Corvids: The Most Intelligent Birds on Earth
In the hierarchy of animal intelligence, one family of birds stands apart. Corvids -- the family Corvidae, encompassing crows, ravens, magpies, jays, jackdaws, rooks, and nutcrackers -- have demonstrated cognitive abilities that rival those of great apes and, in some domains, surpass them. They fashion and use tools, recognize themselves in mirrors, plan for the future, hold grudges against specific human faces for years, and appear to mourn their dead. These are not anecdotal curiosities. They are findings from decades of rigorous peer-reviewed research conducted at institutions from Cambridge to Auckland.
What makes corvid intelligence so remarkable is not merely that it exists, but that it evolved along an entirely separate path from mammalian intelligence. The last common ancestor of birds and mammals lived roughly 320 million years ago. Every cognitive feat a corvid performs was shaped by independent evolutionary pressures, offering profound evidence that complex intelligence is not a fluke of primate brains but a convergent solution to the demands of complex social and physical environments.
The Corvid Family: A Global Dynasty
The family Corvidae comprises over 130 species distributed across every continent except Antarctica. The most studied members include:
- Common crows (Corvus corone and Corvus brachyrhynchos) -- adaptable generalists found across Europe, Asia, and North America.
- Common ravens (Corvus corax) -- the largest passerine birds, with wingspans exceeding 1.3 meters, found throughout the Northern Hemisphere.
- Eurasian magpies (Pica pica) -- striking black-and-white birds recognized for their inquisitive nature and, as research would later reveal, self-awareness.
- Eurasian jays (Garrulus glandarius) -- woodland corvids with an astonishing memory for thousands of food cache locations.
- Western jackdaws (Coloeus monedula) -- smaller, social corvids that form lifelong pair bonds and exhibit cooperative behaviors.
- New Caledonian crows (Corvus moneduloides) -- island specialists that have become the gold standard for avian tool-use research.
What unites these species beyond taxonomy is an unusually large brain-to-body ratio. Corvid brains are proportionally comparable in size to those of chimpanzees when adjusted for body mass, and their neural architecture, though structurally different from the mammalian cortex, performs functionally analogous computations.
New Caledonian Crows: The Master Toolmakers
No discussion of corvid intelligence begins anywhere other than the cloud forests of New Caledonia, a French territory in the southwestern Pacific. Here, a species of crow has developed a tool-manufacturing tradition so sophisticated that it has drawn comparisons to early hominin technology.
Crafting Tools from Nature
New Caledonian crows routinely fashion three types of tools in the wild. They select twigs, strip off side branches and leaves, and shape one end into a functional hook for extracting grubs from deadwood. They tear strips from the stiff, barbed leaves of Pandanus palms, creating tapered probes with serrated edges that grip insect larvae. And they manufacture stepped-cut tools from Pandanus leaves using a standardized, multi-step cutting process that varies between populations -- suggesting cultural transmission rather than instinct alone.
Christian Rutz and his colleagues at the University of St Andrews have documented that different New Caledonian crow populations maintain distinct tool designs, passed from parents to offspring. This mirrors the way human cultural traditions are transmitted across generations, albeit at a simpler scale.
Betty and the Wire
In 2002, a captive New Caledonian crow named Betty astonished researchers at the University of Oxford. Presented with a straight piece of wire and a small bucket of food lodged inside a vertical tube, Betty spontaneously bent the wire into a hook and used it to lift the bucket out. She had never encountered wire before. She had never been trained. She simply innovated a solution to a novel problem using a material that does not exist in her species' natural environment. The study, published in Science by Alex Weir, Jackie Chappell, and Alex Kacelnik, was a watershed moment in animal cognition research.
