Tinbergen's Herring Gull Experiments Explained
In the late 1940s, a Dutch ethologist named Niko Tinbergen and his student A. C. Perdeck carried crude cardboard heads, painted models, and thin wooden rods into a herring gull colony on the Dutch coast. They held these objects in front of newly hatched chicks and counted how often the chicks pecked at them. Some of the crudest objects drew more pecks than a carefully made replica of an adult gull. The study, published in 1950 in the journal Behaviour, showed that a chick's first begging behavior is not guided by a rich picture of "parent" but by a small number of visual features, above all the red spot near the tip of the adult's yellow bill. The work became one of the textbook demonstrations of sign stimuli, the simple cues that release innate behavior, and of supernormal stimuli, artificial cues that beat the real thing. It also helped earn Tinbergen a share of the 1973 Nobel Prize in Physiology or Medicine.
The puzzle: why do chicks peck at their parents?
Herring gulls nest in large colonies on beaches, dunes and cliffs. Their chicks are semi-precocial: they hatch with open eyes and a coat of down, and they can move about within hours, but for weeks they depend on their parents for food. Feeding follows a fixed routine. A parent returns to the nest, lowers its head and points its bill down toward the chick. The chick pecks at the tip of the bill, and the pecking prompts the parent to regurgitate partly digested food onto the ground or hold it in the bill, where the chick takes it.
Adult herring gulls have a yellow bill with a conspicuous red spot on the lower mandible, near the angle known as the gonys. Tinbergen noticed that newly hatched chicks aimed their pecks at that spot, and that they would also peck at objects that only loosely resembled a bill. The question was which features of the parent actually controlled the response. Was the chick responding to the whole bird, to the head, to the bill, or to something narrower still?
The researchers and their setting
Nikolaas "Niko" Tinbergen (1907–1988) grew up in The Hague and spent his youth watching birds and fish on the Dutch coast. Before the war he had already run elegant field experiments at Hulshorst on how digger wasps find their burrows, and laboratory work on courtship and fighting in three-spined sticklebacks. During the German occupation he was held for roughly two years (from 1942) among the prominent Dutch hostages at Kamp Sint-Michielsgestel. After the war he returned to Leiden and built a student group that worked on gulls and other birds, much of it on the island of Terschelling in the Wadden Sea.
Albert Perdeck, his co-author on the 1950 paper, later became known for bird-migration research, including large displacement experiments with starlings. In 1949 Tinbergen moved to the University of Oxford, where he went on to found the Animal Behaviour Research Group and to supervise a generation of behavioral biologists. His popular book The Herring Gull's World, first published in 1953, described the gull work for a general audience and kept the experiments in circulation for decades.
The experimental approach: systematic variation with models
The core method was simple and still worth copying. If you want to know which feature of a complex stimulus drives a behavior, build models that differ in one feature at a time and compare how strongly animals respond to each. Because chick responses vary from moment to moment, Tinbergen and Perdeck presented models in a balanced order, counted pecks in fixed trials, and compared each test model against a standard model shown in the same sessions. That standard gave them a stable reference point, so they could express each variant's effect as more or less than the standard.
Sign stimulus — an external stimulus that releases a fixed action pattern in an animal. It is typically a simple, often abstract cue that triggers innate behavior without learning.
The models were mostly flat cardboard heads with interchangeable bills, along with a three-dimensional plaster head and, later, bare bills and rods. The researchers varied:
- The color of the head
- The color of the bill
- The color of the patch on the bill, and whether there was a patch at all
- The contrast between patch and bill
- The shape and length of the bill
- The height and angle at which the bill was presented, and how it moved
What the models revealed
The results, reported in "On the stimulus situation releasing the begging response in the newly-hatched herring gull chick," overturned the intuitive view that the chick recognizes its parent as a whole.
The head barely mattered. Changing the color of the head, or leaving the head off and showing only a bill, made little difference compared with changes to the bill itself. A chick pecking at a parent is not, in any useful sense, looking at the parent.
The red patch mattered a great deal. A model with a red patch on the bill drew far more pecks than an otherwise identical model with no patch. When patches of other colors were tested, red was the most effective, and contrast between the patch and the surrounding bill also raised the response. Later summaries of the study list the order of effectiveness as a cardboard head with a red spot, then a bill alone with a red spot, then a head with no red spot at all.
Shape and movement mattered too. Long, thin, low-held objects did better than short or high ones, and a bill moved slightly back and forth outperformed a still one. The researchers concluded that several independent features each added to the effect, a pattern Tinbergen later described as "heterogeneous summation": the response grows with the total of separate cues, rather than requiring a complete, integrated image.
