Animal Behavior

Abstract

Animal behavior is a form of behavior: the observable activity of non-human animals, studied as a window onto the mechanisms, development, function, and evolution of action. The scientific study of animal behavior, ethology, matured when Niko Tinbergen set out four complementary questions that any behavior can be asked of, separating the immediate causes of an act from its adaptive value and evolutionary history. This article covers that organizing framework, the classical ethological concepts of the fixed action pattern and the sign stimulus, the shift from instinct toward comparative cognition, and the economic analysis of foraging. A worked example applies the marginal value theorem to patch-leaving decisions, and three interactive demonstrations let the reader sort behaviors into Tinbergen's four questions, release a fixed action pattern, and find the optimal time to leave a depleting patch.

Keywords: animal behavior, ethology, comparative cognition, Tinbergen's four questions, optimal foraging

The study of animal behavior sits at the junction of biology and psychology. It asks what animals do, how the doing is controlled from moment to moment, how it develops over a lifetime, and why it exists at all, questions that range from the firing of a neuron to the pressure of natural selection across millions of years. For most of the twentieth century these questions were pursued by two traditions that rarely spoke: a European ethology built on patient observation of animals in their natural surroundings, and an American comparative psychology built on controlled laboratory experiments on learning.

This article treats animal behavior as the unified field those traditions eventually became. It begins with the framework that made the unification possible, Tinbergen's four questions, then follows the subject from its classical ethological core through its modern expansion into comparative cognition, the experimental study of how animal minds represent and solve problems. Throughout, the aim is to show that a behavior is explained not by a single cause but by a layered set of answers, each legitimate, each addressing a different sense of the word why.

What Animal Behavior Is

Animal behavior is the observable activity of an organism, the coordinated movement, signalling, and response through which an animal interacts with its environment and with other animals. As a field of study it is the scientific investigation of that activity across every level of explanation, from the sensory and neural mechanisms that generate an action to the selective pressures that shaped it over evolutionary time. The Medical Subject Headings vocabulary classifies it as a form of behavior, the broad category of observable activity, and places it within the hierarchy of psychological and behavioral phenomena because the questions it raises, about perception, learning, memory, and decision, are continuous with those of psychology.

The modern field grew out of ethology, the biological study of behavior under natural conditions, whose founders, Konrad Lorenz, Karl von Frisch, and Niko Tinbergen, shared the 1973 Nobel Prize in Physiology or Medicine for establishing that behavior could be studied with the same rigor as anatomy or physiology (Tinbergen, 1963). Their central insight was methodological: behavior is a biological character, as much a product of evolution as a bone or an organ, and it can be described, compared across species, and explained at several distinct levels at once. That insight, and the discipline it created, is the subject of this article.

Types of Animal Behavior

The Medical Subject Headings vocabulary indexes animal behavior as a parent descriptor with a set of narrower descriptors directly beneath it. These narrower kinds partition the subject not by a single principle but by several overlapping ones, some naming a functional category of action (feeding, elimination, tool use), some a mode of spatial movement (homing, kinesis, taxis), and some a defensive state (tonic immobility, cataleptic freezing). The categories are therefore not mutually exclusive: a single episode of behavior may be appetitive, directed by a taxis, and part of a feeding sequence at once. They are an indexing classification, a controlled vocabulary for retrieving literature, rather than a theory of how behavior is organized, and they should be read as such.