Sequential Tool Use
Perhaps the most striking demonstration came from experiments on meta-tool use -- using one tool to obtain another tool. In a 2007 study by Alex Taylor and colleagues at the University of Auckland, New Caledonian crows used a short stick to retrieve a longer stick from a toolbox, then used the longer stick to extract food from a deep tube. This sequential, multi-step planning had previously been documented only in great apes. Some individuals solved the problem on the very first attempt, suggesting genuine insight rather than trial-and-error learning.
The Aesop's Fable Test
Drawing inspiration from the ancient fable in which a thirsty crow drops stones into a pitcher to raise the water level, researchers designed the "Aesop's fable paradigm." New Caledonian crows were presented with a tube of water with a floating food reward just out of reach. The crows dropped stones into the tube, raising the water level until the food was accessible. Critically, they chose solid objects over hollow ones (which would float rather than displace water), heavy objects over light ones, and tubes with water over tubes with sand. They understood -- or at least behaved as if they understood -- the causal relationship between object density, water displacement, and reward access.
Ravens: The Strategic Planners
If New Caledonian crows are the master craftsmen of the corvid world, ravens are its chess players. The common raven has long captivated human imagination, and modern science has shown that its reputation for cunning is well earned.
Planning for the Future
In a landmark 2017 study published in Science, Mathias Osvath and Can Kabadayi at Lund University in Sweden demonstrated that ravens could plan for events up to 17 hours in the future. The ravens learned that a specific tool could open a puzzle box containing a high-value food reward. When later offered a choice between the tool, a distractor object, and an immediate but less valuable food reward, the ravens chose the tool -- and then waited to use it on the puzzle box the following day. Their success rate of roughly 90 percent matched or exceeded that of great apes tested on similar delayed-gratification tasks.
Bartering and Delayed Gratification
In the same experimental framework, ravens also demonstrated bartering ability. They learned that a specific token (a bottle cap) could be exchanged with a human experimenter for a food reward. When offered a choice between an immediate low-value food item and the token, ravens reliably chose the token, waiting for the experimenter to return and exchange it for a superior reward. This capacity for delayed gratification and symbolic exchange was, until recently, considered a hallmark of human economic cognition.
Play and Social Complexity
Ravens are among the most playful of all bird species. They have been filmed sliding down snow-covered rooftops on their backs, hanging upside down from branches, dropping and catching sticks in midair, and engaging in aerial acrobatics with no apparent purpose beyond apparent enjoyment. Play behavior in animals is closely associated with cognitive flexibility, social learning, and neural development.
Bernd Heinrich, the renowned naturalist and author of Mind of the Raven, spent years observing wild ravens in Maine and Vermont. He wrote:
"Ravens are not just smart. They are inventive, scheming, and aware of what others around them are thinking. They live in a social world as complex as that of any primate."
The Tower of London Ravens
No account of ravens would be complete without mention of the Tower of London, where a group of captive ravens has been maintained since at least the reign of Charles II in the 17th century. Legend holds that if the ravens ever leave the Tower, the Crown and the kingdom will fall. Today, the Ravenmaster -- currently a Yeoman Warder of the Tower -- cares for a minimum of six ravens at all times. Their flight feathers are trimmed to prevent departure, but they are otherwise free to roam the Tower grounds, where they interact with millions of visitors each year. The superstition may be folklore, but the ravens themselves display the full range of corvid intelligence: caching food, manipulating visitors, and engaging in complex social negotiations with one another.
Eurasian Magpies: The Self-Aware Bird
In 2008, a study led by Helmut Prior at the Goethe University in Frankfurt delivered one of the most startling findings in comparative cognition. Eurasian magpies passed the mirror self-recognition test -- the first and, to date, only non-mammal species confirmed to do so.
The Mirror Test
In the classic mirror test, a colored mark is placed on an animal's body in a location visible only with the aid of a mirror. If the animal uses the mirror to inspect and attempt to remove the mark, it is considered to possess some form of self-recognition. Previously, only great apes, bottlenose dolphins, Asian elephants, and arguably orcas had passed this test.
