Python molurus bivittatusScience 151: 694-696.

 

James, W (1884)  What is Emotion  Mind 9:188–205

 

Jenssen TA, Greenberg N, Hovde KA (1995)  Behavioral profile of free-ranging male Anolis carolinensis across breeding and post-breeding seasons.  Herpetological Monograph. Vol. 9 (1995):41-62.

 

Johnson M (2007) The Meaning of the Body: Aesthetics of Human Understanding.  University of Chicago Press.

 

Johnson EO, Kamilaris TC, Chrousos G, Gold PW (1992) Mechanisms of stress: A dynamic overview of hormonal and behavioral homeostasis.  Neuroscience and Biobehavioral Reviews.  16:115-130.

 

Kassahn KSCrozier RHPörtner HOCaley MJ. (2009) Animal performance and stress: responses and tolerance limits at different levels of biological organisation.  Biol Rev Camb Philos Soc. 84(2):277-92. doi: 10.1111/j.1469-185X.2008.00073.x. Epub 2009 Mar 11.

 

Keifer J and Zheng Z (2017) Cold block of in vitro eyeblink reflexes: evidence supporting the use of hypothermia as an anesthetic in pond turtles. J Exp Biol 220 (Pt 23):4370-4373. Doi:10.1242/jeb.168427

 

Koestler A (1978)  Janus: A Summing Up.   NY: Random House, 1978 ISBN 0-394-50052-0

 

Kuhn T (1959) The Essential Tension: Tradition and Innovation in Scientific Research, in The Third (1959) University of Utah Research Conference on the Identification of Scientific Talent C. Taylor, Salt Lake City: University of Utah Press: 162–74.

 

Lakoff G, Johnson M (1999) Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought.  N.Y. Basic Books

 

Langer SK (1957) Problems of Art. Scribner’s, N.Y.

 

Langkilde T, Shine R (2006) How much stress do researchers inflict on their study animals? A case study using a scincid lizard, Eulamprus heatwolei.  J Experimental Biology 209, 1035-1043.  Doi:10.1242/jeb.02112

 

Lapiedra O, Schoener TW, Leal M Losos JBKolbe JJ. (2018) Predator-driven natural selection on risk-taking behavior in anole lizards.  Science  360(6392):1017-1020.  https://science.sciencemag.org/content/sci/360/6392/1017.full.pdf

 

Larson MW (1961)  The critical thermal maximum of the lizard Sceloporus occidentalis occidentalis Baird and Girard.  Herpetologica 17(2):113-122.

 

Leshner A (1978) An introduction to Behavioral Endocrinology.  Oxford, N.Y.  361 pp.

 

Licht P (1964a)  The temperature dependence of myosin-adenosinetriphosphatase and alkaline phosphatase in lizards.  Comparative Biochemistry and Physiology.  12:331-340.

 

Licht P (1964b)  A comparative study of the thermal dependence of contractility in saurian skeletal muscle.  Comp. Biochem. Physiol.  13:27-34.

 

Licht P (1967)  Environmental control of annual testicular cycles in the  lizard Anolis carolinensis.  I.  Interaction of light and temperature in the  initiation of testicular recrudescence.  Journal Exp. Zool.., 165:  505-616.

 

Licht P (1968).  Response of the thermal preferendum and heat resistance to thermal acclimation under different photoperiods in the lizard Anolis carolinensis.  American Midland Naturalist 79(1):149-158.

 

Licht P (1971) Regulation of the annual testis cycle by photoperiod and temperature in the lizard Anolis carolinensisEcology 52:240-252. 

 

Licht P, Bradshaw SD (1969)  A demonstration of corticotropic activity and its distribution in the pars distalis of the reptile.  General and Comparative Endocrinology., 13:226-235.