The twelve narrower descriptors MeSH files directly under animal behavior are:

- Animal Communication — the transmission of information between animals through signals, as in alarm calls or the honeybee waggle dance. - Animal Distribution — the spatial arrangement of animals across a habitat or range. - Appetitive Behavior — the searching, approach phase of a motivated sequence that precedes the consummatory act. - Consummatory Behavior — the terminal, goal-completing phase of a motivated sequence, such as the act of eating itself. - Eliminative Behavior, Animal — the patterned behaviors surrounding excretion, often carrying a scent-marking function. - Feeding Behavior — the acquisition and ingestion of food, including foraging, prey capture, and diet selection. - Freezing Reaction, Cataleptic — a sustained, posture-holding immobility seen in some defensive and pharmacological states. - Homing Behavior — the capacity to return to a home site or territory after displacement, as in migratory and navigating species. - Immobility Response, Tonic — a reflexive state of motor inhibition, sometimes called animal hypnosis, elicited by restraint or threat. - Kinesis — an undirected change in the rate or form of movement in response to stimulus intensity. - Taxis Response — a directed movement oriented toward or away from a stimulus such as light or a chemical gradient. - Tool Use Behavior — the manipulation of an external object to achieve a goal, documented in primates, corvids, and other taxa.

None of these narrower descriptors is yet a separate article on this site; they are listed here to show how the parent topic is subdivided in the controlled vocabulary. The parent descriptor directly above animal behavior in the hierarchy is behavior itself.

Tinbergen's Four Questions

The single most important organizing idea in the study of animal behavior is that a complete explanation requires answering four distinct questions, which Tinbergen set out in 1963 in a paper that defined the aims and methods of ethology (Tinbergen, 1963). Faced with any behavior, one can ask about its causation (what internal and external stimuli trigger it, and through what physiological machinery), its ontogeny (how it develops over the individual's lifetime, and what mix of genes and experience shapes it), its survival value (what it does for the animal's reproductive success), and its evolution (what phylogenetic history produced it). The four are not competitors: a song is explained at once by the hormones and neural circuits that produce it, by the learning through which the young bird acquired it, by the mates it attracts, and by the ancestral signals from which it descended.

Tinbergen's deeper contribution was to separate two pairs of questions that are easily confused. Causation and ontogeny are proximate questions, about the mechanism and development of a behavior in the individual. Survival value and evolution are ultimate questions, about the adaptive function and phylogenetic origin of the behavior in the lineage. A proximate answer and an ultimate answer never conflict, because they address different senses of why; the error the framework was built to prevent is treating an answer at one level as if it refuted or replaced an answer at another. The framework remains the standard scaffolding of the field half a century on, and has been revisited and refined rather than displaced (Bateson & Laland, 2013). The first demonstration presents a series of behaviors and asks the reader to sort each explanation into the correct one of Tinbergen's four questions.

Instinct and the Fixed Action Pattern

Classical ethology built its science on the concept of instinct, and its central unit was the fixed action pattern: a stereotyped, species-typical sequence of movements that, once triggered, runs to completion in much the same form every time. Lorenz and Tinbergen showed that such patterns are released by a specific feature of the environment, the sign stimulus or releaser, rather than by the whole object. A greylag goose retrieving an egg that has rolled from the nest performs the same rolling motion even if the egg is removed mid-action; a male stickleback attacks the red belly of a rival, and will attack a crude model with a red underside more vigorously than an accurate model without one. The behavior is keyed to the sign, not to the realistic whole.

This selectivity gives rise to the supernormal stimulus: an exaggerated version of a sign stimulus that elicits a stronger response than the natural object ever would. A bird will preferentially try to incubate an artificially enlarged egg over its own, because the releasing mechanism responds to a dimension, size, that nature never pushed to that extreme. Lorenz also described imprinting, the rapid and largely irreversible attachment a young animal forms to the first suitable moving object it encounters during a sensitive period, which binds a flexible learning process to a narrow developmental window. These concepts remain foundational even as the strict innate-versus-learned dichotomy that framed them has softened into the modern understanding that development interweaves the two. The second demonstration lets the reader vary a sign stimulus and watch a modeled fixed action pattern fire once a releasing threshold is crossed, including the supernormal case.

