Chicken cognition: learning, memory, social behaviour, navigation

Chickens learn from experience, remember where hidden goals are, use information from flock-mates and orient themselves with spatial cues. Controlled experiments have demonstrated associative learning, short-term retention of hidden locations, discrimination between quantities, sensitivity to simple physical constraints, social learning and reorientation by environmental geometry. These findings show flexible performance across several domains, but they do not amount to a single score of 'intelligence'. Each result depends on a particular task, age, motivation and testing context, so broad claims need to be tied back to what the birds actually had to do.

Christine J. Nicol's 2015 chapter 'Learning, intelligence and cognition' provides a starting point for examining research on memory, quantity, object occlusion, navigation and social information in chickens. Peer-reviewed experiments show what these birds can learn, retain and use under different testing conditions. The evidence helps clarify the cognitive abilities involved in everyday behaviour without reducing them to a single measure of intelligence. These capacities also have practical relevance for understanding chicken behaviour and welfare.

What cognition experiments measure in chickens

Cognition includes learning, attention, memory, categorisation, expectation, decision-making and social or spatial information use. Researchers study these processes through controlled behavioural tasks, not general intelligence tests. Chickens may choose visual cues, recall hidden objects, observe other birds or search after a setting changes. Such tasks identify what information influences behaviour, but success does not establish human-like understanding. Failure may reflect motivation, sensory demands or unsuitable testing conditions.

Christine J. Nicol's 2015 synthesis presents learned and unlearned behaviour as interacting processes. Chicks hatch with predispositions affecting attention and attachment, while experience shapes what they approach or avoid. Later reviews describe substantial cognitive and social abilities, although evidence varies across domains and often comes from laboratory studies of young chicks or small groups. Rather than asking whether chickens are 'smart', researchers examine which information they use under defined conditions.

Learning, discrimination and categories

Associations can become flexible rules

Associative learning connects cues, actions or places with outcomes. Poultry experiments use colours, shapes, positions, key-pecking responses and food rewards to test these links. Researchers then examine whether chickens memorise specific stimuli, generalise to new examples or distinguish categories. Birds can learn rewarded choices and adjust when conditions change. Such learning also underlies experiments on numbers, objects and social information, which require attention, memory and choice.

Working memory and hidden objects

Remembering where a goal disappeared

Regolin and colleagues tested five-day-old chicks searching for a target hidden behind one of two identical screens after delays of 10–180 seconds. Targets were either artificial social companions or food. With screens visible throughout the delay, chicks chose correctly above chance at every interval, particularly for social targets. Performance also remained above chance overall when the screens were hidden during the delay. These findings demonstrate short-term location memory in this task, while highlighting the effects of goal type and motivation.

Occlusion tests probe expectations about objects

Object-permanence and 'naive physics' experiments test expectations about unseen objects. Chiandetti and Vallortigara raised newborn chicks with an imprinting object, then hid it behind screens differing in height, width or slope. Chicks preferred screens capable of concealing the object, even when prior physical interaction with it was restricted. This suggests early sensitivity to physical constraints without extensive manipulation experience, but only within the tested occlusion task, not a broad human-like grasp of physics.

What representative experiments actually showed

The studies below illustrate why task design and interpretation need to stay together. They use different ages, goals and measures, so their outcomes should not be combined into a single score. Instead, each experiment identifies a particular capacity under particular conditions.

Ability	Experimental task	Observed result	Main caution
Working memory	Five-day-old chicks searched after a target disappeared behind one of two screens; delays of 10-180 s.	Correct location chosen above chance across tested delays; social targets produced stronger performance.	A short, controlled delayed-search task; it does not define a general memory limit.
Quantity tracking	Newly hatched chicks watched two sets of imprinting objects disappear and then observed transfers between screens.	Birds preferentially approached the screen containing the larger final set.	Supports tracking of small quantities and transformations, not symbolic school arithmetic.
Occluded objects	An imprinting object disappeared behind screens whose dimensions could or could not conceal it.	Chicks preferred physically compatible screens.	Evidence is specific to early-life occlusion expectations and the imprinting paradigm.
Spatial reorientation	Chicks were disoriented and tested with three-dimensional layout geometry versus two-dimensional image cues.	Search was guided by three-dimensional environmental geometry.	Laboratory reorientation is narrower than real-world navigation over long distances.
Social learning	Hens watched trained demonstrators perform a food-rewarded key-pecking task.	Observation changed later task performance; demonstrator identity and status affected influence.	Social context changes learning, so observer and demonstrator characteristics matter.
Social inference	Hens observed dominance interactions before meeting a previously unseen opponent.	Subsequent behaviour changed according to the conflict outcome the hen had witnessed.	The experiment supports use of third-party social information, not human-like social reasoning in general.