Prior's magpies were marked with a small yellow or red sticker below their beak. When placed in front of a mirror, the magpies scratched at the mark, attempted to remove it, and showed increased self-directed behavior. When given a black sticker (invisible against their dark plumage) or no sticker at all, they showed no such response. The controls were clean. The conclusion was unavoidable: Eurasian magpies recognize their own reflection.
This finding was significant not only for what it revealed about magpies but for what it implied about the neural substrates of self-awareness. Magpie brains lack a neocortex entirely. Whatever neural architecture supports their self-recognition must be fundamentally different from the cortical structures responsible for the same ability in mammals.
Jays: Memory Champions of the Bird World
Eurasian jays and their North American relatives, the scrub-jays, have provided some of the most elegant evidence for complex memory in non-human animals.
Caching on an Extraordinary Scale
Clark's nutcrackers (Nucifraga columbiana), a corvid of the North American mountain West, cache an estimated 30,000 or more pine seeds each autumn across thousands of individual hiding spots scattered over areas up to 15 miles wide. They recover these caches months later, even after heavy snowfall has blanketed and transformed the landscape. Spatial memory on this scale is staggering: the birds appear to use a combination of landmarks, triangulation, and possibly sun-compass orientation to relocate their stores with remarkable accuracy. Recovery rates consistently exceed 70 percent.
Episodic-Like Memory: Nicola Clayton's Research
The most celebrated work on jay cognition comes from Nicola Clayton, a professor at the University of Cambridge who has spent over two decades investigating the mental lives of western scrub-jays. Clayton's research demonstrated that scrub-jays remember not only where they cached food but what they cached and when they cached it -- the "what-where-when" triad that constitutes episodic-like memory.
In a key experiment, scrub-jays cached both perishable wax moth larvae (a preferred food) and non-perishable peanuts. When allowed to recover their caches after a short delay (within the larvae's freshness window), the jays preferentially retrieved the larvae. After a longer delay (when the larvae would have decayed), they switched to recovering peanuts instead. The birds had integrated information about food type, location, and elapsed time -- a feat that was thought to require the kind of mental time travel previously attributed only to humans.
Clayton reflected on her findings:
"What the jays are doing is not simply remembering facts. They are traveling back in time in their minds, re-experiencing past events and using those memories to make decisions about the future. That is something we once thought only humans could do."
Social Cognition and Cache Protection
Scrub-jays that have stolen food from other birds' caches are more likely to re-hide their own caches when they know they were observed while caching. Jays that have never stolen do not show this behavior. This suggests a form of experience projection -- the thieving jays attribute their own pilfering tendencies to potential observers, a cognitive ability related to theory of mind. Clayton's work on this phenomenon has been published in Nature and Animal Behaviour and remains one of the strongest cases for theory-of-mind-like abilities outside of primates.
Crows and Facial Recognition: The Seattle Mask Experiment
One of the most famous experiments in corvid research began on the campus of the University of Washington in Seattle in 2006.
The Experiment
John Marzluff and his students wanted to test whether wild American crows could recognize individual human faces. They purchased two rubber masks: a "dangerous" caveman mask worn by researchers who trapped, banded, and released crows on campus, and a "neutral" mask worn by people who simply walked through the area without disturbing the birds.
In the weeks and months following the trapping events, crows aggressively scolded anyone wearing the caveman mask -- cawing, dive-bombing, and mobbing the wearer -- while ignoring the neutral mask. The response was specific to the mask itself: it occurred regardless of the wearer's body type, clothing, gait, or other characteristics. The crows had isolated the face as the identifying feature of the threat.
A Grudge That Spreads
What made the experiment extraordinary was what happened next. Over the following years, the number of crows that scolded the caveman mask increased, even though no further trapping had occurred. Crows that had never been trapped -- including juveniles born after the original capture events -- joined the mobbing response. The threat knowledge was being transmitted socially, from experienced birds to naive ones, across generations.