Learning and Comparative Cognition

The instinct framework could not contain everything animals do. Much behavior is acquired, and the American tradition of comparative psychology had long studied how animals learn through association and consequence. Over the second half of the twentieth century the two traditions merged into comparative cognition, the experimental study of how animals perceive, represent, remember, and reason, synthesized by Sara Shettleworth into a field that insists cognitive claims about animals be grounded in their ecology and evolutionary function (Shettleworth, 2010). The questions moved beyond whether an animal can learn to what it represents: space, time, number, causal structure, and the behavior of other minds.

The evidence that animals possess rich cognition is now substantial. Nicola Clayton and Anthony Dickinson showed that western scrub-jays remember what food they cached, where, and how long ago, recovering perishable items before they rot, the first behavioral evidence of episodic-like memory in a non-human (Clayton & Dickinson, 1998), and later that the same birds cache for anticipated future need, evidence of planning that is not driven by current motivation (Raby et al., 2007). Corvids rival great apes on many tests despite a brain organized without a layered cortex: New Caledonian crows manufacture and use tools, including solving multi-step metatool problems (Taylor et al., 2007), ravens plan for tool use and bartering across delays in a manner parallel to apes (Kabadayi & Osvath, 2017), and the convergent evolution of this intelligence in such distant lineages is itself a major finding (Emery & Clayton, 2004). Vervet monkeys give acoustically distinct alarm calls for different predators, each eliciting the appropriate escape response, an early demonstration of functionally referential communication (Seyfarth et al., 1980). Frans de Waal has argued that the recurring tendency to underestimate these capacities reflects the limits of our tests rather than the limits of animal minds (de Waal, 2016).

The deepest interpretive question in this literature is where to draw the line between a flexible cognitive process and a simpler associative or perceptual one, a line the field polices with controlled experiments precisely because the intuitive reading of a clever-looking behavior is so often wrong. This caution is the modern descendant of Morgan's Canon, the principle that no behavior should be interpreted as the outcome of a higher psychological faculty if it can be explained by one standing lower on the psychological scale (Shettleworth, 2010). That caution is the legacy of the long argument over whether animal behavior should be explained by mind at all.

Optimal Foraging and the Economics of Behavior

If ultimate questions ask what a behavior does for reproductive success, then behavior should, on average, be shaped toward solutions that maximize fitness, and this makes it amenable to the mathematics of optimization. Behavioral ecology treats an animal as a decision-maker allocating limited time and energy, and optimal foraging theory is its most developed branch. A foraging animal that exploits a patch of food, a bush of berries, a flower bed, a leaf-litter patch, faces a universal problem: the patch depletes as it is worked, so the rate of gain falls the longer the animal stays, yet moving to a fresh patch costs travel time during which nothing is gained. The animal must decide when to leave.

The marginal value theorem, introduced by Eric Charnov, gives the optimal rule: leave the current patch when the instantaneous rate of gain within it drops to the average rate of gain available in the environment as a whole. An animal that stays past that point is wasting time on a patch that now pays worse than the average alternative; one that leaves too early forgoes gain it could still have taken cheaply. The optimum balances the two, and it shifts with travel time: when patches are far apart, the long trip between them makes it worth staying longer and depleting each patch further. The worked example below derives the optimal leaving time for a concrete gain curve, and the third demonstration lets the reader move the travel time and watch the optimal patch-residence time change.

Worked Example: The Marginal Value Theorem

Consider an animal foraging in patches whose yield accumulates with diminishing returns. Let the cumulative food gained after spending time t in a patch be g(t) = 30 × (1 − e^−0.5t), a curve that rises steeply at first and then flattens as the patch depletes. Let T be the average travel time between patches. The long-term intake rate for a strategy of spending t in each patch is the gain per patch divided by the total time per patch, R(t) = g(t) / (T + t).

The rule. The marginal value theorem says the optimum t\ is where the within-patch marginal rate g′(t) equals the long-term rate R(t\). Table 1 evaluates R(t) for a travel time of T = 4, to locate the leaving time that maximizes intake.

