Tag: comparative cognition

  • Corvid Intelligence: Tool Use, Planning, and Why Crows Hold Funerals

    A New Caledonian crow named Betty, in a 2002 experiment at Oxford, bent a straight piece of wire into a hook to retrieve food from a tube. She had never seen wire before. She wasn’t trained to bend it. She looked at the problem — food at the bottom of a vertical tube, a straight wire that couldn’t reach it — and manufactured a tool from a novel material, on the spot, to solve a problem she’d never encountered. That single observation kicked off two decades of research into corvid cognition that has systematically demolished the assumption that complex intelligence requires a primate brain, a mammalian cortex, or 300 million years of shared evolutionary history with humans.

    Corvids — the family that includes crows, ravens, jays, magpies, and jackdaws — have brains the size of a human thumb. They have no neocortex, the structure that in mammals is responsible for the cognitive functions we associate with intelligence: planning, reasoning, abstract thought, self-awareness. They produce comparable cognitive outputs using entirely different neural architecture, which means either intelligence is less dependent on specific brain structures than neuroscience assumed, or corvids evolved their way to the same destination through a route nobody predicted.

    https://open.spotify.com/show/7kYYn1bsA68hcfeT1YYehh?si=ae8656690bc643fb

    Tool use: not the party trick it looks like

    New Caledonian crows are the corvid species with the most sophisticated tool use, and the research on them has gone well beyond “crow uses stick to get food.” In wild populations on the Pacific island of New Caledonia, these crows manufacture tools from pandanus leaves by tearing them into specific shapes — stepped, tapered, or wide — to probe insect larvae from tree bark. The tool shapes are consistent within populations and vary between populations, which means the techniques are culturally transmitted rather than genetically encoded. A young crow learns to make tools by watching older crows. If the older crows die before transmitting the technique, the knowledge disappears. This is culture — the same mechanism that transmits human skills across generations — operating in a bird with a brain that weighs 14 grams.

    In laboratory settings, the cognitive demands of corvid tool use have been tested with increasing rigor. Gruber and colleagues, in experiments published in Current Biology, presented New Caledonian crows with metatool problems: multi-step tasks where one tool must be used to obtain another tool, which is then used to reach food, with each stage of the problem out of sight of the others. The crows had to mentally represent the location and identity of tools and apparatuses they couldn’t see while planning and executing a sequence of tool behaviors. They succeeded — maintaining working representations of objects across spatial separation and planning one to two steps ahead. The researchers concluded that New Caledonian crows can use mental representations to solve sequential problems, a capacity previously attributed only to humans and great apes.

    A 2020 study in Proceedings of the Royal Society B pushed further. Crows learned a temporal sequence: they were shown a baited apparatus, given a choice of five objects five minutes later, and given access to the apparatus ten minutes after that. At test, the crows selected the correct tool for the specific apparatus they’d been shown — choosing the right tool for the right future task while ignoring previously useful tools and a low-value food item. The study’s conclusion: New Caledonian crows plan for specific future tool use. This capacity — selecting a tool now for a task that will occur later, based on a mental representation of what that future task requires — was previously considered a defining feature of human intelligence. Corvids and humans shared a common ancestor over 300 million years ago. Whatever cognitive machinery the crows are using, they evolved it independently.

    Funerals: danger assessment, not grief

    When a crow dies, other crows gather. They emit alarm calls — loud, repetitive scolding vocalizations — that attract additional crows to the scene. Dozens of birds may congregate around the body, observing it from nearby perches, sometimes flying down to inspect it, sometimes sitting in silence. To a human observer, it looks like mourning. The scientific explanation is more interesting than mourning.

    Kaeli Swift, a behavioral ecologist at the University of Washington working under corvid cognition researcher John Marzluff, conducted a two-year experiment across over a hundred sites in Washington State. She established feeding stations to attract local crows, then introduced a dead crow (a taxidermied specimen) while a masked human volunteer stood nearby. The crows responded to the dead crow with alarm calls and gathering behavior. More importantly, they subsequently avoided feeding at that location — and they associated the masked person with danger, responding with alarm calls when that person appeared again, even without the dead crow present. The crows learned from the death scene. They identified a potential threat (the person near the dead crow), memorized the threat’s face, and modified their behavior to avoid the area and the individual. Weeks and months later, they still recognized and responded to the mask.