Chicks can track quantity without symbolic numbers

What the arithmetic experiments test

Rugani and colleagues used filial imprinting to make small objects socially attractive to newly hatched chicks. The birds were reared with five identical objects and later watched groups of objects disappear behind two opaque screens. In further trials, the experimenters visibly transferred objects from one hidden set to the other. Chicks preferentially approached the screen that should contain the larger final number of objects. The work is often described as arithmetic because successful choices required keeping track of additions or subtractions to small sets. A more precise description is that chicks represented approximate or discrete small quantities well enough to update them after visible transfers. They were not using numerals, number words or learned written operations.

Geometry can guide reorientation

Spatial cognition is essential to a ground-dwelling bird that must move among feeding sites, shelter, nest locations and other flock members. Laboratory reorientation experiments simplify that challenge by disorienting a chick and asking where it searches for a goal. Lee, Spelke and Vallortigara compared three-dimensional environmental geometry with two-dimensional image information. Chicks spontaneously used the three-dimensional shape of the surrounding terrain or layout to reorient rather than merely matching a visual picture. This supports the idea that spatial representation can encode structural relationships in the environment. Other chicken studies also show use of landmarks and learned local cues, so geometry is one component of navigation rather than a universal replacement for visual landmarks.

Watching another bird can alter later choices

Social information may help chickens reduce risk and locate food. Nicol and Pope studied hens observing trained birds peck a key for food. Observers subsequently performed differently from controls, with effects varying by demonstrator. Dominant demonstrators were especially influential in the 1994 study; a 1999 experiment also found effects of demonstrator category and observer status. Social learning therefore depends on who demonstrates a behaviour and the observer's place in the group.

Calls are sensitive to both food and audience

Communication experiments show a related sensitivity to context. Evans and Marler studied male food calls and found that calling increased when food became available, while the presence of a hen further increased calling after food presentations. The audience effect was specific: social context changed the signalling response without simply increasing all activity. Such experiments do not establish that chicken calls carry language-like meanings, but they do show that vocal production can depend jointly on an external event and on who is present. That is consistent with a communication system in which receivers, relationships and immediate circumstances influence behaviour.

Chicken cognition is best understood as a set of flexible, task-specific abilities shaped by what birds need to remember, predict and do.

Observed relationships can guide later behaviour

Dominance relationships create another information problem: an individual cannot safely test every possible opponent through direct conflict. Hogue, Beaugrand and Laguë examined hens that watched a familiar dominant bird either defeat or lose to a stranger before the observer met that stranger. The bystanders changed their subsequent behaviour according to the outcome they had witnessed, showing that third-party social interactions could inform later decisions. The design does not require the observer to have fought the stranger first. This is a useful demonstration of information use across relationships, although labels such as 'logical reasoning' should be reserved for experiments that explicitly rule out simpler associative explanations.

More recent work has directly tested transitive inference with ordered relationships. Daisley, Vallortigara and Regolin reported in 2021 that performance in a transitive-inference task varied with sex and social rank, with different rank patterns in female and male chicks. That finding links reasoning performance with the social structure in which the birds live, but it also warns against assuming that every individual will perform identically. Transitive tasks can often be solved through several cognitive strategies, including learned representations of ordered pairs. The strongest conclusion is therefore behavioural: chickens can use learned relational information to make choices about combinations they have not been trained on directly, under the conditions of the experiment.

Why cognition matters for chicken welfare

Cognition affects welfare because environments create expectations, choices and opportunities for control. A bird that learns where food, perches or shelter are located can be affected when access changes unpredictably. A social animal can also be affected by unstable group composition, competition or the availability of information from flock-mates. The 2015 chapter explicitly connects cognitive capacities with welfare, and that connection is best treated as a design question rather than as a claim that one housing feature is automatically good or bad. Environments differ in space, resources, social density, predictability and opportunities to perform motivated behaviour, and those factors interact with learning and memory.

Cognitive research suggests several questions worth asking when evaluating management or enrichment. These are not a checklist that guarantees welfare, but they follow from the abilities documented in behavioural studies:

Can birds predict where important resources are and reach them without unnecessary confusion or competition?
Does the environment provide meaningful opportunities to explore, forage and use spatial information rather than remaining behaviourally barren?
Are changes in routines, group membership or resource location introduced in ways that reduce avoidable uncertainty?
Can birds use social information without being forced into persistent conflict around limited resources?
Do enrichment objects and housing features remain usable and relevant after the birds have learned their layout and consequences?

The broader welfare lesson is that chickens are active information users. They learn regularities, remember locations, respond to social partners and adjust choices after experience. This does not mean that every complex-looking behaviour requires a complex mental explanation, nor does it settle debates about particular production systems. It does mean that welfare assessment should not treat chickens as behaviourally interchangeable units responding only to immediate stimuli. Their prior learning, social history and expectations can influence how they experience and use an environment.