The intake rate rises to a maximum of about 3.33 near t = 3 and declines on either side, so the optimal strategy is to leave each patch after roughly three time units. A precise solution of g′(t\) = R(t\) gives t\ ≈ 3.0 with a maximized rate R\ ≈ 3.33, confirming the tabulated peak. The theorem's signature prediction is what happens when travel time changes: raising T from 4 to 16 lowers the best achievable rate from 3.33 to about 1.31 and pushes the optimal leaving time out from 3.0 to about 4.9, because when the next patch is far away it pays to deplete the current one more thoroughly before making the costly trip. Lowering T to 1 raises the achievable rate to about 6.36 and pulls the optimal leaving time in to about 1.7. The animal does not compute this; selection shapes behavior as if it had, which is exactly the sense in which an ultimate, functional analysis explains an animal's choices without attributing the arithmetic to the animal.

Discussion

The study of animal behavior is held together by Tinbergen's insistence that explanation is layered. The classical ethologists supplied a powerful account of the proximate machinery of innate behavior, the releaser, the fixed action pattern, the sensitive period; behavioral ecology supplied the ultimate, functional account in the language of optimization; and comparative cognition reopened the question the behaviorists had tried to close, of what animals represent and compute. These are not three competing theories of behavior but three of Tinbergen's four questions pursued with their own methods, and the health of the field depends on keeping them distinct while letting them inform one another.

The through-line of the modern discipline is a disciplined humility about interpretation. A behavior that looks like foresight may be a cached association; a behavior that looks like blind instinct may be finely tuned by early experience; a rate-maximizing forager computes nothing. The field's characteristic move, from Tinbergen's four questions to the controlled cognitive experiment to the foraging model, is to resist the single intuitive story and insist on specifying which question is being answered and at what level. That discipline is what turned the watching of animals into a science.

Cognitive Implications

For cognitive psychology, animal behavior is both a mirror and a testing ground. It is a mirror because the capacities the field documents, spatial memory in a food-caching bird, numerical and temporal representation, the tracking of others' knowledge, are the same capacities cognitive psychology studies in humans, and the comparative method reveals which are widespread and ancient and which, if any, are distinctively human (Laland & Seed, 2021). An ability found across distant lineages is likely a basic solution to a recurring problem; one found only in humans marks a genuine discontinuity worth explaining.

It is a testing ground because animal work can dissociate cognition from language and culture, isolating the computational core of a capacity from the human-specific scaffolding that normally surrounds it. The finding that corvids achieve ape-like cognition with a brain lacking a layered cortex is a direct challenge to any theory that ties a cognitive function too tightly to a particular piece of neural hardware, and shows that the same computation can be implemented by different architectures (Güntürkün & Bugnyar, 2016). Studying how minds unlike ours solve problems like ours is one of the strongest available constraints on what a theory of cognition must explain.

Current Directions

One active frontier is the neuroscience of non-mammalian cognition. The discovery that the avian pallium supports flexible, ape-like problem-solving without the mammalian six-layer cortex has driven a search for the common computational principles that intelligence requires, independent of the specific tissue that implements them (Güntürkün & Bugnyar, 2016). That search reached a milestone when single-neuron recordings in crows identified an activity signature that tracks the birds' subjective perceptual reports, offering a neural correlate of sensory consciousness in an animal whose brain is organized nothing like our own (Nieder et al., 2020). The question of animal consciousness, once dismissed as unscientific when Donald Griffin raised it, has become an empirical research program (Griffin, 2001).

A second frontier returns to the comparative question of human uniqueness with better tools. Rather than asking whether animals are intelligent, researchers now ask which specific components of human cognition, cumulative culture, flexible planning, recursive social reasoning, are shared and which are derived, using matched tasks across species to locate the differences precisely (Laland & Seed, 2021). The payoff is a sharper account of the few genuine discontinuities against a background of deep continuity, replacing the old binary of human versus animal with a map of which capacities evolved when.

