Historically, a functional approach has helped biology recover coherence whenever perspectives have become fragmented. Darwin’s work is the most obvious and far-reaching example: by showing that species’ traits could be explained by natural selection, he provided a unifying framework that integrated anatomy, behaviour, and ecology under a common principle of function and evolutionary adaptation to particular conditions (Darwin, 1859). Mayr later emphasised that understanding biological phenomena requires both proximate causes (“how” traits operate within an organism) and ultimate causes (“why” those traits evolved) (Mayr, 1961, 1988). Tinbergen’s four questions built on this by operationalising those distinctions and providing a system of inquiry that encourages researchers to engage with four dimensions that distinguish between evolutionary and within-lifetime adaptation (Burkhardt and Hauber, 2014; Tinbergen, 1963). Physiology has similarly used functional questions to link mechanism to survival value (Noble, 2011; Noble and Noble, 2021; Ramsay and Woods, 2024; Schulkin, 2004). In physiology and neurobiology, the concept of allostasis has extended homeostatic models by emphasising that, during their lifetime, organisms anticipate and adapt to changing conditions (McEwen, 1998; Schulkin, 2004; Sterling and Eyer, 1988). These functional syntheses share a common thread because they reinterpret biological mechanisms and explain that evolutionary adaptation has shaped organisms’ adaptive capabilities (and limitations) for within-lifetime adjustment.

Nevertheless, the idea that living organisms can be better understood in terms of their ends, functions, or purposes (i.e., that it helps to ask “what is this attribute good for?”) has deep roots in Western thought, reaching back to Aristotle’s teleological metaphysics (Brennan, 2002; Cameron, 2010; Logan, 1897; Manning, 1998; Woodfield, 1998, 2016). For Aristotle, nature was inherently purposeful; every living thing had a telos, a set of functions and capabilities that constitute its nature, by which it could be explained (Hauskeller, 2005; Logan, 1897; Soontiëns, 1992; Woodfield, 1998). As philosophers tried to explain not only “what” but “why” such purposes exist, attention shifted away from organisms’ internal organisation to the question of who or what imposed those purposes – a preoccupation that would dominate Western thought for centuries (Brennan, 2002; De Cruz and Smedt, 2015; Logan, 1897; Sedley, 2017; Soontiëns, 1992). Subsequent philosophical and theological traditions reframed teleology’s intrinsic purposiveness in terms of externally imposed divine design until the Enlightenment, when efforts to separate science from theology elevated mechanistic accounts of life and diminished teleology’s credibility (Boyle, 1688; Des Chene, 2001; Dresow and Love, 2023; Ghiselin, 2002; Hempel, 1994; Mayr, 1988, 1992). Even Darwin’s functional explanations of evolutionary adaptation were interpreted by some contemporaries and followers to imply finalism [the idea that evolution is invariably propelled toward increasingly perfected forms, culminating in humans (De Chardin, 2017; Vidal, 2021)], another theory that was eventually rejected as scientifically untenable (Driscoll et al., 2009; Ghiselin, 1994; Soontiëns, 1992).

In the twentieth century, the biologists Pittendrigh and Monod sought to recover the explanatory value of functional explanation without its metaphysical burden (Monod et al., 1974; Pittendrigh, 1958). Pittendrigh introduced the term teleonomy to describe the autonomous goal-directed (telos-directed) organisation of living systems as a driver of evolution by natural selection without a designer (Pittendrigh, 1958). Teleonomy continues to be a live topic of debate in evolutionary biology, genetics, and the philosophy of science, where questions of evolutionary causation, directionality, and agency continue to provoke disagreement (Dawkins, 2024; Dresow and Love, 2023; Hennig, 2011a; Noble, 2011). Yet despite ongoing conceptual tensions, recent developments suggest a renewed and multidisciplinary interest in teleonomic thinking (West-Eberhard, 2025) including fields such as systems biology, theoretical neuroscience, and affective science. Here, metaphysical disputes are increasingly being set aside in favour of integrative frameworks that better reflect how complex, goal-directed biological systems operate and influence each other in real time. Recent interdisciplinary efforts that illustrate this revival include the Linnean Society’s 2021 Evolution “On Purpose” meeting (Linnean Society, 2021), its associated multi-author volume (Corning et al., 2023), the companion overview article (Vane-Wright and Cornig, 2023), and the Human Affectome Project (Schiller et al., 2024). The last of these shows how framing biological and psychological processes in terms of their evolutionary origin and within-lifetime adaptive purpose can bring coherence to the multidisciplinary study of affective phenomena in humans.