    Crows respond far more strongly to dead crows than to dead birds of other species. They largely ignore dead pigeons, robins, or other non-corvid birds placed in their territories but react intensely to dead members of their own species. Some studies suggest they respond more strongly to familiar individuals than to unfamiliar crows, indicating they may recognize specific community members even in death.

    The “funeral” is not a ceremony. It’s a threat assessment protocol. The crows are investigating the scene to determine what killed the dead crow, whether that threat persists, and how to avoid it. The alarm calls broadcast the danger to the wider community. The subsequent avoidance behavior encodes the lesson into the population’s behavioral repertoire. Marzluff’s research demonstrated that crows can remember human faces that posed a threat for years — and they transmit this knowledge to crows that weren’t present for the original event. A crow that never saw the masked person holding a dead crow will nonetheless scold that person if other crows in the community do, because the social alarm response propagates through the group.

    The behavioral function is pragmatic: collective intelligence applied to mortality data. The emotional dimension — whether crows experience something analogous to grief — remains scientifically unresolvable. Swift’s position is candid: she believes crows have emotional intelligence, but testing that scientifically is impossible because there’s no way to access what’s happening at an emotional level inside an animal’s brain. What’s measurable is the behavioral output: crows process death, learn from it, remember the context, and share the information. Whether they feel anything while doing it is a question the methodology can’t answer.

    What corvid brains do differently

    The corvid brain lacks a neocortex. In mammals, the neocortex is the seat of higher cognitive function — the structure that expanded dramatically in primates and reached its maximum density in humans. Corvids achieve comparable cognitive outputs using a structure called the pallium, which is organized differently from the mammalian cortex but performs analogous functions. The neuron density in the corvid pallium is remarkably high relative to brain volume — corvid brains pack more neurons per gram than most mammalian brains.

    A 2025 paper in Animal Cognition by Veit and colleagues explored the “dimensions of corvid consciousness” — a research framework asking not whether corvids are conscious but what aspects of consciousness their neural architecture could support. The paper argues that corvid brains process sensory information, maintain working memory, and generate flexible behavioral responses through neural pathways that are structurally distinct from but functionally analogous to mammalian circuits. A German neurobiologist trained two crows — Glenn and Ozzy — to peck at “yes” or “no” targets to indicate whether they had detected a faint light, demonstrating analytical introspection: the crows reported on their own perceptual states, a capacity associated with subjective experience.

    The convergent evolution angle is what makes corvids matter for neuroscience rather than just for animal behavior. If complex cognition can evolve independently in a brain that is structurally unrelated to the primate brain, then intelligence is not a property of a specific neural architecture. It’s a property of certain computational principles — neuron density, connectivity patterns, feedback loops — that can be instantiated in multiple biological substrates. The corvid brain is evidence that there is more than one way to build a mind, and that the way mammals did it is not the only way it can be done.

    Corvids sit alongside octopuses as the strongest natural evidence that intelligence is convergent rather than unique. We cover corvid cognition alongside cuttlefish camouflage, electroreception, and the full landscape of how animal brains solve problems humans assumed required human brains across our Neurozoology course — including why a 14-gram brain that last shared an ancestor with yours 300 million years ago can plan for the future, manufacture tools from materials it’s never seen, and hold a funeral that’s more operationally useful than most of ours.

  • The CIA’s Cat and Pigeon Spy Programs: The Strangest Operations in Intelligence History

    In the early 1960s, the CIA’s Directorate of Science and Technology surgically implanted a microphone in a cat’s ear canal, embedded a three-quarter-inch radio transmitter near the base of its skull, wove a fine wire antenna through its fur all the way to its tail, and placed a power pack in its abdomen. Additional wires connected to the cat’s brain allowed handlers to detect when the animal was hungry or sexually aroused, and to override those urges so the cat wouldn’t abandon its mission to chase a pigeon or find a mate. The project took five years to develop and cost an estimated $20 million. Then they put the cat in a van, drove it to a location near the Soviet embassy in Washington, D.C., and released it to eavesdrop on two men sitting on a park bench.

    According to former CIA officer Victor Marchetti, the cat waddled across the street and was immediately hit and killed by a taxi. Twenty million dollars, five years of surgical development, and the most expensive domestic animal in American intelligence history, dead on contact with reality. A former CIA technical officer named Robert Wallace later disputed this, claiming the cat survived and the project was cancelled for other reasons. The CIA’s own website says the cat was treated humanely and the equipment was removed when the program ended. Whether the cat died under a taxi or retired to a quiet life remains, appropriately, classified.