Common Misconceptions

A behavior has one true explanation.

No. Tinbergen's four questions show that causation, development, survival value, and evolution are complementary levels of explanation; an answer about mechanism does not compete with an answer about adaptive function, and treating them as rivals is the field's most common reasoning error (Tinbergen, 1963; Bateson & Laland, 2013).

Instinct and learning are opposites, and a behavior is one or the other.

Not as a clean dichotomy. Even the classic fixed action patterns develop through an interaction of inherited structure and experience, and imprinting is precisely a learning process bound to an innately specified sensitive window, so development interweaves the two rather than assigning a behavior to a single origin (Bateson & Laland, 2013).

Clever-looking animal behavior proves human-like reasoning.

Not on its own. The field's whole method is to test whether an apparently insightful behavior reflects flexible cognition or a simpler associative or perceptual process, because the intuitive reading is so often wrong; cognitive claims are accepted only when controlled experiments rule out the leaner explanation (Shettleworth, 2010).

Intelligence requires a mammalian cortex.

No. Corvids match great apes on many cognitive tasks with a brain that has no layered neocortex, showing that the same cognitive computations can be implemented by very different neural architectures (Emery & Clayton, 2004; Güntürkün & Bugnyar, 2016).

Glossary

Animal behavior.

The observable activity of non-human animals, studied across the mechanistic, developmental, functional, and evolutionary levels of explanation.

Cognitive ethology.

The study of the mental experiences and consciousness of animals in their natural context, a program named and defended by Donald Griffin.

Comparative cognition.

The experimental study of how animals perceive, represent, remember, and reason, grounded in the ecology and evolution of each species.

Episodic-like memory.

Memory for what happened, where, and when, demonstrated behaviorally in food-caching scrub-jays without claiming the subjective recollection that defines human episodic memory.

Ethology.

The biological study of behavior under natural conditions, founded by Lorenz, von Frisch, and Tinbergen.

Fixed action pattern.

A stereotyped, species-typical sequence of movements that, once released by a sign stimulus, runs to completion largely unmodified.

Imprinting.

The rapid, largely irreversible attachment a young animal forms to the first suitable object encountered during a sensitive developmental period.

Instinct.

An inherited disposition toward a species-typical behavior that develops reliably without specific training, the organizing concept of classical ethology.

Marginal value theorem.

The optimal-foraging rule that an animal should leave a depleting patch when its within-patch rate of gain falls to the average rate available in the environment.

Morgan's Canon.

The principle of parsimony in comparative psychology: a behavior should not be read as the product of a higher mental faculty when a simpler process suffices to explain it.

Optimal foraging theory.

The modeling of foraging as an optimization problem in which an animal allocates time and effort to maximize net energetic gain.

Proximate cause.

An explanation of a behavior in terms of its immediate mechanism or its development within the individual, as opposed to its evolutionary function.

Sign stimulus.

A specific environmental feature, also called a releaser, that triggers a fixed action pattern independently of the rest of the object.

Supernormal stimulus.

An exaggerated version of a sign stimulus that elicits a stronger response than the natural stimulus ever would.

Ultimate cause.

An explanation of a behavior in terms of its adaptive survival value or its evolutionary history, as opposed to its immediate mechanism.

Waggle dance.

The symbolic figure-eight movement by which a honeybee communicates the distance and direction of a food source, decoded by Karl von Frisch.