Teleonomic approaches recognise organisms as systems shaped by evolutionary pressures to meet the recurrent challenge of maintaining viability and reproduction in dynamic environments (Birch, 2009, 2020; Vane-Wright and Cornig, 2023). Through a teleonomic lens, the components of the system (i.e., the organism’s teleonomic subsystems, traits, and regulatory processes) are interpreted in relation to the viability problems they evolved to address. In animal sciences, problem-solving activities matter because they are biologically relevant to the organism under study (i.e., they affect their survival, and reproductive success) (Fraser and Duncan, 1998; McEwen, 1998; Schulkin, 2004). In animal welfare science, a field that depends on contributions from many biological disciplines, teleonomy offers a way to unify perspectives and generate new questions and insights. Furthermore, the organism-centric perspective required in animal welfare science could benefit all biological fields.

To our knowledge, there is no single term that captures the organisation that we see realised in the distinctive forms of life that evolution has produced to “fit biologically” in the extraordinarily diverse environmental conditions of planet Earth (Carroll, 2001; Crow and Wagner, 2006; Page, 2011). Bernard Rollin’s concept of animal telos was an early attempt to do so (Rollin, 2012, 2014, 2017). When he wrote that telos is “what is encoded in an animal’s genetics, as expressed in their normal environment” (Rollin, 2014, p. 100), he was describing animals as a suite of interconnected traits that were not merely present in the genome, but functionally expressed in real-world contexts. His memorable metaphors “the pigness of the pig, the dogness of the dog” condense species-specific anatomical, physiological, and behavioural features into an intuitive shorthand for how animals have evolved their functionality alongside how they express it.

For Rollin, an animal’s nature implies a purposive structure, and telos (literally “end” or “purpose”) means the unity of form, function, and context. He argued that evolved biological structures carry with them functional expectations that are obvious when they are considered within the environments to which their species is adapted. By appealing to commonsense with phrases such as “fish gotta swim; birds gotta fly,” Rollin was not just making a normative claim, but an ontological one: a fish and a bird are the kind of biological entities that make sense only in relation to the aquatic and aerial environments that allow their morphology, physiology, and behaviour to be expressed. If you alter the animal or remove the ecological context, the functional coherence of the organism collapses. For example, clipping a bird’s wings forecloses a central mode of life that the organism’s body and behavioural repertoire evolved to express. In Rollin’s ethical framework, if we constrain an animal so that it cannot be the animal it evolved to be, we violate something fundamental to its being, even if it is productive and we avoid causing it pain, hunger, or disease (Rollin, 2003, 2015, 2019a, 2019b).

Rollin’s theory was a major advance in animal ethics because it grounded moral consideration in species-specific interests (Harfeld, 2013; Rollin, 2014, 2020; Rollin and Hickey, 2019). However, outside ethics, and particularly in animal welfare science, “telos” has seen limited uptake. One reason may be that the term carries connotations of extrinsic or divine purpose that remind some of biology’s rejection of teleological causation and finalism. However, and more fundamentally, by its own meaning, telos names the goal but not the organised system by which that goal is realised, making it difficult to operationalise in research and practice.

Recognising survival and reproductive success as the universal biological goals of life (telos) and acknowledging that each lineage has evolved a distinct system for pursuing telos, shifts the focus from the common teleonomic purpose to the unique expression of it in individual organisms. Accordingly, we propose a term for the integrated, self-organising system that is embodied by living beings and their connection to the environments that shaped them – the teleonome.

What follows develops the teleonome and the teleonomic principle as integrative constructs. Section 3 defines the teleonome and outlines the supporting concepts required to interpret it in animal welfare science. Section 4 shows the teleonome in action, its implications for individual adaptation, relevance, learning, and the organising role of affect. Section 5 develops focused topics and future directions, before turning to limitations and safeguards in Section 6 and concluding reflections in Section 7.

Our aim is to provide researchers with a principled way to situate their work within a broader understanding of animals as evolved, telos-directed organisms. In doing so, we emphasise that animals are not passive subjects of environmental forces but active agents whose biological organisation equips and motivates them to pursue adaptive ends. We believe this framing can motivate more integrative approaches to welfare assessment and theory building that are grounded in the animal’s perspective. We invite you to consider how familiar research questions and methods align with the teleonomic principle; how this can help explain not just what animals are and how they function, but why they are organised and function in those ways; as well as the ways in which their teleonome provides a framework through which those principles are expressed in living form.

The teleonome is the organised biological system through which organisms pursue biological goals (telos). It comprises the integrated system of perceptual, physiological, behavioural, and adaptive capabilities shaped by natural selection to pursue survival and reproduction in dynamic environments. Each species, population, and ultimately each organism possesses its own teleonome: a set of inherited traits and interconnected regulatory processes that provide them with potential for solving problems of viability in particular ecological contexts. As such, all autonomous living systems, from the simplest bacteria to humans, have a teleonome.

An organism’s teleonome can be understood only in relation to the ancestral environments and selection pressures that shaped it. However, because environments are never static, the teleonome includes inherited capabilities for individual adaptive modification within phylogenetically constrained bounds. It embodies a multi-generational gamble: that offspring will encounter a sufficiently similar ecological structure, and that their inherited adaptive mechanisms will be adequate to accommodate differences (Dawkins, 2024). The teleonome thus represents a heritage of potential and limitations for achieving teleonomic goals that are continually tested against changing conditions, linking ancestral efficacy to ongoing, goal-directed striving, while remaining open to variation and failure.