 

Lovern MB, Holmes MM, Wade J (2004) The Green Anole (Anolis carolinensis): A Reptilian Model for Laboratory Studies of Reproductive Morphology and Behavior ILAR Journal Volume 45, Number 1:54-64

 

Luther D, Baptista L (2010) Urban noise and the cultural evolution of bird songs Proc. R. Soc. B (2010) 277: 469–473   doi:10.1098/rspb.2009.1571

 

Lillywhite HB, Shine R, Jacobson DF, Denardo MS, Gordon CA, Navas TW, Seymour RS, Storey KB, Heatwole H, Heard D, Brattstrom B, Burghardt GM (2017)  Anesthesia And Euthanasia Of Amphibians And Reptiles Used In Scientific Research: Should Hypothermia And Freezing Be Prohibited?  Bioscience  Jan 2017  67(1): 51-61

 

MacLeod KJ, McCormick GL. & Langkilde T. (2019) Glucocorticoids do not influence a secondary sexual trait or its behavioral expression in eastern fence lizards. Sci Rep 9, 5229  DOI:10.1038/s41598-019-41596-1

 

Mahoney CJ (1992) Opinion: Some thoughts of psychological enrichment.  Lab Animal May:27-37.

 

Marcellini DL Jenssen TA (1991) Avoidance learning by the curly-tailed lizard, Leiocephalus schreibersi: Implications for anti-predator behavior.  Journal Herpetol. 25(2):238-241.

 

Martínez-Silvestre, A. (2014) How to assess stress in reptiles. Journal of exotic Pet Medicine, 23: 240-243.

 

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370–396. doi:10.1037/h0054346

 

Mayr E (1974) Behavior programs and evolutionary strategies. American Scientist.  62:650-659.

 

McBride G (1976) The study of social organizations.  Behaviour 59(1-2):96-115.

 

McNight RR, Copperberg GF, Ginter EJ (1978).  Duration of tonic immobility in lizards (Anolis carolinensis) as a function of repeated immobilization, frequent handling, and laboratory maintenance.  Psychol. Rec. 28:549-556.

 

Medvin MB (1990)  Sex differences in coloration and optical signaling in the lizard Anolis carolinensis (Reptilia, Lacertilia, Iguanidae).  Animal Behaviour 39:192-193.

 

Meier AH, Trobec TN, Haymaker R, MacGregar R. III, Russo AC (1973)  Daily variations in the effects of handling on fat storage and testicular weight in several vertebrates.  Journal Exp. Zool. 184:281-288.

 

Mellor DJ, Beausoleil, NJ (2015) Extending the ‘five domains’ model for animal welfare assessment to incorporate positive welfare states.  Animal Welfare 24(3):241-253. 

 

Mill JS (1882) A system of logic, ratiocinative and inductive, being a connected view of the principles of evidence and the methods of scientific investigation. 8th edition. New York: Harper and brothers [online] Available from http://www.gutenberg.org/ebooks/27942 (Accessed 7th December 2017).

 

Moehn LD (1974) The effect of quality of light on agonistic behavior of iguanid and agamid lizards. Herpetologica 8(2):175-183.

 

Moore IT, Jessop TS (2003)  Stress, reproduction, and adrenocortical modulation in amphibians and reptiles Ignacio T. Moore and Tim S. Jessop (2003) Hormones and Behavior 43 (2003) 39–47. 

 

Morton DB, Burghardt GM, Smith JA (1990) Critical anthropomorphism, animal suffering, and the ecological context.  In: Animals, Science, and Ethics.  S. Donnelley and K. Nolan, editors.  Hastings Center Report  20(3): 13-19. 

 

Morton ES (1975) Ecological Sources of Selection on Avian Sounds. Amer Naturalist 109(965)17-34.

 

Nature (2012) Animal Odd Couples.  Nature,  Season 31 Episode 3 https://www.pbs.org/video/nature-animal-odd-couples/ premiered Wednesday, November 7, 2012.

 

Nomura TKawaguchi MOno KMurakami Y (2013) Reptiles: a new model for brain evo-devo research. J Exp Zool B Mol Dev Evol. 2013 Mar;320(2):57-73. Doi: 10.1002/jez.b.22484. Epub 2013 Jan 14.

 

Norris KS (1953) The ecology of the desert iguana, Dipsosaurus dorsalis.  Ecology 34:265-287.

 

Novak MA, Suomi SJ (1988)  Psychological well-being of primates in captivity.  American Psychologist.  43(1):765-773.

 

OIE (2019) Office International Epizooties and World Organisation for Animal Health. Killing of reptiles for their skin, meat and other products. Available at: www.oie.int/standard-setting/ specialists-commissions-working-groups/scientific-commission- reports/ad-hoc-groups-reports

 

Orlans FB (1990) Policy issues in the use of animals in research, testing, and education. In: Animals, Science, and Ethics.  S. Donnelley and K. Nolan, eds,  Hastings Center Report, May/June Supplement. Pp. 25-30.