How to read claims about chicken intelligence

Claims about animal intelligence are most reliable when they preserve the structure of the experiment. Ask what the birds could see, how they were trained, what counted as a correct response, what alternatives were controlled and whether a result was replicated. Young chicks are especially useful for studying early-emerging capacities because researchers can control experience from hatching, but that also makes those paradigms different from the daily lives of adult hens. Likewise, an experiment with six or twenty birds can reveal a mechanism without estimating how common a trait is across breeds, ages and housing conditions. Sample size and population therefore matter as much as an attention-grabbing task label.

The current evidence supports a coherent picture: domestic chickens possess functional learning, memory, spatial and social-cognitive abilities that are flexible enough to solve a range of controlled problems. Some behaviours emerge very early; others depend strongly on experience and social context. The evidence is strongest when conclusions stay close to observable choices and experimental controls, and weaker when terms such as 'counting', 'physics' or 'logic' are used without explaining the task. Seen this way, chicken cognition is neither mysterious nor trivial. It is a set of evolved and learned capacities that help birds organise behaviour in a changing physical and social world.

Frequently asked questions

Are chickens intelligent?

Chickens show learning, memory, quantity discrimination, spatial and social-information use in controlled experiments. 'Intelligence' is too broad to be a single scientific result, so the strongest claims describe the task the birds solved. Performance also varies with age, motivation, social context and the cues available during testing.

Do chickens remember hidden objects?

Experiments with young chicks show that they can remember where an attractive object disappeared after delays and can search for objects that are no longer visible. These results support working memory and object-permanence-related abilities in specific laboratory tasks; they do not define a universal memory duration for every chicken or situation.

Can chickens count?

Newly hatched chicks have distinguished small quantities and tracked visible transfers of objects between hidden sets. Researchers describe these findings as numerical or arithmetic-like cognition because the birds update which set is larger. The experiments do not involve symbols, number words or learned written arithmetic, so 'counting' should be used carefully.

How do chickens navigate?

Chickens can use remembered locations, landmarks and geometric properties of an environment. In reorientation experiments, chicks have used three-dimensional layout geometry to search after disorientation. Laboratory arenas isolate particular cues, however, so these studies explain components of spatial cognition rather than reproducing all of the challenges of outdoor navigation.

Do chickens learn from one another?

Yes. Hens that watch trained demonstrators can later perform differently from birds that had no demonstration, and the identity or social status of the demonstrator can affect the outcome. Chickens also use observed social interactions when responding to unfamiliar birds, showing that social information can influence both learning and decision-making.

Sources

Nicol CJ. 'Learning, intelligence and cognition.' In: The Behavioural Biology of Chickens. 2015; pp. 128-144.
Marino L. Thinking chickens: a review of cognition, emotion, and behavior in the domestic chicken. Animal Cognition. 2017;20(2):127-147.
Regolin L, Rugani R, Pagni P, Vallortigara G. Delayed search for social and nonsocial goals by young domestic chicks, Gallus gallus domesticus. Animal Behaviour. 2005;70(4):855-864.
Rugani R, Fontanari L, Simoni E, Regolin L, Vallortigara G. Arithmetic in newborn chicks. Proceedings of the Royal Society B. 2009;276(1666):2451-2460.
Chiandetti C, Vallortigara G. Intuitive physical reasoning about occluded objects by inexperienced chicks. Proceedings of the Royal Society B. 2011;278(1718):2621-2627.
Lee SA, Spelke ES, Vallortigara G. Chicks, like children, spontaneously reorient by three-dimensional environmental geometry, not by image matching. Biology Letters. 2012;8(4):492-494.
Nicol CJ, Pope SJ. Social learning in small flocks of laying hens. Animal Behaviour. 1994;47(6):1289-1296.
Nicol CJ, Pope SJ. The effects of demonstrator social status and prior foraging success on social learning in laying hens. Animal Behaviour. 1999;57(1):163-171.
Evans CS, Marler P. Food calling and audience effects in male chickens, Gallus gallus: their relationships to food availability, courtship and social facilitation. Animal Behaviour. 1994;47(5):1159-1170.
Hogue ME, Beaugrand JP, Laguë PC. Coherent use of information by hens observing their former dominant defeating or being defeated by a stranger. Behavioural Processes. 1996;38(3):241-252.
Daisley JN, Vallortigara G, Regolin L. Low-rank Gallus gallus domesticus chicks are better at transitive inference reasoning. Communications Biology. 2021;4:1344.

Imogen Farrant — author

Imogen Farrant is an animal health and welfare writer with a background in animal science and a particular interest in companion animal welfare. Her work focuses on translating veterinary and animal-welfare research into clear, accessible information for pet o...

Chicken cognition: learning, memory, social behaviour, navigation

What cognition experiments measure in chickens