Compiling a teleonome (i.e., an inventory of teleonomic traits) for any type of organism or animal should follow a structured, systematic approach identifying traits, systems, and capabilities oriented toward viability and reproductive success. (See also section 5.1).

Etymologically, the word teleonome follows established conventions in biological nomenclature, constructed from the Greek teleo- meaning “end,” “goal,” or “purpose” (as in telos), and -nome from nomos, signifying “law,” “system,” or “management.” This construction parallels other biological terms such as “genome” (the complete set of genetic instructions) and “proteome” (the complete set of proteins), indicating a systematic collection organised toward particular functions. The term is meant to complement and refine established conceptual frameworks across multiple biological disciplines, including animal welfare science, animal husbandry, behavioural ecology, evolutionary biology, comparative physiology and psychology, conservation biology, and veterinary medicine. The teleonome provides the missing North Star – a fixed reference point, external to human preference, that reflects what is most likely to be biologically relevant to each animal.

This progression toward conceptual and semantic precision can be understood at three complementary levels: (a) telos identifies the universal biological goal of viability, survival, and reproduction, but does not specify how this goal is pursued; (b) teleonomy explains that goal- or telos-directed biological processes result from natural selection rather than divine design, but focuses on the principle rather than cataloguing specific traits; and (c) teleonome provides a comprehensive term for the complete, integrated biological system through which organisms actually pursue and achieve their biological goals. Table 1 summarises the relationships among telos, teleonomy, teleology, and teleonome, clarifying how these interrelated concepts operate at different explanatory levels. In this way, the teleonome provides the biological specifications for what Rollin intuitively grasped with telos – a framework that preserves his ethical insights while offering the conceptual precision needed for empirical investigation.

As noted earlier, each organism possesses a teleonome, a system or traits that can be considered on at least three levels (Figure 1). At the broader, species-level (e.g., domestic dogs, Canis lupus familiaris), the teleonome encompasses the inherited teleonomic traits shared by all members of the species: the species-typical morphology, physiology, sensory and affective capabilities, and behavioural tendencies that define its adaptive organisation. Species offer a useful reference point, even though the sharing of teleonomic traits may extend beyond species boundaries and be altered by natural and artificial selection. Within species, morphotype-level teleonomes represent stable subpopulation refinements of different ecotypes, subspecies, breeds, or strains (e.g., herding dog types) where selection, whether natural or artificial, has produced specialised adjustments. At the individual level, each organism inherits a unique configuration through genetic recombination and epigenetic marking, such that even siblings possess distinct teleonomic potential.

Finally, the expressed teleonome represents the real-time manifestation of the flexibility of these inherited capabilities as they are deployed in the present moment, modulated during the individual’s lifetime by environmental context and other factors including life stage, learning, and current physical and mental state. This distinction between inherited teleonomic traits and their real-time expression acknowledges that what organisms possess as evolved potential must be understood alongside how that potential is actualised under specific conditions (See also Section 4.5).

Critically, the teleonome is not synonymous with the genome. It refers to the organised functional capabilities that emerge from genetic information as expressed through development and in relation to environmental context (ancestral and present). Nor is it equivalent to the phenotype, which includes traits that may be non-functional byproducts or selectively neutral variations. Rather, the teleonome specifically encompasses those inherited, functionally organised capabilities oriented toward biological goals (telos).

Although the teleonome as a concept applies to all autonomous living systems, from the simplest bacteria to humans, welfare science focuses specifically on sentient, non-human animals (hereon ‘animals’) whose teleonomic capabilities include the generation of subjective experiences that are felt as pleasant or unpleasant from the animal’s perspective (Low et al., 2012). For sentient animals, the teleonome includes affective subsystems that evaluate environmental conditions (physical and social) and internal states in terms of biological relevance, producing experiences of valence and urgency that guide individual adaptive responses (behavioural and physiological) (Mendl et al., 2010; Mendl and Paul, 2020). These affective phenomena are themselves teleonomic, being evolved tools for flexible, rapid assessment and response (e.g., pain isn’t just information; it is a teleonomic alarm). This framing parallels the Human Affectome (Schiller et al., 2024) which formalised affective phenomena as hierarchically organised, teleonomic processes. In animals, the affectome can be understood as the subset of the teleonome responsible for generating and regulating affective experience. It is primarily the functioning of these affective-regulatory subsystems, rather than teleonomic organisation as a whole, that gives rise to welfare-relevant states in individual animals. Accordingly, while the teleonome characterises any living system, its welfare significance emerges most clearly in animals whose organised capabilities include the generation of subjective experience. From this point, we focus on the teleonome and the teleonomic principle as they apply to non-human sentient animals, with particular attention to how affective experience functions as an evolved system for adaptive regulation during individual lifetimes. As well as sentience, two other features, affordances and agency, are essential to understanding the teleonome in welfare terms. The following subsections outline how each of these three components specifies a different aspect of how the teleonome is expressed in real-time.