 

Painter Corinne, Lotz C (2007) (Eds.) Phenomenology And The Non-Human Animal: At The Limits Of Experience   Springer, Dordrecht, The Netherlands.  ISBN 978-1-4020-6306-0

 

Pearson AK, Tsui HW, Licht P (1976)  Effect of temperatures on spermatogenesis, on the production and action of androgens and on the ultrastructure of gonadotropic cells in the lizard Anolis carolinensisJournal Exp. Zoo., 195:  291-304.

 

Pough FH (1980) The advantages of ectothermy for tetrapods. The American Naturalist, 115(1):92-112.

 

Pound P, Ritsket-Hoitinga M (2018) Is it possible to overcome issues of external validity in preclinical animal research? Why most animal models are bound to fail. J Transl Med   16:304 8 pages.  https://translational-medicine.biomedcentral.com/track/pdf/10.1186/s12967-018-1678-1

 

Preston SD and de Waal, FBM (2002) Empathy: Its ultimate and proximate bases Behavioral and brain sciences 25(1):1-20 https://doi.org/10.1017/S0140525X02000018

 

Prinz J (2003) Emotions Embodied  in  Thinking about Feeling. R. Solomon (ed) Oxford University Press

 

Prosser CL (1986) Adaptational Biology: Molecules to Organisms. John Wiley & Sons, N.Y.

 

Putman BJ, Drury JP, Blumstein DT, Pauly GB (2017) Fear no colors? Observer clothing color influences lizard escape behavior. PLoS ONE 12(8): e0182146. https://doi.org/10.1371/journal.pone.0182146

 

Ragland IM, Wit LC, Sellers JC (1981)  Temperature acclimation in the lizards Cnemidophorus sexlineatus and Anolis carolinensis.  Comparative Biochemistry and Physiology 70A:33-36.

 

Rand AS (1964) Inverse relationship between temperature and shyness in the lizard Anolis lineatopus. Ecology 45:863-864.

 

Rand AS, Gorman GC, and Rand WM (1975) Natural history, behavior, and ecology of Anolis agassizi. Smithsonian Contributions in Zoology 174:27–38.

 

Rappaport R (1971) The Sacred in Human Evolution.  Annual Review of Ecology and Systematics Vol. 2:23-44. https://doi.org/10.1146/annurev.es.02.110171.000323

 

Regal PJ  (1966) Thermophilic response following feeding in certain reptiles.  Copeia. 1966(3):589-590.

 

Regal PJ (1980) Temperature and light requirements of captive reptiles.  In: Reproductive Biology and Diseases of Captive Reptiles, SSAR Contributions to Herpetology Number 1, J.B. Murphy and J.T. Collins, editors. Society for the Study of Amphibians and Reptiles.  Pp. 79-89.

 

Rociia CFD, Bergallo HG (1990) Thermal biology and flight distance of Tropidurus oreadicus (Sauria, Iguanidae) in an area of Amazonian Brazil.  Ethology, ecology, & Evolution 2:263-268.

 

Rodda GH, Bock BC, Burghardt GM, Rand AS (1988) Techniques for identifying individual lizards at a distance reveal influences of handling.  Copeia 1988(4):905-913.

 

Romm C (2015) Can a Predator Really Be Friends With Its Prey?  Atlantic Dec.11, 2015.     https://www.theatlantic.com/science/archive/2015/12/interspecies-friendships/420054/

 

Rose, P,  O’Brian M (2020) Welfare Assessment for Captive Anseriformes: A Guide for Practitioners and Animal Keepers Animals 2020, 10(7): 1132 (16pp);  https://doi.org/10.3390/ani10071132

Rothenberg A (1979) Einstein’s creative thinking and the general theory of relativity: a documented report. Am J Psychiatry. 1979 Jan;136(1):38-43.

 

Schwarzkopf L, Shine R (1991) Thermal biology of reproduction in viviparous skinks, Eulamprus tympanum: why do gravid females bask more?  Oecologia 88:562-569.

 

Seligman MEP   (1970On the generality of the laws of learning. Psychol Rev 77(5):406418.   CrossRefGoogle Scholar

 

Shah P,  Michal, A, Ibrahim A,  Rhodes R, Rodriguez F  (2017) What makes everyday scientific reasoning so challenging?  Psychology of Learning and Motivation  Volume 66, Chapter 7: Pages 251-299.  https://doi.org/10.1016/bs.plm.2016.11.006

 

Sherman PW (1988) The levels of analysis.  Animal Behaviour 36(2): 616-619.