Five years after the original trapping, Marzluff's team walked predetermined routes across campus wearing the masks. Approximately 60 percent of crows encountered along the route scolded the dangerous mask. At some locations where the original trapping had taken place, the response rate approached 100 percent.
Brain Imaging Confirms Fear Responses
In a follow-up study, Marzluff and his colleague Donna Cross used PET (positron emission tomography) scans to image the brains of crows exposed to the dangerous and neutral masks. Crows viewing the dangerous mask showed significantly increased activation in the amygdala analog (the arcopallium), the hippocampal formation, and areas associated with visual processing and fear conditioning. The neural response pattern was strikingly parallel to what is seen in mammalian brains processing learned threats.
Crow Funerals: Responding to Death
When a crow dies, something remarkable often happens. Other crows gather around the body, sometimes dozens of them, in near-silence. They may remain for 15 to 30 minutes before dispersing. To human observers, the gatherings look unmistakably like funerals, and the term "crow funeral" has entered both popular and scientific literature.
Kaeli Swift, working under Marzluff at the University of Washington, conducted a rigorous study of this behavior. She placed a taxidermied dead crow in various locations and recorded the responses of wild crows. The living crows reliably gathered around the dead crow, sounded alarm calls, and avoided feeding in the area for up to 48 hours. When Swift paired the dead crow with a novel "predator" -- a person holding the dead crow -- the crows subsequently treated that person as a threat, scolding and mobbing them in future encounters.
The conclusion: crow funerals are not grief rituals in the human sense, but they serve a vital function. By gathering around a dead conspecific and associating the death with nearby threats, crows learn about dangers in their environment without having to experience them directly. It is social learning about mortality -- a survival strategy of extraordinary sophistication.
The Corvid Brain: Small but Mighty
How do birds with brains weighing roughly 7 to 14 grams (for crows and ravens, respectively) achieve cognitive feats that challenge those of great apes with brains 50 to 200 times larger?
Neural Architecture
The answer lies not in brain size but in neuron density. A 2016 study by Seweryna Olkowicz and colleagues, published in Proceedings of the National Academy of Sciences, revealed that bird brains contain dramatically more neurons per gram of brain tissue than mammalian brains. A crow brain weighing approximately 10 grams contains roughly 1.5 billion neurons. By comparison, a capuchin monkey brain weighing 52 grams contains approximately 1.14 billion neurons. Gram for gram, the crow brain is far more densely packed with processing power.
The Nidopallium Caudolaterale
The cognitive feats of corvids are centered in a brain region called the nidopallium caudolaterale (NCL). This region is functionally analogous to the mammalian prefrontal cortex -- the seat of executive function, working memory, planning, and flexible decision-making. Lesion studies and electrophysiological recordings have shown that NCL neurons in crows encode abstract rules, maintain information in working memory, and signal reward prediction errors in ways virtually indistinguishable from prefrontal cortex neurons in monkeys.
In 2020, Andreas Nieder and colleagues at the University of Tubingen published a study in Science demonstrating that crows possess a form of sensory consciousness. Single-neuron recordings from the NCL showed that crow neurons responded not just to the presence of a visual stimulus but to the crow's subjective perception of the stimulus -- firing differently depending on whether the crow reported "seeing" or "not seeing" a faint visual target. This was the first evidence of subjective experience correlates in a non-mammalian brain.