Understood physiologically, the bodies of sentient animals of welfare interest are constituted of numerous mechanistic networks that have whole body, organ system, organ, cellular, and subcellular functions. These networks are so extensively integrated that every function of each one interacts directly and/or indirectly with the functions of every other network. From a teleonomic perspective it is apparent that this edifice of intricately interactive functionality serves to preserve structural and functional integrity in pursuit of biological goals (telos) (Corning et al., 2023; Pittendrigh, 1958; Vane-Wright and Cornig, 2023). Importantly, these include sustaining the individual’s life until reproductive age is reached and securing the possibility of producing viable offspring. To achieve this, it is necessary to maintain conditions within the body’s ‘internal environment’ that are compatible with sustaining life in the face of an ever-fluctuating ‘external environment.’ There is also a need to support internal developmental changes from birth through infancy to youthful maturation, followed by puberty and mature reproductive or alloparenting capabilities. Within this complexity, two generalised functional phenomena of relevance to sustaining life-critical features in the internal environment have been described – namely homeostasis and allostasis.

In the teleonomic view, affect is not merely a subjective byproduct of neural processing; rather, it is a functionally integrated system for managing biological risk and opportunity in fluctuating environments (Schiller et al., 2024). At the heart of this adaptability, we propose a four-step process through which sentient animals execute adaptive responses:

The processes of detection, evaluation, and anticipatory forecasting generate the experience, i.e., felt affective phenomena that signal both the importance (salience) and the emotional value (valence: negative/positive) of the perceived change according to its relevance (Posner et al., 2005). The experience motivates, triggers, and controls the response; it tells the organism how much the situation matters, in what direction (approach or avoid) to respond, and how long to persist.

This four-step model reflects the functional architecture of affective experience as an evolved solution to the problem of adaptive decision-making. Experience is not incidental or epiphenomenal; rather it is central to the organism’s ability to regulate itself in the face of change, exercise agency to resolve challenges, seize opportunities, engage with available affordances, restore function, and maintain viability.

Animals identify and organise meaning by determining what is threatening, rewarding, or worth paying attention to, and what they should and can do about it. These evaluative processes arise from the interplay between their teleonome’s genetic inheritance and the non-inherited factors that influence its moment-to-moment expression. What matters to an animal at any given moment emerges dynamically from the ongoing process of filtering and prioritising information from both internal and external environments. This filtering keeps the animal attuned to conditions that are biologically relevant. By evaluating changes in relation to their current needs, capabilities, and vulnerabilities, animals focus attention and effort on the opportunities and challenges deemed most significant for maintaining viability.

In a teleonomic framing, relevance is actively enacted rather than passively received. The animal assigns significance to stimuli by relating them to what they can currently do, and what they need in that moment and situation. Although many of these tendencies reflect evolved predispositions, the teleonomic perspective treats them not as fixed ‘instincts’ but as flexible capabilities whose expression is shaped by developmental history, learning, and current conditions. This is not a static process; each encounter with the world updates the animal’s internal model, adjusting priorities and refining responses over time.

The four-step detection, evaluation, forecasting, and response process does not operate as a linear sequence; rather, it cycles continuously as the affective, physiological, and environmental consequences of each response generate new feedback that informs subsequent cycles (Figure 2). This dynamic process is well described in terms of an iterative Bayesian algorithm (Friston, 2010, 2012; Lecorps and Weary, 2024). That is, the animal begins with a prior expectation based on inherited tendencies and previous (learned) experiences about what a given stimulus might mean and how to respond. As new evidence accumulates through sensory input and the outcomes of action, these priors are updated. The system recalibrates, refining future predictions and narrowing response options to those perceived to be most adaptive.

Over time, this iterative loop supports increasingly efficient, context-sensitive regulation of behaviour and physiology. Behaviourally, it functions as a learning algorithm, using prediction errors to refine which actions are reinforced and repeated, and those that are punished and avoided (Emanuel and Eldar, 2023; Friston, 2012; Lecorps and Weary, 2024; Mendl and Paul, 2020). Physiologically, it allows animals to adjust not only their immediate responses, but the parameters they defend according to changing internal states and environmental demands (Ramsay and Woods, 2014, 2016, 2024). In this sense, the process reflects the principles of allostasis (the capability for dynamic, experience-dependent adaptation through anticipatory regulation and resource allocation). At a more detailed physiological level, there is a flexibly integrated interaction between homeostatic “steady state” fluctuations of salient parameters around specific “set points,” and an allostatic shifting of these set points and parameter fluctuations around them as a response to change. This process helps explain how animals can adapt rapidly to changing environments without deliberation, through embodied mechanisms of learning and regulation that are inherently probabilistic and error-corrective. Species differ in how rapidly and flexibly they adapt and, in some, this behavioural and physiological plasticity has played a central role in facilitating domestication.

