Animal Behavior

When a Canada goose rolls a displaced egg back into its nest using a stereotyped head-tucking motion — even if the egg is replaced with a tin can or a volleyball — it is not being foolish. It is executing a fixed action pattern that evolved to solve a problem encountered reliably enough over millions of generations that the nervous system encoded a solution without requiring the goose to figure it out anew each time. When a honeybee performs a waggle dance to communicate the direction and distance of a food source to thousands of nestmates who have never left the hive, it is demonstrating that animal communication can encode symbolic information about locations the communicator has visited but the receiver has not. And when a ground squirrel gives an alarm call that attracts predator attention to itself but benefits the relatives in nearby burrows, it is behaving in a way that natural selection has favored not because it helps the individual survive, but because the genes underlying that behavior are more likely to be present in surviving relatives. Animal behavior — what animals do, how they do it, and why — is the bridge between genetics and ecology, between neuroscience and evolution, between the individual animal and the species-wide patterns that natural selection produces.

Animal Behavior as a Discipline — Ethology, Comparative Psychology, and Behavioral Ecology

Animal behavior — the scientific study of what animals do and why — is one of the oldest questions in natural history but one of the youngest scientific disciplines. Its modern form emerged from three largely independent traditions that converged in the mid-twentieth century: ethology (the European naturalist tradition of studying behavior in natural settings, associated with Konrad Lorenz, Nikolaas Tinbergen, and Karl von Frisch — who shared the Nobel Prize in Physiology or Medicine in 1973), comparative psychology (the North American laboratory tradition associated with Ivan Pavlov, B.F. Skinner, and Edward Thorndike, focusing on learning mechanisms in controlled settings), and sociobiology/behavioral ecology (the evolutionary tradition, crystallised by E.O. Wilson’s 1975 Sociobiology and John Maynard Smith’s game-theoretic approach, analyzing behavior in terms of adaptive function and evolutionary history).

These three traditions asked different questions and favored different methods: ethologists watched animals in the field and asked what behaviors occurred in nature; comparative psychologists ran controlled laboratory experiments and asked what learning rules governed behavior change; behavioral ecologists built mathematical models and tested them with field data, asking whether observed behaviors maximised fitness. Contemporary animal behavior integrates all three approaches, recognising that a complete explanation of any behavior requires multiple levels of analysis — a framework that Tinbergen formalised as his four questions.

Tinbergen’s Four Questions — The Framework for Complete Behavioral Explanation

In a 1963 paper, Nikolaas Tinbergen articulated what has become the most widely used organising framework in animal behavior: four distinct but complementary questions that must all be answered to fully understand any behavior. These questions are not alternatives — they operate at different levels of analysis and are not in competition. A complete explanation of why a male European robin sings in spring requires answers at all four levels, and answers at one level do not replace or contradict answers at another.

Innate Behavior — Fixed Action Patterns, Sign Stimuli, and Orientation Mechanisms

Innate behaviors are behavioral responses that appear in functionally complete form without requiring specific prior experience — they are encoded in the nervous system through evolutionary history and expressed in all normally developing individuals of the species. The concept of “innate” does not mean environmentally independent: all behaviors require some physical environment for development. Rather, innate behaviors are those that develop reliably across the wide range of environments a species normally encounters, without requiring specific learning experiences. They represent evolutionary solutions to recurrent problems that were stable enough over evolutionary time to be reliably encoded in the genome.

Fixed Action Patterns and Sign Stimuli

A fixed action pattern (FAP) is a species-typical, stereotyped behavioral sequence that, once triggered by a specific releasing stimulus, runs to completion regardless of whether the stimulus remains present. FAPs are the classic unit of Lorenz and Tinbergen’s early ethology. The greylag goose egg-rolling behavior is the textbook example: a goose that sees an egg outside its nest will extend its neck, hook the egg with its bill, and roll it back using a combination of directed lateral movements (to keep the egg on course) and a fixed longitudinal component (which continues even if the egg is removed midway, resulting in the bird rolling nothing back to the nest). The longitudinal component is the FAP — it runs without sensory feedback once triggered. The lateral corrections are a taxis (directed component) that requires ongoing sensory input.