Key Researchers

Nicola S. Clayton

(University of Cambridge, Professor of Comparative Cognition). Comparative psychologist whose scrub-jay experiments provided the first behavioral evidence of episodic-like memory in a non-human and of flexible planning for the future; Fellow of the Royal Society. ORCID · Wikipedia · Wikidata · Faculty

Karl von Frisch

(University of Munich; 1886–1982). Austrian ethologist who decoded the honeybee waggle dance, showing that an invertebrate communicates the distance and direction of food through a symbolic signal; shared the 1973 Nobel Prize. Wikipedia

Donald R. Griffin

(Rockefeller University; 1915–2003). American zoologist who discovered echolocation in bats and later founded cognitive ethology, arguing that animal consciousness is a legitimate object of scientific study. Wikipedia

Alex Kacelnik

(University of Oxford, Emeritus Professor of Behavioural Ecology). Behavioural ecologist whose work bridges evolutionary biology, experimental psychology, and decision theory, including the New Caledonian crow tool studies and the scalar utility theory of risk-sensitive foraging; Fellow of the Royal Society. Wikipedia · Faculty · Scholar

Konrad Lorenz

(Max Planck Institute for Behavioral Physiology; 1903–1989). Austrian zoologist and co-founder of ethology who established the concepts of the fixed action pattern and filial imprinting; shared the 1973 Nobel Prize. Wikipedia

Sara J. Shettleworth

(University of Toronto, Professor Emerita of Psychology and of Ecology & Evolutionary Biology). Experimental psychologist whose synthesis defined the modern field of comparative cognition and insisted that cognitive claims about animals be grounded in adaptive function. Wikipedia · Faculty

Nikolaas Tinbergen

(University of Oxford, Professor of Animal Behaviour; 1907–1988). Dutch ethologist who, with Lorenz and von Frisch, founded modern ethology and framed the four complementary questions that still organize the study of animal behavior; shared the 1973 Nobel Prize. Wikipedia

Frequently Asked Questions

What is animal behavior as a field of study?

It is the scientific investigation of what animals do, pursued across every level of explanation from the neural mechanism of an action to the evolutionary pressure that shaped it. It grew out of ethology, the biological study of behavior in natural conditions, and now includes comparative cognition and behavioral ecology.

What are Tinbergen's four questions?

They are the four complementary questions any behavior can be asked of: its causation (immediate trigger and mechanism), its ontogeny (how it develops in the individual), its survival value (what it does for reproductive success), and its evolution (its phylogenetic history). The first two are proximate questions, the last two ultimate (Tinbergen, 1963).

What is the difference between a proximate and an ultimate cause?

A proximate cause explains the immediate mechanism or development of a behavior; an ultimate cause explains its adaptive function or evolutionary origin. They are different senses of why and never compete, so a mechanistic answer does not refute a functional one (Bateson & Laland, 2013).

What is a fixed action pattern?

A fixed action pattern is a stereotyped, species-typical sequence of movements that runs to completion once triggered by a specific sign stimulus or releaser, such as a goose's egg-rolling motion. It was the central unit of classical ethology's account of instinct.

Are animals genuinely intelligent, or do they just react?

Many animals show flexible cognition that goes well beyond reflex. Scrub-jays remember what they cached, where, and when and plan for future need, and corvids solve multi-step tool problems, though the field accepts such claims only after controlled experiments rule out simpler associative explanations (Clayton & Dickinson, 1998; Raby et al., 2007).

How can birds be so intelligent without a cortex?

Corvids match great apes on many cognitive tasks using the avian pallium, a brain organized without the mammalian six-layer neocortex, which shows that comparable cognition can be implemented by different neural architectures (Emery & Clayton, 2004; Güntürkün & Bugnyar, 2016).

What does optimal foraging theory predict?

It predicts that an animal allocates foraging effort to maximize net energetic gain. The marginal value theorem specifies when to leave a depleting patch: at the point where the within-patch rate of gain drops to the average rate available in the environment, a leaving time that comes later when patches are farther apart.

Can animal communication carry meaning?

It can carry information that functions referentially. Vervet monkeys give acoustically distinct alarm calls for different predators, each prompting the appropriate escape, and honeybees encode the distance and direction of food in the waggle dance, though whether these signals carry meaning in the full human sense remains debated (Seyfarth et al., 1980).

References