However, when animals face conditions outside the range to which they are evolutionarily adapted, the regulatory systems that work to maintain stability during change can become overtaxed. This may occur when animals are confronted with extreme physical environments, or are exposed to pathogens, parasites, or other factors to which the species lacks immunity. It may also occur when their sensory systems cannot detect, or their behavioural repertoires cannot resolve, the challenges with which they are faced. Whilst short-term shifts in physiology and behaviour support survival, prolonged or repeated demands may generate cumulative dysfunction, colloquially “wear and tear,” known as allostatic load (Cicchetti, 2011). In such cases, the very mechanisms that enable flexible adjustment risk undermining long-term function, setting animals on declining welfare trajectories (Edes et al., 2018; McEwen, 1998; Schulkin, 2004; Seeley et al., 2022; Wolfe and Edes, 2023) (Figure 3). Most animal welfare research has tracked these processes using a limited range of biomarkers, most commonly glucocorticoids and changes in heart rate, which can be informative but are notoriously ambiguous, since both elevated and suppressed concentrations and rates can reflect dysregulation, and our interpretations depend on context, baselines, and individual variation [e.g (Monteiro et al., 2023; Stafford and Mellor, 2005)]. In response, researchers have called for the development of species-specific allostatic load indices, arguing that integrated measures are needed to capture the cumulative physiological costs of lifetime adaptation (Edes et al., 2018; Seeley et al., 2022; Wolfe and Edes, 2023). These proposed indices would align well with, and be strengthened by, the construction of species-specific teleonomes and affectomes. Taken together, teleonomic, affective and allostatic load profiles could provide the conceptual scaffolding and the practical inventories needed to identify which biomarkers, behavioural indicators, and affective phenomena are most relevant to each species’ evolved capabilities and vulnerabilities.

At any given moment, the way in which an individual’s teleonome is expressed reflects the interaction between inherited capabilities and a suite of non-inherited influences. Phylogenetically inherited perceptual, physiological, and behavioural tendencies form the structural basis of teleonomic relevance (Hennig, 2011b; Linnean Society, 2021; Pagel, 1997; Pittendrigh, 1958). They give rise to ‘evolutionary priors,’ or expectations about what is likely to matter in the animal’s world and how they are likely to respond (Nave et al., 2020; Tooby et al., 2024). These inherited expectations may be modulated by epigenetic processes that fine-tune responsiveness (Gibney and Nolan, 2010; Jaenisch and Bird, 2003). However, the real-time manifestation of the animal’s biological potential is modulated and governed by non-inherited factors during the individual animal’s lifetime (Moczek, 2015; West-Eberhard, 2020) (Figure 4).

Non-inherited influences on the expressed teleonome include:

These influences interact continuously, allowing the animal to regulate their position within a tolerable range of states (adaptive operational zones) and to shift behaviour in ways that support viability. When challenges are too intense, too frequent, and/or too prolonged, they may overwhelm the organism’s capability to update or recover (sometimes labelled coping), leading to high allostatic load, reduced resilience, and compromised welfare (Arndt et al., 2022; Wolfe and Edes, 2023).

Understanding this dynamic architecture of experience is essential in animal welfare science, where the core aim is to assess how animals feel, and how their feelings shift in response to the environments in which we choose to locate them [e.g (Green and Mellor, 2011; Littlewood et al., 2023; Mellor et al., 2020; Rault et al., 2025)]. Accordingly, science has moved from snapshots in time to welfare trajectories in order to capture the temporal and contextual nature of animal welfare [e.g (Kirkland and Cunningham, 2012; Paul et al., 2023)]. Monitoring welfare involves more than checking for physical health or abnormal behaviour; it requires interpreting how animals appraise their conditions based on what matters to them (Mellor et al., 2020). This interpretive step requires recognition of when their inherited expectations and capabilities, accumulated experience, or current capabilities are functioning well or being over-taxed. By tracing these interacting influences, welfare assessments can become more biologically based and more attuned to the individual animal’s perspective, helping identify not only when welfare is compromised, but why.

The nested architecture proposed in the Human Affectome (Schiller et al., 2024) provides a biological rationale for weighting animal welfare indicators according to their evolutionary and functional relevance. Schiller et al. (2024) propose that affective processes are organised hierarchically from immediate, survival-related concerns to broader, long-term ones, each operating over different temporal and spatial scales. This hierarchy reflects a continuous evaluation of change in terms of urgency and ongoing viability. This aligns with animal studies that see affective experience as a structured system for prioritising different levels of adaptive relevance (Panksepp, 2005; Panksepp et al., 2011; Panksepp and Zellner, 2004). Proximal concerns (e.g., breathlessness, pain, hunger, thirst, and thermoregulatory perception) signal immediate threats to physiological stability and may, therefore, command higher priority action (Beausoleil and Mellor, 2015b; Mellor, 2010). Distal concerns operate over longer timeframes, shaping future behaviour, learning and mood (Špinka, 2019). The layers interact such that chronic disruption of distal concerns can eventually alter physiological regulation, just as unresolved acute needs can cascade into longer-term dysregulation (Colditz and Hine, 2016; Mason and Rushen, 2006).