    The project was code-named Acoustic Kitty. It was declassified in 2001. The closing memorandum, dated 1967 and still heavily redacted, concluded that while the CIA had proven “cats can indeed be trained to move short distances”—described without irony as “a remarkable scientific achievement”—”the environmental and security factors in using this technique in a real foreign situation force us to conclude that, for our purposes, it would not be practical.”

    Anyone who has ever owned a cat could have told them that for free.

    Why animals seemed like a good idea

    The logic behind CIA animal programs wasn’t insane. It was 5 percent good idea and 95 percent bad execution, repeated across multiple species with consistent results. The core insight was genuine: animals can access places humans can’t, and they do it without triggering suspicion. A stray cat near an embassy is invisible. A pigeon on a windowsill is furniture. A raven on a ledge is scenery. In an era when electronic surveillance devices were the size of textbooks and human agents were tailed by KGB counterintelligence teams, the idea of using a biological platform that could move freely through denied areas had real appeal.

    The CIA’s own historical review, published on the agency’s website under the title “Natural Spies: Animals in Espionage,” is remarkably candid about the programs. The agency acknowledges that “many of the animal programs studied by CIA were never deployed operationally—or failed for a variety of technical, logistical, or behavioral reasons.” The candor is unusual for an organization that typically lets misconceptions stand rather than correcting them. The fact that they published the review suggests they’ve decided the programs are more charming than embarrassing at this distance.

    The pigeons that actually worked

    Project Tacana, the CIA’s pigeon camera program, was the animal operation that came closest to producing operational intelligence. During the 1970s, the agency trained pigeons to carry miniature cameras weighing roughly 35 grams and fly over Soviet military installations—shipyards, naval bases, and other targets that were difficult to photograph from satellites or high-altitude aircraft.

    The theory was sound for a specific technical reason: a pigeon flying at low altitude could capture higher-resolution photographs than a spy satellite orbiting hundreds of miles above the target. Satellite imagery in the 1970s was good enough to identify buildings and vehicles but often lacked the resolution to read markings, count components, or assess equipment condition. A pigeon at rooftop height with a miniature camera could, in principle, deliver imagery that filled that gap.

    Tests showed that approximately half of the 140 photographs taken during trials achieved good image quality—a success rate that was encouraging enough to continue development but insufficient to justify full operational deployment. The program faced the same fundamental problem as Acoustic Kitty: you could get the animal to the right general area, but you couldn’t guarantee it would do what you wanted once it got there. Pigeons are trainable—far more so than cats—but they’re navigating by instinct and training, not by mission briefing. They have no concept of which building is the target or which angle produces the most useful photograph. The camera fires on a timer or by altitude trigger, and the resulting images are whatever the pigeon happened to be flying over.

    The program never became fully operational. Satellite imagery improved, the U-2 and SR-71 reconnaissance aircraft covered much of the gap, and the era of miniaturized unmanned drones eventually made biological platforms obsolete for aerial surveillance. But the pigeon program came closer to working than most people realize, and the CIA’s acknowledgment that the concept was sound—even if the execution was impractical—suggests the agency viewed pigeons as a near-miss rather than a failure.

    The rest of the menagerie

    The CIA tested ravens for precision delivery of surveillance devices. Ravens were trained to carry miniaturized eavesdropping equipment and deposit it on window ledges using specially designed carrying mechanisms. In at least one operation, a raven successfully delivered a bugging device to a European target—though no usable audio was ever captured. The delivery worked. The intelligence didn’t.

    Under MKUltra Subproject 94, the agency implanted electrodes in dogs’ brains to create remote-controlled animals that could be directed to run, turn, and stop via radio signals. Six dogs achieved “field operational” status, meaning they could be reliably directed through basic movement commands. The program was never deployed operationally, and the ethical dimensions of surgically implanting brain electrodes in dogs for remote control are exactly as uncomfortable as they sound.

    The Insectothopter was a mechanical dragonfly—a miniaturized unmanned aerial vehicle designed to carry a listening device. It was selected after an initial bumblebee design proved too erratic in flight. The dragonfly could fly 200 meters in 60 seconds, guided by a laser beam, but proved inoperable in crosswinds above five miles per hour. Charlie and Charlene were robotic catfish developed by the CIA’s Office of Advanced Technologies and Programs to study unmanned underwater vehicle technology—robot fish designed for aquatic surveillance.

