Learning (zoology)

Learning, as defined by ethologists, is simply any change or modification in behavior that is directed by previous experience and involves the nervous system but cannot be attributed to the effects of development, maturation, fatigue, or injury. These latter phenomena contribute to changes in behavior that generally do not constitute learning. Learning takes many forms, including habituation, sensitization, associative learning, perceptual or programmed learning, and insight. Each type has its own basic characteristics and adaptive significance. Habituation and sensitization are the simplest and most widespread forms of learning, and insight is the most complex and least understood form. Insight involves the ability to put two previous experiences together to solve an unrelated problem.

Habituation and Sensitization

Habituation and sensitization, forms of simple nonassociative learning, are considered the simplest forms of learning. Habituation involves a decrease in a behavioral response that results from repeated presentation of a stimulus. A young, naïve duck, for example, will exhibit an innate startle response when any hawk-shaped object is passed overhead. With the repeated presentation of the hawk model, however, the intensity of the bird’s reflex declines as the animal becomes habituated or learns that the stimulus has no immediate significance. Habituation learning is common throughout the animal kingdom, occurring in animals as complex as humans and as simple as single-celled protozoans. It has tremendous adaptive significance in that it prevents repeated responses to irrelevant stimuli that could otherwise overwhelm the animal’s senses and prevent it from accomplishing other critical tasks. One of the common characteristics of habituation is that after a short period, usually defined by the particular species and the stimulus in question, the animal will completely recover from the habituation experience and will again exhibit a full response to the stimulus. This, too, has important survival implications, especially for species that rely on stereotypic alarm responses to avoid predation.

In contrast to habituation, sensitization is the increase in intensity of a response that results from the repeated presentation of a stimulus. A good example is the heightened sensitivity to even relatively soft sounds that results from the initial presentation of a loud, startling noise, such as a gunshot. Sensitization differs from habituation in important ways. First, the specific stimulus that elicits a sensitization response is different from the stimulus to which the animal becomes sensitized. Second, the underlying physiological mechanisms that control these two processes are fundamentally different.

The third and broadest category of learning is associative learning. In this type of learning, an animal makes a connection between some primary environmental stimulus (that involves either a reward or punishment) and a novel or neutral stimulus that is paired with the first stimulus.

Classical and Operant Conditioning

The simplest form of associative learning is classical conditioning, first studied by Ivan Pavlov. Pavlov observed that when a dog is presented with food, the dog will begin to salivate. He referred to the food, in this case, as an unconditioned stimulus (US) and to the salivation reflex as the unconditioned response (UCR). When the unconditioned response is effectively paired with a second, novel stimulus, such as a light or bell (called the conditioned stimulus, or CS), the dog will, after several trials, associate this second stimulus (CS) with the US and begin to salivate whenever the CS is presented. The salivation reflex that occurs following the presentation of the CS is termed the conditioned response (CR).

Although classically conditioned learning is often associated with the controlled experiments of psychologists, it undoubtedly occurs throughout the animal kingdom, and it may be one of the most common ways by which animals learn about their immediate environment. A good example of this is the phenomenon of taste-aversion learning. Taste-aversion learning occurs when an animal associates a specific odor or visual stimulus with an unpleasant experience resulting from the consumption of an unpalatable or poisonous food item. After even a single experience with the distasteful food, the animal will subsequently avoid ingestion, even if it means starvation. Taste-aversion learning is especially important for nonspecialist feeders that forage on a variety of foods and must periodically sample unfamiliar food items. This learning phenomenon also serves as the basis for the evolution of many warning signals in animals.

Through natural selection, many distasteful prey have evolved distinctive marks, colorations, odors, or behavioral characteristics that serve as a reminder (a CS) to predators that it is distasteful or harmful. After one negative experience with this prey, the predator learns to associate these characteristics with the sight or smell of the animal. Such characteristics have obvious survival benefits for the prey. Taste-aversion learning differs from classical conditioning in that the critical time between the CS and the US is usually much longer and that only one trial is necessary for the learning to occur. This latter effect has important implications for animals that rely on aversion learning to avoid poisoning. Many animals, such as bluejays and rats, wait a specific length of time after ingesting a novel food item to determine whether they will become ill.