The animal affectome, here regarded as the subset of the teleonome responsible for generating, interpreting and prioritising affective responses, defines how the animal responds to change. It encompasses both self-regulatory, interoceptive systems (e.g., breathlessness, pain, hunger, thermoregulation) and social affective systems that monitor and regulate relationships with conspecifics (e.g., isolation distress, affiliative bonding, social confidence). For social species, both interoceptive and social-affective systems are equally fundamental to maintaining viability. This means that welfare assessments that prioritise physiological indicators, while treating social deprivation as secondary, fail to recognise the integrated teleonomic architecture through which these animals regulate their existence. At the individual level, consistency in affective and behavioural responses, often called temperament or personality, has been documented across diverse taxa and shown to have heritable bases and fitness consequences (Réale et al., 2007). Teleonomy explains why such individual differences persist: different response strategies represent alternative solutions to the same regulatory challenges under varying ecological and social conditions.

Teleonomic reasoning thus offers a biological rationale for weighting welfare indicators according to their biological relevance, offering a principled way to address the persistent challenge of determining the relative significance of different experiences on an animal’s overall welfare (Wilkins et al., 2024). Mapping affective hierarchies within species-specific affective profiles may help interpret how multiple experiences interact, compete, or compound to influence behaviour and welfare trajectories. This approach links affective neuroscience and welfare assessment under a shared biological logic; namely, one that infers what matters to animals not by analogy with human feelings, but by reference to their evolved systems for staying alive and well.

Future work should focus on developing and refining knowledge of specific perceptual, affective, and motivational systems at species- and morphotype-levels, based on empirical evidence and ethological insights. These profiles should be informed by knowledge of the evolutionary pressures that shaped the species’ teleonomes. They should remain flexible and open to revision as new evidence emerges and be co-developed with welfare practitioners to ensure relevance and practical application. These requirements will ensure that the teleonomic framework will remain relevant, useful, and scientifically progressive.

As noted earlier, compiling a teleonome (i.e., an inventory of teleonomic traits) for any type of animal should follow a structured, systematic approach identifying traits, systems, and adaptive capabilities shaped by natural and artificial selection to achieve survival and reproductive goals. Inclusion may be guided by evidence that: (a) the trait is historically adaptive (i.e., its structure, integration, and distribution across related species suggests functional contributions to ancestral fitness); (b) the underlying motivational system remains functionally intact and capable of being triggered and expressed given appropriate environmental conditions and affordances (even if currently latent); and (c) the trait is organised around detecting or exploiting specific ancestral affordances.

Evidence that preventing a trait’s expression leads to physiological stress, stereotypies, injury and/or disease, impaired development, or negative affective states may strengthen the case for inclusion. Such outcomes indicate that the associated motivational and regulatory systems remain active, even if the trait is currently constrained or unsupported by the environment. The teleonome also encompasses inherited capabilities and limitations for teleonomically relevant learning and allostasis. Grounds for provisional exclusion could include the following: traits lacking current evidence of historical selection for functional outcomes; those showing no apparent integration with internal regulatory or affective systems; and those whose manipulation produces no detectable welfare impacts. An example is variations in traits for which there is currently no evidence of integration with viability relevant systems. Nevertheless, teleonomes should be regularly updated as new evidence emerges (Biémont and Vieira, 2006; Brancalion et al., 2022; Pennisi, 2012).

Understanding a species’ teleonome, including the affective capabilities and the ecological conditions to which it is evolutionarily fitted, may support more refined welfare assessments by shifting focus from momentary measures to patterns of chronic affective dysregulation. Indicators of sustained allostatic load, persistent hyperarousal, or blunted affect reveal long-term strain on regulatory systems that short-term ‘snapshot’ evaluations can overlook (Edes et al., 2018; Seeley et al., 2022). Adopting a teleonomic perspective emphasises welfare as an ongoing process of adaptive regulation rather than a static state by its linking of physiology, affect, and behaviour through the common principle of maintaining viability under changing conditions.

Recent work has called for the development of species-specific allostatic load indices to quantify cumulative physiological costs of adaptation over time (Wolfe and Edes, 2023). The teleonome provides the conceptual foundation for this effort: because each lineage has evolved distinct regulatory architectures and affective sensitivities, the biomarkers that best indicate chronic dysregulation at different life stages will differ accordingly.