An unnatural model won. The most striking result came from a thin red rod with three white bands near its tip. It looked nothing like a gull. Yet chicks pecked at it at a higher rate than they pecked at the other models, including a cardboard head with a red-spotted bill. The rod's slender shape and strong contrasts exaggerated exactly the features the chick's response appeared to be tuned to.
| Model | Main features | Relative response (direction only) |
|---|---|---|
| Thin red rod with three white bands | Long, slender, red with strong white contrast, no head | Highest; exceeded the head models |
| Cardboard head with red bill spot | Flat head, yellow bill, red patch | High |
| Bill alone with red spot | No head, yellow bill, red patch | Somewhat lower |
| Head with no red spot | Realistic head shape, uniform bill | Markedly lower |
Direction of effects as reported in Tinbergen and Perdeck (1950) and secondary summaries. Exact counts varied between test series, so this table shows the ranking rather than percentages.
The discovery of supernormal stimuli
A sign stimulus that works better than the natural one is called a supernormal stimulus. The red-and-white rod is one of the classic examples, and Tinbergen collected others. Oystercatchers offered a choice between their own egg and an artificial egg several times larger would try to incubate the giant one. Ringed plovers, in earlier work by Otto Koehler and Zagarus, favored more boldly speckled eggs than their own. Male sticklebacks, whose red belly signals breeding condition, attacked crude models with red undersides more readily than accurate fish models that lacked the red.
These cases have a common logic. A response tuned to "the bigger the egg, the better" or "the redder and more contrasting, the better" works well when nature never supplies extreme values. Real eggs come in a narrow range of sizes; real bills come in one color pattern. A rule with no upper limit costs nothing in that world, so selection never shaped one. An experimenter, or a human-made object, can step outside that range and expose the rule.
Why sign stimuli matter: innate behavior is rule-based
Tinbergen's interpretation was that the chick's response is controlled by a mechanism that filters incoming information and passes on only a few features. In the classical ethological vocabulary of Tinbergen and Konrad Lorenz, the sign stimulus acts on an "innate releasing mechanism," which in turn triggers a "fixed action pattern" such as pecking. The terms have aged, but the observation behind them holds: early in life, a small set of cues is enough to switch on an essential behavior.
That design makes sense under the conditions a hatchling faces. In the first hours after hatching, the most important object in a chick's world that is long, thin, pointing down, and carrying a red spot is a parent's bill. A detector that responds to those features does not need to be accurate in any broader sense. It needs to be fast and to work with almost no experience. Its occasional false alarms, such as pecking at a pencil, cost little. Missing a meal could cost a great deal.
The four questions: what the gull work was really about
The gull experiments are often presented as a story about "stupid" animals, but Tinbergen saw them as part of a broader scientific program. In 1963 he published "On aims and methods of ethology," which set out four distinct questions that any behavior raises:
- Mechanism (causation): What stimuli and internal processes produce the behavior? For the chick, this is the red spot, the bill's shape and movement, and the neural filtering of those cues.
- Development (ontogeny): How does the behavior appear and change over an individual's life?
- Function (survival value): How does the behavior help the animal survive and reproduce? Pecking gets the chick fed.
- Evolution (phylogeny): How did the behavior arise in the species' history, and how does it compare with relatives?
The pecking study speaks most directly to mechanism. Tinbergen's emphasis on keeping the questions separate, while answering all four, remains one of his most cited contributions. A mechanism explanation ("the chick pecks because it sees a red spot") does not compete with an evolutionary, survival-value explanation ("chicks that pecked effectively were fed and survived"); both can be true at once.
Tinbergen's own group pursued the survival-value question with field experiments of the same style. A well-known example is the work on black-headed gulls removing eggshells from the nest after hatching. By laying out eggs with and without broken shells nearby, the researchers showed that the white inner surface of a shell attracts predators such as crows, which helped explain why parents carry the shells away.
Hailman's challenge: the pecking response develops
The strongest follow-up came from the American ethologist Jack P. Hailman. In a 1964 dissertation, published by Brill in Leiden in 1967 as The Ontogeny of an Instinct, Hailman studied pecking in chicks of the laughing gull, a species whose adults have a dark red bill on a black head, along with related gulls.
Hailman confirmed that newly hatched chicks respond to simple features of a bill-like stimulus. But he also showed that the response changes rapidly. Pecking accuracy improves over the first days as chicks steady their stance and refine their aim. Their preferences also narrow: older chicks become more selective, responding more to stimuli that resemble their own species' parents. Hailman argued that the behavior is shaped by experience, including the reward of being fed.
Hailman's results did not overturn the sign-stimulus finding. They added a developmental layer to it. The newly hatched chick starts with a crude, broad detector, and experience turns that detector into a more precise skill. "Instinct" in this view is a starting point that development builds on, not a finished behavior locked in at hatching. Hailman summarized this work for a general audience in Scientific American in 1969 under the title "How an Instinct Is Learned."