Comparison: Corvid vs. Primate Cognition
| Cognitive Ability | Corvids | Great Apes | Notes |
|---|---|---|---|
| Tool manufacture | Yes (New Caledonian crows) | Yes (chimpanzees, orangutans) | Corvids craft hooked tools; chimps use modified sticks and stones |
| Sequential tool use | Yes | Yes | Both solve multi-step tool problems |
| Mirror self-recognition | Yes (Eurasian magpie) | Yes (chimpanzees, orangutans, gorillas debated) | Magpies are the only non-mammal to pass |
| Future planning | Yes (ravens) | Yes (bonobos, orangutans) | Ravens plan up to 17 hours ahead |
| Episodic-like memory | Yes (scrub-jays) | Yes (great apes, rats debated) | Jays remember what, where, and when |
| Facial recognition | Yes (American crows) | Yes (chimpanzees) | Crows hold grudges for 5+ years |
| Causal reasoning | Yes (New Caledonian crows) | Yes | Both pass Aesop's fable-type tests |
| Social deception | Yes (ravens, jays) | Yes | Cache protection, false caching |
| Neuron count (forebrain) | ~1.5 billion (crow) | ~6.2 billion (chimpanzee) | Corvids pack more neurons per gram |
| Brain mass | 7-14 g | 300-500 g | Corvid brains are 30-50x smaller |
Urban Adaptation: Traffic Nut Cracking in Japan
Corvid intelligence is not confined to the laboratory. In the city of Sendai, Japan, carrion crows (Corvus corone) have developed a remarkable strategy for cracking walnuts. The crows carry hard-shelled walnuts to busy intersections and place them on the road in the path of oncoming traffic. When cars run over the nuts, the shells are crushed. The crows then wait for the traffic light to turn red and pedestrians to cross before hopping into the crosswalk to collect the exposed nut meat.
This behavior, first documented by Yoshiaki Nihei in the 1990s, involves multiple cognitive demands: understanding that cars crush nuts, selecting appropriate drop sites (intersections rather than random road segments), associating traffic signals with safe retrieval windows, and timing behavior to the rhythm of urban infrastructure. The technique has been observed spreading through crow populations in multiple Japanese cities, consistent with social learning and cultural transmission.
Corvids in Culture and Mythology
Long before scientists documented corvid intelligence, human cultures recognized something exceptional about these birds.
Odin's Ravens: Huginn and Muninn
In Norse mythology, the god Odin is accompanied by two ravens: Huginn (from hugr, meaning "thought") and Muninn (from munr, meaning "memory" or "mind"). Each day, the ravens fly across the world of Midgard, observing everything that transpires, and return to Odin's shoulders at dusk to whisper what they have learned. Odin himself reportedly said, "I fear for Huginn, that he may not return, but I worry more for Muninn." The pairing of thought and memory in corvid form is a striking echo of the cognitive abilities science would confirm centuries later.
The Trickster Crow in Native American Traditions
Across many Indigenous cultures of North America, Raven or Crow serves as a central mythological figure -- a trickster, creator, and transformer. In Pacific Northwest traditions, particularly among the Tlingit, Haida, and Tsimshian peoples, Raven is credited with stealing the sun and bringing light to the world. The trickster archetype captures the corvid qualities that are most visible even to casual observation: curiosity, opportunism, cleverness, and an apparent delight in mischief. These cultural narratives, developed over thousands of years of close observation, align remarkably well with what controlled experiments have since revealed about corvid cognition.
Conclusion
The intelligence of corvids is not a single trait but a constellation of abilities -- tool use, planning, memory, social cognition, self-awareness, and flexible problem-solving -- that together compose a cognitive profile of extraordinary depth. That this profile evolved independently from mammalian intelligence, in a brain organized along fundamentally different architectural principles, suggests that the pressures of complex social environments and demanding ecological niches can produce convergent cognitive solutions across vast evolutionary distances.
The study of corvid minds has reshaped our understanding of what brains can do, and what it means to be intelligent. As Bernd Heinrich observed after decades of fieldwork with ravens:
"Every time I think I have figured out the limits of what a raven can do, one of them does something that forces me to revise my assumptions upward."
The corvids, it seems, are still ahead of us.
References
Weir, A. A. S., Chappell, J., & Kacelnik, A. (2002). "Shaping of hooks in New Caledonian crows." Science, 297(5583), 981. -- Betty the crow bending wire into a hook tool.