SPECIES              FAP                         SIGN STIMULUS (Releaser)

Greylag goose        Egg retrieval               Object outside nest of approx. egg size
Stickleback (male)   Aggressive display/attack   Red underside on male-shaped dummy
Herring gull chick   Begging peck at parent bill  Red dot on long yellow bill (supernormal stim.)
Digger wasp         Prey burial sequence         Stung, paralysed prey item
Male robin           Attack of rival             Red breast patch (even on tuft of red feathers)
Toad                 Tongue strike               Small dark object moving horizontally (prey)
Frog                 Escape response             Large dark object moving from above (predator)
Female silk moth     Orientation upwind          Trace of bombykol (male pheromone)

SUPERNORMAL STIMULI:
  Releasers exaggerated beyond natural range → stronger response than natural stimulus
  Herring gull chicks peck more at artificial bills with larger/brighter red dots
  Oystercatchers prefer giant artificial eggs over their own eggs
  Evolutionary basis: preference for extremes of releasing features has not been
  counter-selected because extremes rarely occur in nature

Orientation Behaviors — Taxes and Kineses

Orientation behaviors are innate mechanisms that direct animals toward or away from stimuli in the environment. A taxis is a directed movement toward (positive) or away from (negative) a stimulus gradient. In a klinotaxis, the animal compares stimulus intensity at successive time points as it moves; in a tropotaxis, the animal simultaneously compares intensity at two spatially separated receptor sites and turns to equalise them. A kinesis is an undirected response in which movement rate or turning frequency changes with stimulus intensity — the animal does not move toward or away from the stimulus directly, but random movements bring it into favourable areas where its movement rate decreases, tending to keep it there. Woodlice (pill bugs) show hygrotaxis — a klinokinesis in which they move more rapidly and turn more frequently in dry air, eventually accumulating in humid microhabitats where movement slows.

Learned Behavior — Habituation, Classical Conditioning, and Operant Conditioning

Learning is a relatively permanent change in behavior that results from experience. It is distinct from fatigue (a temporary performance decrement from repeated stimulation), sensory adaptation (reduced sensory responsiveness), maturation (developmental changes independent of specific experience), and injury. Learning allows animals to modify behavior in response to individual experience, adapting to local conditions, novel environments, and unpredictable resource distributions in ways that genetically fixed responses cannot. The capacity for learning is itself an evolved trait — natural selection has shaped learning rules that make certain associations easy to form and others nearly impossible, tuning learning capacity to the ecological challenges each species faces.

Imprinting and Critical Periods — When the Window for Learning Opens and Closes

Imprinting is a form of learning that occurs during a critical (sensitive) period in early development, is acquired rapidly (sometimes in a single exposure), and produces lasting, often irreversible effects on future behavior. It differs from other learning forms in its time-limited nature, its resistance to extinction once formed, and the fact that it modifies behavioral tendencies rather than specific responses — imprinting does not produce a reflexive reaction but a broad preference or attachment that influences a wide range of subsequent social, sexual, and communicative behaviors.

Social Learning and Animal Culture — Learning from Others Without Individual Trial and Error

Social learning encompasses all mechanisms by which information about the world is acquired via observation of, or interaction with, other individuals — rather than through individual trial-and-error or innate response. It is a fundamentally different process from individual learning: the observer bypasses the costs and risks of personal experience by exploiting the information embedded in others’ behavior. Social learning has been documented across a wide taxonomic range — from invertebrates (bees learn floral preferences by observing experienced foragers) to primates (chimpanzees learn tool-use techniques from group members) — and is now understood as a potentially important route to culturally transmitted behavioral traditions.

Animal Communication — Signals, Channels, and the Evolution of Honest Signalling

Animal communication occurs whenever the behavior of one individual (the sender) modifies the behavior or internal state of another (the receiver) in a way that benefits, on average, the fitness of both parties — the modification being mediated by a signal specifically evolved to transmit information. This definition distinguishes communication from simple environmental effects: a predator’s footstep that a prey animal hears is information, but it is not a signal — the predator did not evolve to produce that sound for communication purposes. Signals are structures, substances, movements, or sounds that evolved primarily to transmit information — they are costly to produce, specific to the channel and receiver, and typically conspicuous in ways adapted to the communication context.

Communication Channels and Signal Types

The Waggle Dance — A Symbolic Communication System

Karl von Frisch’s decoding of honeybee (Apis mellifera) waggle dance communication is one of the great achievements of ethology. Worker bees returning from a profitable food source perform a figure-eight dance on the vertical comb surface of the hive. The straight-run component of the figure eight (the waggle run) encodes two pieces of information simultaneously: its duration (and number of waggles) encodes distance to the food source — approximately 1 second of waggling indicates ~1 km; and its angle relative to vertical encodes the direction of the food source relative to the current sun position — a dance straight up indicates flying directly toward the sun; one 60° to the right indicates flying 60° to the right of the sun. Nestmates following the dance in the dark hive decode this information through antennal contact with the dancer and fly to the indicated location. The dance is symbolic — it represents information about a location the receiver has not visited — and is the most complex non-human communication system known to use symbolic representation of spatial information external to the signaller.