A second major form of associative learning is operant, or instrumental, conditioning. Unlike classic conditioning, in which the animal is passively involved in the learning experience, in operant conditioning, the animal learns by manipulating some part of its environment. In traditional operant learning experiments, the animal, for example, presses a lever or rings a bell to receive some reward. Because this kind of learning usually improves with practice, it is often referred to as trial-and-error learning. This kind of learning has obvious adaptive significance under natural conditions. Perhaps the best example of this is the reinforced trial-and-error learning that is necessary for many young vertebrates to perfect their feeding techniques. The naïve young of many mammals and birds greatly enhance their feeding efficiency when repeatedly allowed to manipulate their food. Similarly, many animals, including insects, use this reinforced practice to learn their way around in their habitat, home range, or territory, much the way a rat learns its way around in a maze.

Programmed Learning

In addition to associative learning, there are several types of learning that seem to involve mechanisms that aremore complex than simple associations. The most common examples include song learning (in birds) and imprinting. Ethologists often refer to these types of learning as programmed learning since they only take place at certain times and under very restricted circumstances.

Imprinting is the process whereby a young animal develops a behavioral attachment to some other animal or object. Animals have been observed to imprint naturally on their parents, individuals of the opposite sex, food items, preferred habitats, and home streams (in the case of salmon). All such types of imprinting have two general features in common. First, the imprinting must occur during some critical period. The most familiar type of critical period is that which occurs in parental imprinting, a specific imprinting routine whereby a newborn becomes behaviorally fixed on a parent. First described by Konrad Lorenz, this type of learning requires a critical period shortly after birth, in which the young learn to recognize and follow the parent. Outside this period, learning simply cannot occur. The second characteristic common to all types of imprinting is that the young animal must be actively involved in the learning process. In fact, the strength of the imprinting seems to depend largely on the degree of this involvement.

Song learning in birds is fundamentally quite similar to imprinting in that it, too, requires a specific learning period. White-crowned sparrows, for example, learn their song from their fathers, usually from one to six weeks after birth. During this critical period, these young birds learn to imitate the song that is specific to their species as well as the variations and dialect characteristics of their population. When young birds are raised in isolation and prevented from hearing their own species’ song, they develop an abnormal vocalization. If given the opportunity to hear a recording of a normal adult song of its species during the critical learning period, the young bird will learn to sing normally. If, on the other hand, the animal is exposed to the song of some other closely related species, the animal will not develop a normal song. This suggests that birds are, somehow, innately programmed to learn their species-specific songs. Thus, it seems that both imprinting and song learning are, in many ways, quite similar and may be controlled by the same underlying mechanisms. The tendency to classify these as complex behaviors, however, may be attributable to ethologists’ lack of understanding of these mechanisms.

Insight

Perhaps the most advanced and least understood form of learning is insight. Insight is said to differ from other forms of learning in that it is characterized by a modification in behavior that is not contingent on some particular recent experience. Instead, insight behavior involves the ability to put two independent ideas together to solve a third, unrelated problem. Wolfgang Köhler’s classic observations on learning in chimpanzees illustrate the phenomenon of insight. He observed that when a preferred food item (such as a banana) was placed out of reach of a caged chimpanzee, the animal quickly learned to use a pole as an extension of its arm to pull in the food; when the food was hung overhead, the animal would learn to stack boxes to reach the food. Examples of tools used by chimpanzees observed under natural field conditions include using sticks as probes for gathering insects, as spoons to fish ants and termites from their homes, and using small branches to ward off potential predators. They have also been observed using stones to crush open nut shells and using branches and leaves to stay dry when it rains. Some chimpanzees give their tools to younger chimpanzees in their group, presumably encouraging them to learn the skill.

Although this type of problem-solving behavior seems fundamentally more complex than any other type of learning, it has been suggested that many of the specific behaviors cited as examples of insight may be nothing more than extensions of associative learning. Pigeons, for example, can be conditioned to perform certain activities that they use later in solving more complex problems. A pigeon conditioned at one time to push a box across its cage floor and at another time to climb on a box and peck at a food lever will later push and position a box under a lever so that it can peck at the lever and receive a reward. While this seems to reflect some type of problem-solving ability, it is interesting that birds that are not previously conditioned cannot solve the problem. Thus, insight may build on some form of associative learning.