 

      Shettlesworth SJ  (1972)   Constraints on Learning.  Advances in the Study of Behavior 4:1-68

 

Shine R (1980) “Costs” of reproduction in reptiles. Oecologia 46:92-100.

 

Shine R, Munn AJ, Stewart M, Vyssotski AL, Lesku JA (2015) Is “cooling then freezing” a humane way to kill amphibians and reptiles? Biology Open 4 (7):760-763. Doi:10.1242/bio.012179

 

Siegforda J, Jensen MB, Haskell M (2018) Preface. Applied Animal Behaviour Science 205 (2018) 181–182.

 

Stamps JA (1976)  Egg retention, rainfall and egg laying in a tropical lizard, Anolis aeneus  Copeia 1976:759-764.

 

Suboski MD (1992)  Releaser-induced recognition learning by amphibians and reptiles. Animal Learning and Behavior 20(1):63-82.

 

Sugerman RA (1990)  Observer effects in Anolis sagreiJournal of Herpetology. 24(3):316-317.

 

Sugerman RA, Hacker RA (1980) Observer effects on collared lizards.  Journal of Herpetology. 14(2):188-190.

 

Summers CH (1988) Chronic low-humidity stress in the lizard, Anolis carolinensis: Effects on ovarian and oviductal recrudescence.  Journal Experimental Zoology 248:192-198.

 

Summers CH (2002) Social Interaction Over Time, Implications for Stress Responsiveness. Integrative and Comparative Biology, 42(3):591-599. https://doi.org/10.1093/icb/42.3.591

 

Tinbergen N (1951)  The Study of Instinct.  Clarendon Press, Oxford.

 

Toadvine T (2007) How Not to be a Jellyfish: Human Exceptionalism and the Ontology of Reflection.  In: C. Painter and C. Lotz (eds.), Phenomenology and the Non-Human Animal, 39–55. (e-book) Springer, Dordrecht, The Netherlands.

 

Tokarz RR, Summers CH (2011) Stress and reproduction in reptiles. In: Norris DO, Lopez KH (eds) Hormones and Reproduction of Vertebrates. Academic Press, pp 169-213.

 

Tollis MHutchins EDKusumi K (2014)   Reptile genomes open the frontier for comparative analysis of amniote development and regeneration. Int J Dev Biol. 2014;58(10-12):863-71. Doi: 10.1387/ijdb.140316kk  https://www.ncbi.nlm.nih.gov/pubmed/26154326

 

Vogt JL, Coe CL, Levine S (1981)   Behavioral and adrenocorticoid responsiveness of squirrel monkeys to a live snake:  Is flight necessarily stressful?  Behavioral and Neural Biology 32:391-405.

 

Von Uexküll J (1909)  Umwelt und Innenwelt der Tiere, Berlin, Springer. Translated and excerpted in Foundations of Comparative Ethology, G.M. Burghardt, editor. Van Nostrand Reinhold, N.Y.  pp. 222-245

 

Wade J (2012) Sculpting reproductive circuits: relationships among hormones, morphology and behavior in anole lizards.  Gen Comp Endocrinol. 2012 May 1;176(3):456-60. Doi: 10.1016/j.ygcen.2011.12.011. https://www.ncbi.nlm.nih.gov/pubmed/22202602

 

Warwick C (1990)  Reptilian ethology in captivity:  Observations of some problems and an evaluation of their aetiology.  Applied Animal Behaviour Science 26:1-13.

 

Warwick C (1991)  Observations on disease-associated preferred body temperatures in reptiles.  Applied Animal Behaviour Science 28:375-380.

 

Warwick C, Steedman C, Kimm V, Holford T (1992)  Ethological and biological considerations of reptiles in scientific research.  A Report to the Commission of the European Communities. November 1992.  50 pp.