The broader criticism of the instinct concept
Hailman's work fits a larger debate. In 1953, the American comparative psychologist Daniel Lehrman published "A critique of Konrad Lorenz's theory of instinctive behavior" in The Quarterly Review of Biology. Lehrman argued that labeling a behavior "innate" says little, because every behavior emerges through development in a particular environment. He objected that Lorenz effectively defined instinct in negative terms, as whatever was not learned, rather than tracing how the behavior actually came about.
Tinbergen took the criticism seriously, more so than Lorenz. His 1963 framework, with development as one of four separate questions, was partly a response. In later years he became more cautious about the phrase "innate behavior" and more interested in how experience and inherited predispositions interact.
Other criticisms are more technical. Model experiments measure a response to a stimulus held by a human, not a natural interaction between parent and chick. Pecking rates also vary with hunger, time of day and fatigue, which is why careful comparisons against a standard model were essential. Finally, supernormal results with artificial objects tell us about the limits of a detector, but they do not mean the animal regularly makes mistakes in the wild.
Extension to other animals: supernormal stimuli everywhere
Researchers have since reported supernormal and sign-stimulus effects in many groups:
- Sticklebacks: Tinbergen's earlier work showed that males attack simple models with red undersides more than accurate fish models without the red.
- Oystercatchers and other ground-nesting birds: Birds offered oversized dummy eggs attempt to incubate them, preferring them over normal eggs.
- Australian jewel beetles: In a widely cited 1983 report, Darryl Gwynne and David Rentz described male Julodimorpha bakewelli beetles trying to mate with discarded brown beer bottles. The bottles' glossy brown surface and dimpled glass apparently resembled an exaggerated female.
- Cuckoos and their hosts: Studies of common cuckoos in reed warbler nests found that a single cuckoo chick produces rapid begging calls that sound like a whole brood of warbler chicks, helping it obtain enough food from its foster parents.
The examples share the original lesson. A response that evolved to track a natural cue can be captured by a stimulus that pushes that cue beyond the natural range.
Common misconceptions about the experiment
"Chicks cannot tell their parents from a stick." The experiments show that the first begging response is driven by simple features. They do not show that gulls never learn to recognize parents. Herring gull chicks do learn to distinguish their own parents, particularly by voice, as they grow.
"The red rod was the most realistic stimulus." The opposite is the point. The rod looked nothing like a gull, which is why its success was so informative.
"The chicks were being fooled." The response is working as designed. Under natural conditions, the cues it uses are reliable. The "mistake" appears only when a human introduces an object that never occurs in a gull colony.
"The behavior is pure reflex and cannot change." Hailman's work shows that pecking accuracy and selectivity both improve with experience in the first days.
"Tinbergen did it all in 1946 on one beach." The work was a series of experiments run by Tinbergen and Perdeck in the late 1940s and published in 1950, building on earlier observations by Tinbergen and on related work by other ethologists.
Implications for understanding animal intelligence
The gull work invites a question: if animals rely on such simple cues, how intelligent are they? Tinbergen's framing suggests that the question is badly posed. A hatchling with a few hours of experience cannot afford elaborate recognition. A detector that picks out a few dependable features is the efficient solution to the problem it actually faces. Later in life, the same bird uses learning, memory, and individual recognition to handle more complex problems.
This view remains central to modern behavioral ecology and cognitive science. Animals are not miniature humans failing at human tasks. Their perceptual systems are tuned to particular problems in particular environments. Studying where those systems break, as with supernormal stimuli, is one of the best ways to discover what they are tuned to in the first place. The same approach underlies the study of learning more generally; compare the conditioned responses in Pavlov's classical conditioning experiments, where experience links a new cue to an existing response.
Legacy and modern applications
Tinbergen shared the 1973 Nobel Prize in Physiology or Medicine with Konrad Lorenz and Karl von Frisch "for their discoveries concerning organization and elicitation of individual and social behaviour patterns." The award recognized ethology as a branch of biology, and the herring gull study was part of that body of work.
The idea of supernormal stimuli has spread far beyond gulls. In her 2010 book Supernormal Stimuli: How Primal Urges Overran Their Evolutionary Purpose, Harvard psychologist Deirdre Barrett argued that many modern human products, from processed foods to entertainment, act as supernormal versions of cues our ancestors evolved to seek out. That application is debated, and human behavior is shaped far more by culture and learning than a gull chick's first pecks, but the comparison has been influential.
The method has also had a long life in neuroscience and engineering. Recording from sensory neurons while presenting simplified stimuli echoes the model approach, and the logic of "which feature drives the response" underlies work on feature detectors in visual systems, discussed in how the nervous system works. Researchers studying animal behavior now use robotic and computer-animated models, which can vary single features with precision Tinbergen could not achieve with cardboard.