Teleonomic systems keep adjusting until they are no longer able to; therefore, endurance in suboptimal conditions should not be misinterpreted as resilience. Animals may continue to function while accruing cumulative costs, and it is precisely the organism’s persistence in adapting that can systematically hide deteriorating welfare trajectories. Teleonomically grounded indices could therefore integrate endocrine, behavioural, and autonomic measures in ways that reflect the species’ evolved organisation and ecological history (McEwen, 1998; Wolfe and Edes, 2023). Such longitudinal tools could help identify early warning signals of affective compromise before overt pathology emerges, advancing both the predictive and preventive dimensions of welfare science.

The teleonomic principle offers a foundation for aligning experimental design with the animal’s evolved capabilities and affective relevance. Future research should evaluate whether experimental tasks activate meaningful concerns or impose artificial constraints. Teleonomic reasoning may help guide the design of control conditions that correspond to the species-evolved features, while also accommodating individual variation. The purpose is to avoid defaulting to established management and husbandry practices that may conflict with the animal’s teleonome, impose unnecessary allostatic load, or reflect anthropocentric assumptions.

For example, in domestic horses, the use of equipment such as bits or nosebands, or housing in individual stables, may be considered standard control conditions in equitation and equine science studies. Yet such conditions are teleonomically inconsistent: horses did not evolve to solve the challenges these conditions impose, and the conditions are far removed from the ecological and social contexts that shaped the equine teleonome. So, using such baselines can confound interpretation by introducing unacknowledged stressors that mask or distort the measured effects of experimental interventions, leading to underestimation of their true impacts on affective experience and welfare.

Understanding the focal animal’s teleonome also enables interpretation of findings in terms of motivational relevance, not merely performance metrics or statistical significance. A teleonomic framework can also help avoid common interpretive errors. These include: misclassifying high arousal as preference; mistaking decreases in physiological measures for calm states; assuming that no measurable change implies no welfare effect; construing behavioural inhibition due to an animal’s low expectations of control as indicating resolution; or restrained or controlled behaviour as evidence of assent, especially during interactions with humans. These risks are increasingly recognised in welfare-focused behaviour modification research, and are explicitly addressed in recent methodological guidance such as the COMPASS Guidelines (McGreevy et al., 2026), which emphasise species-relevant controls, motivational validity, precautionary stopping criteria, and careful interpretation of behavioural and physiological indicators in studies involving animal training, handling, or restraint.

A teleonomic lens can reveal the ways in which evolved motivational systems shape learning, prediction, and decision-making. As framed by teleonomic principles, learning is a process by which organisms update internal models to avoid threats and pursue opportunities in dynamic environments (Lecorps and Weary, 2024). The teleonome provides the species-specific architecture (motivational, perceptual, and affective) that constrains what an animal can learn and how.

Reinforcement and punishment are therefore more than behavioural contingencies; they carry affective weight that modulates how the animal evaluates contexts, predicts outcomes, and adapts to change. Future research should focus on characterising the affective valence, salience, and subjective value of stimuli used in training, and on examining how affective states influence prediction errors, learning speed, and generalisation. Attention should especially be given to how these processes interact with agency, control, and persistence of positive expectation, all of which emerge from the organism’s teleonomic organisation (Colditz and Tilbrook, 2022; Littlewood et al., 2023; Maier and Seligman, 2016; Webster, 2008).

Such investigations can illuminate how Bayesian updating of expectations contributes to adaptive stability and competence building or, conversely, to chronic dysregulation, pessimism, and learned helplessness (Lecorps and Weary, 2024), especially in contexts involving ambiguity or loss of control. Taken together, teleonomic assumptions will support more nuanced interpretations of animal training contexts, outcomes, coping strategies, and the mechanism through which experience shapes adaptive engagement with the social and physical environments.

As species-specific teleonomic profiles mature, they may inform veterinary decision-making (e.g., balancing treatment goals with affective impact), policy and regulation (e.g., transport and slaughter standards based on species), as well as enrichment and husbandry design (e.g., aligning environmental features with species-specific motivational systems) and end-of-life decisions. In each case, the goal would be to design these experiential events not just around the animal, but with the animal’s affective and motivational world in focus.

Teleonomic reasoning centres analysis on the individual organism, seeking to understand the world from their perspective and recognising that the pursuit of viability is immanent – arising from within the animal. This principle aligns with interest-based ethical frameworks such as Bernard Rollin’s telos (Rollin, 2017) and Christine Korsgaard’s extension of Kantian ethics, which treats animals as ends in themselves, and therefore grounds the moral question “what ought to matter to us?” in what actually matters to them, from their own perspective (Birch, 2020; Korsgaard, 2004, 2023, 2013). It is compatible with core aspects of Martha Nussbaum’s (2023) interests and capabilities approach, which focuses on enabling animals to exercise their species-specific capabilities and fulfil the lives their natures make possible (Nussbaum, 2023; Tulloch, 2011). Together, these perspectives support the idea that welfare assessment should consider whether animals have the agency and affordances that enable them to engage in the activities and experiences that their evolved systems are designed to support.