 

Warwick, C., Bates, G., Arena, P. & Steedman, C. (2018) Reevaluating the use of hypothermia for anesthetizing and euthanizing amphibians and reptiles. Journal of the American Veterinary Medical Association, 253:12;1536-1539. https://avmajournals.avma.org/doi/10.2460/javma.253.12.1536

 

Whittier JM, Crews D (1987) Seasonal Reproduction: Patterns and Control. In: Norris DO, Jones RE (eds) Hormones and Reproduction in Fishes, Amphibians, and Reptiles. Springer, Boston, MA  DOI:  https://doi.org/10.1007/978-1-4613-1869-9_13

 

Wingfield JC, Maney DL, Breuner CW, Jacobs JD, Lynn S, Ramenofsky M, Richardson RD. (1998)  Ecological bases of hormone-behavior interactions: The “emergency life history stage”. American Zoologist / now Integrative and Comparative Zoology. 38:191-206.

 

Zani, P. A., T. D. Jones, R. A. Neuhaus, J. E. Milgrom (2009) Effect of refuge distance on escape behavior of side-blotched lizards (Uta stansburiana). Canadian Journal of Zoology, 2009, Vol. 87, No. 5 : pp. 407-414.  https://doi.org/10.1139/Z09-029.

 

Zelena, D. (2015) The Janus Face of Stress on Reproduction: From Health to Disease. Int J Endocrinol  2015, Article ID 458129, 10 pages.  http://dx.doi.org/10.1155/2015/458129

 

 

 

 

 

Figure 1.  Dermal Chromatophore model of stress hormone interactions in Anolis carolinensis (adapted from Greenberg and Crews 1983). Acute and chronic stressors are integrated to evoke the release of hormones that interact with each other and converge in their effect on the darkness of a chromatophore. Epinephrine preferentially stimulates beta-receptors, but when betas are saturated, alphas are then stimulated. Activating beta-adrenergic receptor and MSH receptors evoke a darker (brown) cell when melanin disperses; alpha-receptors when stimulated can override the other responses and lead to a lighter coloured (green) cell. The chromatophores of the ‘eyespot’ (just behind the eye) is unique in possessing no alpha receptors so it will look dark even when the rest of the body is green. Animals in extreme distress may look mottled. ACTH, adrenocorticotropic hormone; CS, corticosterone; E, epinephrine (adrenalin); MSH, melanocyte stimulating hormone; NE, norepinephrine (noradrenalin). CS can elevate the ratio of E to NE released when stimulated by facilitating a key enzyme within the adrenal gland. E stimulates beta-receptors preferentially and then alpha-receptors, which has the opposite effects.

 

 


Table 1. Practice Informed by the Ethological Attitude*


  • Prioritise the Organism: Our direct perception of the animal we interact with is the center of our concern. A sense of the whole animal and its welfare should precede any analysis of its parts.
  • Define the Behaviour: Utilise a shared vocabulary at all levels of complexity. Describe behaviour in as much detail as possible and from multiple perspectives such as those organised in DEEP ethology. Elements of definition involve:
    • Direct Desription: Avoid inferences about behaviour at one level based on expectations from other levels, while being mindful that some participant phenomena are barely or only subliminally presented or perceived.
    • Indirect Description: Reliance on technological prostheses to extend our senses  must be approached with care and provided with independent corroboration utilising alternate techniques.
  • Cultivate a Janusian Perspective: Aim at simultaneously adopting both past/future and analytical/synthetic perspectives. Consider top-down and bottom-up perspectives.  Balance generalities and specifics.
  • Be Aware of and Eschew Congenital and Acquired Bias
    • Consider the respective umwelten of ourselves and of others: This will emerge out of the attention to all four perspectives in DEEP ethology.
    • Employ epoché: A process at the core of the Phenomenological Method that allows for the bracketing—setting aside but not denying–the possible influence of preconceptions and assumptions.

*The ‘ethological attitude’ is informed by a ‘phenomenological attitude’, close to heart of the qualitative data gathering and the interpretative procedures of generations of philosophers, theoreticians, and practitioners of the phenomenological method.

 

Table 2.  DEEP ethology.*

DEVELOPMENT

ECOLOGY

EVOLUTION

PHYSIOLOGY

Definition: Change within individuals from conception to demise involves the progressive expression of genetic potential as enabled or suppressed in specific environments.

 

Developmental change involves neural, physiological and behavioural plasticity, enabling adaptation to environmental change.)

 

Definition: The abiotic contexts of behaviour including climate and geology and the biotic contexts of behaviour including predators, prey, and conspecifics (including networks of reciprocal interactions of organisms with each other and their environments.)