Finally, the experiments matter for conservation. Many birds rely on sign stimuli in feeding, mating and nesting, and human-made objects such as plastic debris, artificial lights, or glossy litter can hijack those responses. Understanding which cues release a behavior helps predict how animals will react to new features of their environment. The broader framework for asking why such behaviors evolved in the first place is natural selection, a theme that also runs through the Kettlewell peppered moth experiment.
Sources
- Universiteit Leiden - About Niko Tinbergen
- Nobel Prize - Niko Tinbergen Lecture (1973)
- Tinbergen, N. & Perdeck, A. C. (1950). On the stimulus situation releasing the begging response in the newly-hatched herring gull chick (Larus argentatus argentatus Pont.). Behaviour, 3, 1–39.
- Tinbergen, N. (1963). On aims and methods of ethology. Zeitschrift für Tierpsychologie, 20, 410–433.
- Hailman, J. P. (1967). The Ontogeny of an Instinct: The Pecking Response in Chicks of the Laughing Gull (Larus atricilla L.) and Related Species. Leiden: Brill. (Jack Hailman - Wikipedia)
- Lehrman, D. S. (1953). A critique of Konrad Lorenz's theory of instinctive behavior. The Quarterly Review of Biology, 28(4), 337–363. (Rutherford Journal - Konrad Lorenz on Instinct)
- Begging in animals - Wikipedia (summary of Tinbergen & Perdeck results)
- The Auk, vol. 74 (1957) - discussion of Tinbergen & Perdeck's herring gull models
- Audubon - Why do gulls have a red spot on their bills?
- Barrett, D. (2010). Supernormal Stimuli: How Primal Urges Overran Their Evolutionary Purpose. W. W. Norton. (Harvard Gazette coverage)
- New World Encyclopedia - Nikolaas Tinbergen
FAQ
Q: Why would chicks peck at a red rod more than at a realistic parent model? A: The rod exaggerates features the pecking response is tuned to: a long, thin shape and strong color contrast near the tip. Because the response adds up separate cues rather than checking for a complete gull, an object that pushes those cues beyond natural values can beat a realistic model. The chick is not malfunctioning; its detector was simply never shaped by objects like that rod.
Q: When were the herring gull experiments carried out and published? A: Tinbergen and A. C. Perdeck ran the model experiments in the late 1940s and published them in 1950 in the journal Behaviour (volume 3, pages 1–39). Tinbergen later described the work for general readers in The Herring Gull's World (1953).
Q: Do herring gull chicks learn to recognize their real parents? A: Yes. The first begging response depends on simple visual cues, but chicks become more selective as they grow and learn to recognize their own parents, especially by their calls. Jack Hailman's research on laughing gulls showed that pecking accuracy and preference both change with experience in the first days.
Q: What is the difference between a sign stimulus and a supernormal stimulus? A: A sign stimulus is the simple feature, such as the red bill spot, that releases a behavior. A supernormal stimulus is an exaggerated version of that feature, such as the red rod with white bands, that releases the behavior more strongly than the natural object does.
Q: Could humans be fooled by supernormal stimuli the way animals are? A: Some researchers argue so. Deirdre Barrett's 2010 book Supernormal Stimuli makes the case for food, media and other products. The idea is debated because human behavior is heavily shaped by culture and learning, but the comparison has been influential.
Q: Does this mean animal behavior is just reflex? A: No. The gull experiments reveal one layer of behavior, the initial trigger. Later work, including Hailman's and Daniel Lehrman's critique of instinct theory, emphasized that even "innate" responses develop and are refined by experience.
Q: What other animals show supernormal stimulus effects? A: Classic examples include sticklebacks attacking red-bellied models, oystercatchers trying to incubate oversized eggs, Australian jewel beetles courting brown beer bottles, and reed warblers feeding cuckoo chicks whose begging calls sound like a whole brood.
Q: How did Tinbergen's work lead to the founding of ethology? A: Tinbergen, with Konrad Lorenz and Karl von Frisch, showed that behavior could be studied with controlled experiments and explained in biological terms. His 1963 "four questions" framework (mechanism, development, function and evolution) gave the field a lasting structure, and the three shared the 1973 Nobel Prize in Physiology or Medicine.
Read next:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

The Fox Domestication Experiment: Breeding Tameness in Real Time
Starting in 1959, Soviet geneticist Dmitry Belyaev selectively bred silver foxes for tameness alone. Within a few generations, the foxes developed floppy ears, piebald coats, and dog-like behavior.


Why Do Flamingos Stand on One Leg? The Science
Flamingos stand on one leg because it's literally easier than two. The biomechanics, thermoregulation, and evolution behind the pink pose.

Why Are We Ticklish? The Biology of Tickling
Why can't you tickle yourself? The neuroscience of knismesis vs gargalesis, the cerebellum's prediction trick, and why feet/armpits are most ticklish.