Insight learning has also been invoked to explain the origin of many types of cultural learning. Cultural learning occurs when one animal in a group discovers a unique or novel behavior, and the other members learn to copy the behavior through the process of observational learning. One of the classic examples of this kind of learning was observed in the blue tit, a small European bird that was observed to strip the caps off milk bottles to drink the cream that surfaced at the top. In relatively little time, the behavior spread and was exhibited by this species across Western Europe. Although there is little doubt that such cultural transmission involves nothing more than the simple imitation of another animal’s behavior, it is not clear whether the origin of such behaviors reflects some form of innovation.

Ethological and Psychological Approaches

The study of behavior and learning has long been characterized by two very different methodological and philosophical approaches: those of ethology and psychology.

Ethology, the study of animal behavior, is built on several very specific assumptions and principles that clearly distinguish it from the field of psychology. First, the study of ethology involves objective, nonanthropomorphic (that is, not biased by human expectations or interpretations) descriptions and experiments of the learning process within a natural context. Konrad Lorenz, one of the founders of the field, insisted that the only way to study behavior and learning was to make objective observations under completely natural field conditions. Building on Lorenz’s purely descriptive approach, Nikolaas Tinbergen conducted rigorous field experiments, similar to those that now characterize modern ethology. The classic work of early ethologists helped demonstrate how an animal’s sensory limitations and capabilities can shape its ability to learn. For example, in a series of classic learning experiments, Karl von Frisch convincingly documented the unusual visual capabilities of the honeybee. He first trained honeybees to forage at small glass dishes of sugar water and then, by attaching different visual cues to each dish, provided the animals with an opportunity to learn where to forage through the simple process of association. From these elegant (but simplistic) experiments, von Frisch found that bees locate and remember foraging sites by the use of specific colors, ultraviolet cues, and polarized light, a discovery that revolutionized how scientists view the sensory capabilities of animals.

A second important feature of ethology is that it is built on the assumption that learning depends not only on environmental experience but also on a variety of underlying physiological, developmental, and genetic factors. The work of countless neurobiologists, for example, clearly demonstrates how behavioral changes are linked to modifications in the function of nerves and neuronal pathways. By observing the response of individual nerves, neurobiologists can observe changes that occur in the nerves when an animal modifies its behavior in response to some stimulus. In a similar way, they can show how learning and behavior are affected when specific nerve fibers are experimentally cut or removed. However, neurobiologists’ understanding of the physiological control of learning is limited to simpler kinds of learning, such as habituation and sensitization.

Like the neurobiologists, behavioral geneticists have shown that much of learning, and behavior in general, is intimately tied to internal mechanisms. The results of hybridization experiments and artificial breeding programs clearly demonstrate a strong genetic influence on learned behaviors. In fact, it has been well documented that many animals (including both invertebrates and vertebrates) are genetically programmed (or at least have a genetic predisposition) to learn only specific kinds of behaviors. Finally, the most important characteristic of ethology is that it places tremendous importance on the evolutionary history of an organism. It assumes that an animal’s ability to learn is shaped largely by its evolutionary background, and it emphasizes the adaptive significance of the various types of learning.

In comparison with ethology, the field of psychology emphasizes the importance of rigorously controlled laboratory experiments in the study of learning. The most widely used methods in this field are those of classic and operant conditioning. The primary objective in these approaches is to eliminate and control as many variables as possible and thereby remove any doubt as to the factors responsible for the behavioral changes. These approaches have met with considerable success at identifying specific external mechanisms responsible for learning. These techniques, however, tend to focus only on the input (stimulus) and output (response) of an experiment and, as a result, de-emphasize the importance of proximate mechanisms, such as physiology and genetics. In addition, these approaches generally ignore the evolutionary considerations that ethologists consider so fundamental to the study of behavior.