  •  
  • Environmental change confronts selection pressures with which organisms must cope to maintain stability.

Definition: Change across generations and the study of probable ancestors and likely descendants.

 

‘Ultimate’ causes and consequence of behaviour

involve biological variations that are naturally selected and determine the nature of future generations.

Definition: How animals function and manifest  cascades of ‘proximate’ biological causes and consequences resulting in the expression of any behavioural trait.

 

Physiological change integrates multiple levels of organisation in coping with selection pressures to maintain the stability of the organism (homeostasis).

Key words: ontogeny, experience, genetics,  epigenesis.

Key words: biotic (predators, prey, conspecifics), abiotic (geology, climate), and umwelt (environment as perceived by animal).

Key words: genes, memes.  adaptation, direct and indirect fitness. natural selection.

Key words: homeostasis; nervous system and endocrine system; stress.

Kinds of questions:

When did the behaviour appear? How does the predictable change in a developing organism interact with the unpredictable change in its environment? Does the likelihood of a specific behavioural pattern change throughout one’s life? Would a specific experience affect the organism in different ways at different ages? Why? 

Kinds of questions:

What resources are available to enable an animal to meet its needs? Is the likelihood of a specific behavioural pattern different in different (physical or social) contexts? What aspect of the environment enables or impairs specific experience and adaptive change? Why? Given a specific ecology, what are the costs and benefits of a particular trait? 

Kinds of questions:

Are comparable patterns seen in kin or other progenitors? What are the ultimate causes and consequences of the likelihood of a specific behavioural pattern? How is it adaptive? How does it cope with selection pressure? How do specific traits affect direct or indirect fitness?

 

​Kinds of questions:

What are the proximate causes and consequences of the likelihood of a specific behavioural pattern? How is a current experience integrated with past (and possible anticipated) experiences? What is the path information takes considering ‘top-down’ and ‘bottom-up’ information transfer); How does error-detection help the organism navigate its umwelt? 

*The DEEP view of behaviour begins with an inventory of elements of behaviour and an ethogram, the patterns these elements manifest. Elements of behaviour may range from very modest (eye blink, tail twitch); patterns are exemplified by assemblies of coordinated behaviour such as courtship or predation.  This approach is also a formula for interdisciplinary collaboration and creative interactions between the disciplines. There is an inevitable, perhaps necessary and useful tension between the relative importance of generalities and specifics.

 

 

Table 3.   Ethologically Informed Design (EID)*


  1. Describe and define  the behaviour of primary concern relative to the problem to be solved or the question asked in terms of:
    1. Units of behaviour (‘behaviour inventory’)
    2. Patterns of behaviour (‘ethogram’)

 

  1. Assess DEEP ethology by asking key questions about the behaviour of interest from each of the DEEP perspectives as described in Table 2:
    1. Development (When is the behavioural trait of interest first seen? How has it manifested over time?)
    2. Ecology (In what environment(s) is the behavioural trait of interest seen?)
      1. Biotic (conspecifics, predators, prey, symbionts)
      2. Abiotic (climate, geology)
    3. Evolution (When is the trait of interest first manifest; propects for transmission)  
      1. Ancestral history
      2. Siblings, or contemporary kin
      3. Descendents
    4. Physiology (What processes act to maintain homeostasis? What are the internal and external expressions of those processes?)

 

  1. Identify and prioritise the biological needs of your subject(s) (a biological version of Maslow’s (1943) hierarchy of motivational needs) How does behaviour of interest serve:
    1. Physiology (homeostasis, health)?
    2. Safety (from environmental extremes, predators, and competitive conspecifics)?
    3. Sociality (nurture, courtship, reproduction)?
    4. Individuality (uniqueness, traits of possible interest to reproductive partners)?
    5. Self-Actualisation (maximising direct fitness (survival and prosperity of progeny) and/or indirect fitness (propagation of genes shared with more-or-less closely related kin)?

*Ethologically Informed Design enables practitioners in research, husbandry, and veterinary practice to more fully contextualise and thereby minimise sources of confound or confusion in planning, executing, and interpreting their actions with respect to the animals for which they are responsible. The questions of a unit or pattern of behaviour suggested by the categories above will help identify the gaps in understanding the causes and consequences of behaviour that could impair our efforts to secure the optimal welfare of an animal in any management situation.