Honest Signalling and the Handicap Principle

For communication to be evolutionarily stable, signals must be honest on average — they must accurately reflect the quality they claim to indicate — otherwise receivers would evolve to ignore them, and the communication system would collapse. Zahavi’s handicap principle explains how honesty can be maintained: signals that are costly to produce can only be afforded by high-quality individuals, making the cost the guarantee of honesty. A peacock’s elaborately costly tail can only be grown by males healthy enough to afford the metabolic cost and predation risk it imposes — females who prefer large-tailed males are therefore reliably choosing high-quality partners. This is the handicap argument: the signal is honest because its cost is prohibitive for low-quality signalers.

Foraging Behavior and Optimal Foraging Theory — How Animals Decide What to Eat and Where

Foraging — the behaviors by which animals locate, identify, select, capture, and consume food — is one of the most studied topics in behavioral ecology because it directly determines energy acquisition and therefore survival, growth, and reproduction. Optimal foraging theory (OFT) applies evolutionary reasoning to foraging behavior: natural selection should favor foraging decisions that maximise the net rate of energy gain per unit time, because energy is the primary currency linking foraging behavior to fitness. OFT does not claim animals consciously optimise — it claims that animals whose foraging behavior approximated the optimum left more descendants, so optimising mechanisms evolved.

Mating Systems and Sexual Selection — How Animals Choose Mates and Why Some Individuals Win

A mating system describes the pattern of mate acquisition in a population — specifically, the number of mates individuals of each sex have, the duration of pair bonds, and the extent of parental investment from each sex. Mating systems are not arbitrary cultural conventions but evolved outcomes shaped by natural selection acting on the fitness consequences of different mating strategies given the ecological and social environment. Understanding why mating systems vary across taxa requires understanding the ecological factors that determine which mating strategies are available and profitable, particularly the distribution of resources, the operational sex ratio, and the degree to which offspring require parental investment from one or both parents.

Sexual Selection — Intrasexual and Intersexual Competition

Charles Darwin identified sexual selection as a distinct evolutionary force — selection acting on traits that affect an individual’s success in obtaining mates rather than in surviving. Sexual selection has two components: intrasexual selection (competition between members of the same sex, usually males, for access to mates — producing weapons like antlers, large body size, and fighting ability) and intersexual selection (mate choice — usually females choosing among males — producing ornaments like peacock tails, elaborate birdsong, and bright coloration). The Fisherian runaway model predicts that female preferences and male ornaments can coevolve to produce exaggerated traits purely through genetic correlation, without the ornament reflecting genuine male quality. The good genes hypothesis (honest advertisement) proposes that ornaments are condition-dependent signals that honestly indicate underlying genetic quality — parasitic load resistance, developmental stability — which females can use to obtain genes for healthy offspring.

Social Behavior — Groups, Dominance Hierarchies, and Cooperation

Many animals spend substantial portions of their lives in social groups — aggregations of conspecifics that interact regularly and whose members influence each other’s behavior, physiology, and fitness. The evolution of sociality requires that the benefits of group living outweigh its costs for at least some group members — a condition that depends on the ecological context, the nature of the benefits (predator dilution, cooperative foraging, shared information, cooperative defence), and the genetic relationships within the group.

Eusociality — The Extreme of Cooperative Social Organisation

Eusocial species show the three defining characteristics of advanced sociality: cooperative brood care, reproductive division of labour (with some individuals — workers — foregoing or reducing their own reproduction to assist others), and overlap of generations in the colony. The eusocial Hymenoptera (ants, bees, wasps) and termites are the most familiar examples, but eusociality has independently evolved in naked mole rats, snapping shrimp, and arguably in some cooperative breeding vertebrates. Haplodiploidy in the Hymenoptera (females are diploid, males haploid) means that sisters share 75% of their genes on average — leading Hamilton to propose that kin selection particularly favoured the evolution of worker behaviour in this group, since a female helping her mother raise sisters gains more genetic representation than she would by reproducing herself.

Altruism, Kin Selection, and Hamilton’s Rule — Why Animals Help Each Other

One of the most debated problems in the early development of behavioral ecology was the evolution of altruistic behavior — behavior that imposes a fitness cost on the actor while benefiting the recipient. If natural selection favors individuals that leave the most offspring, how can self-sacrificing behavior evolve? The answer, provided by W.D. Hamilton in 1964 in two papers that transformed behavioral ecology, is kin selection: natural selection acts on genes, not individuals, and a gene that causes its carrier to help a relative can spread if the relative is likely to share that gene and the fitness benefit to the relative exceeds the cost to the carrier, appropriately weighted by their degree of relatedness.