Understanding the Learning Process

Although the approaches used to study learning vary tremendously, nearly all such studies are directed at two goals: to understand the adaptive value of learning in the animal kingdom and to understand the physiological, genetic, and psychological mechanisms that control learning. For any animal, the adaptive advantages of learning result primarily from the increase in behavioral plasticity that learning provides. This plasticity (ability to be flexible) provides the animal with a greater repertoire of responses to a given stimulus and thereby increases the chances that the animal will survive, reproduce, and pass the genes that control the learning process on to the next generation. In comparison, the value of an innate behavior lies primarily in its ability to provide a nearly stereotypic response to a stimulus on the very first occasion on which it is encountered. Innate reflexes are especially important in situations in which there may not be a second chance for the animal to learn an appropriate response. The best examples are basic feeding responses (for example, the sucking reflex in newborn mammals) and predator-escape behaviors (alarm calls in young birds). It is a common misconception, however, that a learned behavior is attributable entirely to the animal’s environment, whereas instinct is completely controlled by the genes. Many studies have demonstrated that numerous animals are genetically programmed to learn only certain behaviors. In contrast, it has been shown that instinct need not be completely fixed but can be modified with experience. Thus, learning and instinct should not be considered two mutually exclusive events.

Additionally, the study of learning has provided considerable insight into the internal mechanisms that control and regulate behavior. These mechanisms are the cellular and physiological factors providing the hardware that facilitates learning. As neurobiologists and geneticists learn more about these types of control, it is increasingly evident that learning at nearly all levels may involve the same basic mechanisms and processes—the only difference between simple and complex behaviors may be the extent to which the learning is physically constrained by the biology of the animal. Thus, many invertebrates, by virtue of their simple body plan and specific sensory capabilities, are limited to simple learning experiences. Vertebrates, on the other hand, live longer and are not as rigorously programmed for specific kinds of behavior.

Research continues to provide a better understanding of learning and animal cognition in the twenty-first century. Worldwide, scientists have performed studies showing evidence of a per-verbal stage in finches, problem-solving and teaching among bumblebees working together to solve puzzles, and chimpanzees learning how to use tools in more complex ways. Researchers trained crows to count, observed otters using tools, and documented a wild orangutan applying the leaves of a medicinal plant to its wounds. Even animals historically considered intellectually simple, like rats and chickens, have been observed exhibiting learned behaviors. The increased scientific focus on understanding the learning processes of the animal kingdom improves scientists’ abilities to protect and preserve wildlife, but it also provides a foundation for understanding human brains, Earth’s ecosystems, and the interactions between humans, animals, and our habitats.

Principal Terms

Adaptation: Any heritable characteristic that increases the probability that an animal will survive and reproduce in its natural environment

Conditioning: The behavioral association that results from the reinforcement of a response with a stimulus

Innate: Any inborn characteristic or behavior that is determined and controlled largely by the genes

Instinct: Any behavior that is completely functional the first time it is performed

Natural Selection: The process of differential survival and reproduction that leads to heritable characteristics that are best suited for a particular environment

Stimulus: Any environmental cue that is detected by a sensory receptor and can potentially modify an animal’s behavior

Bibliography

Alcock, John. Animal Behavior: An Evolutionary Approach. 12th ed. Sunderland, Sinauer Associates, 2023.

Beran, Michael J. “Animal Learning and Cognition.” Oxford Research Encyclopedia of Psychology, 2020.

Bonner, John T. The Evolution of Culture in Animals. Princeton, Princeton University Press, 2018.

Donahoe, John W., et al. Learning and Complex Behavior. Richmond, Ledgetop Publishing, 2004.

Gould, James, L. Ethology: The Mechanisms and Evolution of Behavior. New York City, W. W. Norton, 1982.

Grier, James W. Biology of Animal Behavior. 3rd ed. New York City, McGraw-Hill, 1999.

Hickman, Cleveland P., et al. Integrated Principles of Zoology. 19th ed. New York City, McGraw Hill, 2023.

Kluger, Jeffrey. "What Animal Studies Are Revealing About Their Minds—and Ours." Time, 5 June 2024, time.com/6985448/animals-human-behavior-research. Accessed 15 Sept. 2024.

Melfi, Vicky A., et al. Zoo Animal Learning and Training. Hoboken, John Wiley & Sons, 2020.

Safina, Carl. "How do Animals Learn How to Be, Well, Animals? Through a Shared Culture." Ted Ideas, 21 May 2020, ideas.ted.com/how-do-animals-learn-how-to-be-well-animals-through-a-shared-culture. Accessed 10 July 2023.

Zentall, Thomas R. The Oxford Handbook of Comparative Cognition. Rev. ed. Oxford, OxfordUniversity Press., 2012.