A teleonomic approach that incorporates species-specific teleonomes highlights that many animals now inhabit environments that no longer match the conditions their functional organisation evolved to meet. As a result, the teleonomic systems that once motivated adaptive behaviour may be under-stimulated, misdirected, or unable to find expression. Captive animals (exotic and domesticated), for example, may retain substantial elements of the motivational architecture of their ancestral teleonome but limited agency and/or affordances (i.e., opportunities) to deploy their physical and cognitive capabilities through exploration, risk taking, and learning to control their surroundings, all of which could be central to their physical and psychological wellbeing.

Domesticated animals face a further layer of complexity because as well as living in managed environments, their teleonome has been altered by artificial selection. Sometimes, the traits favoured by humans reduce the physiological and psychological costs of captivity, while others impair viability or create extreme dependency on human care. Many domesticated species are now equipped to survive in human environments where their wild ancestors could not. However, they are often simultaneously rendered incapable of surviving without artificial care and reproductive assistance, while retaining the motivational systems that artificial environments frequently fail to support. Artificially selected extreme morphotypes are associated with compromised welfare (Asher et al., 2009) but even animals bred for non-extreme attributes can have elevated levels of disease.

Selective breeding practices in pedigree dogs, including the use of closed studbooks and popular sires, inadvertently increase the expression of inherited disorders that are not directly linked to breed standards (Summers et al., 2010). Many of these disorders, including conditions affecting the nervous-sensory, cardiovascular, and immune systems may not manifest until adulthood, complicating breeding decisions and welfare assessment (Collins et al., 2010; Summers et al., 2010). This dissociation explains why domesticated animals may be numerically successful population-wise, while individuals experience chronic welfare compromise. These situations expose the profound epistemic and ethical challenges of intervening in evolved systems. While removing a trait and its motivational architecture through genetic engineering might theoretically eliminate associated suffering in environments that prevent its expression (Rollin, 2014, 2020), the teleonomic perspective suggests that such interventions should be approached with extreme caution. This is because teleonomic traits exist within deeply integrated networks of physiological, neural, behavioural, and ecological systems, and altering one component may have cascading effects that are impossible to predict or detect until significant harm has occurred. What appears to be a targeted modification may compromise adaptive capabilities in ways that only become apparent over extended periods or across multiple generations. In production systems characterised by rapid turnover from birth to culling, these delayed effects are easily obscured. Such systems have created epistemic blind spots that favour short-term productivity while masking the gradual erosion of genetic and experiential resilience. Over time, animals may become increasingly dependent on human management to survive, generating self-reinforcing cycles of intervention that ultimately threaten both welfare and economic sustainability.

Here, it is crucial to clarify that recognising a capability or behaviour as teleonomic does not imply that enabling its expression will necessarily improve welfare, nor that such expression ought to be facilitated in managed contexts. Many teleonomic traits – including agonistic competition, reproductive effort, antipredator responses, illness behaviour, risk-taking, and persistence under extreme physiological challenge – are evolutionarily conserved precisely because they contributed to reproductive success or survival under ancestral conditions, even when they imposed substantial physiological stress, injury, or negative affect at the level of the individual. From a teleonomic perspective, these traits remain biologically relevant, but their expression may conflict with welfare aims in contemporary or human-managed environments. This tension does not undermine the relevance of teleonomy to welfare assessment; rather, it sharpens it. Welfare does not track whether teleonomic capabilities are exercised and fulfilled per se, but whether the motivational and regulatory systems that give rise to them can operate coherently within the conditions provided. In environments that exaggerate, constrain, or distort evolved contingencies – such as captivity, intensive production, or urban free-ranging contexts – allowing unrestricted expression of certain teleonomic behaviours (e.g., resource-guarding, coercive mating, predation, or terminal persistence) may predictably generate harm to the individual or others.

A teleonomic approach therefore requires careful discrimination between recognising a behaviour as part of an animal’s evolved organisation and judging whether enabling its expression under specific conditions supports or undermines welfare. This distinction is especially salient under artificial selection, where selective amplification of particular traits (e.g., fecundity, milk yield, growth rate, or appetite) can disrupt the balance of integrated teleonomic systems, rendering animals increasingly dependent on human intervention. Acknowledging these conflicts does not romanticise “natural” lives or reject domestication. Rather, it foregrounds the ethical responsibility to evaluate how far evolved systems can be pushed before regulatory coherence, resilience, and lived viability are compromised.

These considerations suggest that genetic modifications which impair teleonomic (viability-oriented) systems would require extraordinary justification. A teleonomic framing does not prescribe simple answers to ethical questions, but establishes a biologically grounded basis for evaluating how far interventions into evolved systems can be ethically justified.