Reciprocal Altruism — Cooperation Among Non-Relatives

Hamilton’s kin selection explains altruism among relatives, but what explains costly helping among non-relatives? Robert Trivers (1971) proposed reciprocal altruism: costly acts can evolve when individuals interact repeatedly and the helped individual later reciprocates. For reciprocal altruism to be stable, cheating (accepting help without reciprocating) must be detected and punished — the game must be iterated, the same individuals must interact repeatedly, and cheaters must be identifiable. Vampire bats provide the most compelling empirical example: bats that failed to feed on a given night are fed regurgitated blood by roost-mates; the same individuals that feed another are more likely to receive help when they are hungry; relatedness partially but not completely explains the pattern. Tit-for-tat — a strategy of cooperating on the first move and subsequently copying the opponent’s last move — is the evolutionarily stable strategy in iterated prisoner’s dilemma games, consistent with reciprocal altruism theory.

Aggression, Territoriality, and Conflict — The Evolution of Fighting and Its Limits

Aggression — behavior directed by one animal toward another that has the potential to cause harm — is one of the most studied topics in animal behavior, in part because of its apparent paradox: fighting is dangerous and energetically costly, yet widespread. Understanding why aggression occurs and why it is typically constrained requires understanding it in terms of costs, benefits, and alternative strategies — the framework of game theory.

Evolutionary Game Theory and Conflict Resolution

John Maynard Smith and George Price applied game theory to animal conflict in 1973, introducing the concept of the evolutionarily stable strategy (ESS) — a behavioral strategy that, when adopted by most members of a population, cannot be invaded by any rare alternative strategy. In the Hawk-Dove game, Hawk always escalates and Dove always retreats. In a population of all Doves, a Hawk mutant wins every encounter at no cost, so Hawk spreads. In a population of all Hawks, escalated fights are frequent and injuries are common — a Dove mutant avoids injuries by retreating, gaining some benefit. At the ESS, a mixed strategy (playing Hawk some proportion p* of encounters) or a mixture of Hawks and Doves is stable — neither pure strategy can invade the other when the payoff matrix is appropriately parameterised.

Real animals use assessment strategies that are more sophisticated than the simple Hawk-Dove game: the sequential assessment model predicts that contestants initially use low-cost mutual assessment displays (which provide information about relative fighting ability), escalating to higher-cost contact fighting only if the initial assessment is equivocal. This matches the observed structure of many aggressive encounters — a series of display stages with most contests resolved at the display stage, with escalation proportional to the similarity of opponents’ resource-holding potential (fighting ability).

Migration and Navigation — Long-Distance Movement, Compass Systems, and True Navigation

Animal migration — the seasonal, directional mass movement of individuals between geographically separated breeding and non-breeding areas — is one of the most remarkable phenomena in animal behavior. Arctic terns make the longest annual migration of any animal: approximately 90,000 km round-trip from their Arctic breeding grounds to the Antarctic and back, navigating using sun compass, star compass, and magnetic sense. Monarch butterflies fly up to 4,500 km from eastern North America to a small stand of oyamel fir trees in central Mexico — a destination no individual butterfly has ever visited, navigated using a time-compensated sun compass and magnetoreception.

The Neurobiological and Hormonal Basis of Behavior — How the Brain and Endocrine System Produce and Regulate Actions

Every behavior ultimately has a neural basis — the muscles that produce it are activated by motor neurons, which are driven by circuits in the central nervous system, which integrate sensory inputs and internal states to generate motor outputs. Understanding the neural and hormonal mechanisms underlying behavior (Tinbergen’s causation question) is essential both for a complete explanation of specific behaviors and for understanding how evolutionary changes in behavior occur — since evolutionary change in behavior must be implemented through changes in neural and hormonal systems.

Neural Circuits and Behavior — From the Command Neuron to the Brain Network

The simplest neural basis of behavior is the reflex arc — a sensory neuron synapsing directly (or via an interneuron) onto a motor neuron. More complex behaviors involve central pattern generators (CPGs) — networks of neurons that produce rhythmic motor output without requiring sensory feedback for each cycle. CPGs underlie locomotion in almost all animals (walking, swimming, flying), chewing, breathing, and many other rhythmic behaviors. They are extensively modified by sensory feedback and descending signals from higher brain areas, but the basic rhythm is generated intrinsically.

The birdsong system is the most completely understood vertebrate neural circuit for a complex learned behavior. The song control system consists of two pathways: the motor pathway (HVC → RA → motor neurons controlling the syrinx) that directly produces song, and the anterior forebrain pathway (HVC → Area X → DLM → LMAN → RA) required for song learning during development but not for song production once crystallised. This dual-pathway architecture — a direct production pathway and a learning pathway that modifies the production pathway — is remarkably similar to the corticobasal ganglia-thalamo-cortical loops involved in human motor learning, suggesting deep conservation of circuit organisation between birds and mammals despite their independent evolution of vocal learning.

Frequently Asked Questions About Animal Behavior