Taylor, A. H., Hunt, G. R., Holzhaider, J. C., & Gray, R. D. (2007). "Spontaneous metatool use by New Caledonian crows." Current Biology, 17(17), 1504-1507. -- Sequential tool use in corvids.
Kabadayi, C., & Osvath, M. (2017). "Ravens parallel great apes in flexible planning for tool-use and bartering." Science, 357(6347), 202-204. -- Future planning and bartering in ravens.
Prior, H., Schwarz, A., & Gunturkun, O. (2008). "Mirror-induced behavior in the magpie (Pica pica): Evidence of self-recognition." PLoS Biology, 6(8), e202. -- First non-mammal to pass the mirror self-recognition test.
Clayton, N. S., & Dickinson, A. (1998). "Episodic-like memory during cache recovery by scrub jays." Nature, 395(6699), 272-274. -- Foundational study of what-where-when memory in jays.
Marzluff, J. M., Walls, J., Cornell, H. N., Withey, J. C., & Craig, D. P. (2010). "Lasting recognition of threatening people by wild American crows." Animal Behaviour, 79(3), 699-707. -- The Seattle mask experiment on facial recognition.
Olkowicz, S., et al. (2016). "Birds have primate-like numbers of neurons in the forebrain." Proceedings of the National Academy of Sciences, 113(26), 7255-7260. -- Neuron density in avian versus mammalian brains.
Nieder, A., Wagener, L., & Rinnert, P. (2020). "A neural correlate of sensory consciousness in a corvid bird." Science, 369(6511), 1626-1629. -- Evidence of subjective sensory experience in crows.
Heinrich, B. (1999). Mind of the Raven: Investigations and Adventures with Wolf-Birds. HarperCollins. -- Comprehensive naturalist account of raven cognition and behavior.
Swift, K. N., & Marzluff, J. M. (2015). "Wild American crows gather around their dead to learn about danger." Animal Behaviour, 109, 187-197. -- Systematic study of crow funeral behavior.
Frequently Asked Questions
Can crows really use tools, and how sophisticated is their tool use?
Yes, crows -- particularly New Caledonian crows -- are among the most accomplished tool users in the animal kingdom outside of great apes. They craft hooked stick tools from twigs by stripping leaves and bending the tip, manufacture stepped-cut tools from Pandanus palm leaves with standardized designs, and demonstrate sequential (meta) tool use by employing one tool to retrieve another tool to reach food. In laboratory settings, a New Caledonian crow named Betty spontaneously bent a straight piece of wire into a hook to retrieve a food bucket from a tube, a behavior never previously observed in the wild for that material.
How do ravens demonstrate problem-solving and planning abilities?
Ravens have demonstrated planning abilities that rival those of great apes. In studies led by Mathias Osvath at Lund University, ravens selected and saved a tool they would need to open a puzzle box up to 17 hours in the future, ignoring immediate but less valuable food rewards. Ravens also succeed at bartering, choosing a token over an immediate food reward because they have learned the token can be exchanged for a better reward later. They solve multi-step mechanical puzzles, engage in deliberate play behavior such as sliding down snowy rooftops, and have been observed apparently deceiving other ravens by pretending to cache food in one location to distract competitors while hiding it elsewhere.
Can crows really recognize and remember human faces?
Yes, crows have extraordinary facial recognition abilities. In a landmark experiment at the University of Washington in Seattle, researchers led by John Marzluff wore specific rubber caveman masks while trapping and banding crows on campus. Years later, crows still scolded and dive-bombed anyone wearing the trapping mask, even though the original captured crows had not been disturbed again. The grudge persisted for over five years and spread to crows that had never been trapped -- adults apparently taught their offspring and flock-mates to recognize the threatening face. Brain imaging using PET scans later confirmed that crows viewing the dangerous mask showed activation in brain areas associated with fear and memory.