    What the programs actually tell us

    The pattern across all of these operations—cat, pigeon, raven, dog, dragonfly, catfish—is consistent and diagnostic. The CIA could build the technology. Miniaturizing transmitters, embedding recording devices, engineering mechanical insects—the engineering was ahead of its time. What they couldn’t do was solve the interface between human intent and animal behavior. A cat with a working transmitter in its skull is still a cat. It will chase a bird, wander toward food, lose interest in the park bench, or walk into traffic. The technology was the easy part. Biology was the hard part, and biology won every time.

    A 2023 comparative cognition study quantified the problem: cats made “considerably fewer choices than dogs in laboratory environments, and their tendency to make a choice declined during trials.” The CIA discovered this empirically, at a cost of $20 million, six decades before the paper was published. Cats evolved as solitary ambush predators whose attention is stimulus-driven, not command-driven. Their brains prioritize potential prey over instructions. Asking a cat to eavesdrop on a Soviet diplomat instead of chasing a squirrel is asking the cat to override 30 million years of predatory evolution for a food pellet. The cat’s answer, delivered at a behavioral level that no amount of surgical modification could change, was no.

    The pigeon program came closest because pigeons have social structures and can be trained through operant conditioning to fly specific routes and return to specific locations—behaviors that align with their natural homing instincts. Dogs performed better than cats because their social cognition is command-oriented rather than stimulus-oriented. Ravens succeeded at precision delivery because corvids are problem-solvers that can learn sequential tasks. The CIA’s animal programs, read as a body of work, are an accidentally rigorous experiment in comparative cognition: which species can be directed to perform tasks that conflict with their natural behavioral repertoire, and what determines the answer?

    The answer, demonstrated across two decades of classified research, is that animals with social structures and reward-oriented learning systems (dogs, pigeons, ravens) outperform solitary predators (cats) at human-directed tasks—but none of them can be reliably directed to perform context-dependent intelligence operations that require judgment, sustained attention, and goal persistence in uncontrolled environments. The technology worked. The biology was not negotiable. And a taxi, if Marchetti is to be believed, delivered the final verdict.

    We cover the CIA’s animal programs alongside navy dolphins, anti-poaching dogs, and the full history of animals deployed in human conflicts across our Animal Heroes course—including why the most expensive spy the CIA ever built had whiskers, a tail, and absolutely no interest in Soviet diplomats.

  • Fish That Use Tools: The Species That Shattered Assumptions About What Fish Can Do

    In 2006, a diver named Scott Gardner was ascending from an 18-meter dive in the Keppel Islands region of the Great Barrier Reef when he heard a cracking noise. He looked over and saw a blackspot tuskfish hovering above a sand patch, holding a cockle shell in its jaws. The fish was rolling onto its side and slamming the shell against a rock—alternating left and right blows, aimed at the pointed section of the rock for maximum impact—until the shell cracked open. Scattered around the rock were broken shells from previous meals. This wasn’t an isolated event. It was a feeding station. The fish had a preferred anvil, and it had been using it long enough to accumulate a midden of shattered prey.

    Gardner photographed the sequence. The images were published in Coral Reefs in 2011, and the paper posed a question in its title that a generation of biologists had considered already answered: “Tool use in the tuskfish?” The question mark was doing heavy lifting. By the definitions that Jane Goodall had established—the use of an external object as a functional extension of mouth or hand in the attainment of an immediate goal—the tuskfish was using a tool. The external object was the rock. The goal was food. The behavior was deliberate, sequential, and repeated. The only reason anyone hesitated to call it tool use was that the animal doing it was a fish.

    Why this matters more than it should

    For most of the history of comparative cognition, the assumption was straightforward: fish are simple. They operate on instinct. They have small brains, short memories, and minimal behavioral flexibility. Tool use—the cognitive capacity to identify an external object, recognize its functional utility, and deploy it to achieve a goal—was reserved for the clever animals: primates, corvids, maybe elephants and sea otters. The hierarchy was implicit and rarely questioned. Mammals and birds think. Fish react.

    The tuskfish broke that hierarchy not by being unusually smart but by doing something that forced the definition of intelligence to either expand or become incoherent. If tool use is a marker of advanced cognition, and a fish uses tools, then either the fish is cognitively advanced or tool use isn’t the marker we thought it was. Both conclusions are uncomfortable for the framework that produced the hierarchy in the first place.

    The discomfort deepened as evidence accumulated. The tuskfish observation wasn’t a one-off. A 2025 study led by Macquarie University, published in Coral Reefs, documented anvil use in five species of Halichoeres wrasses across the western Atlantic—the first evidence of tool use for three of those species and the first video evidence for the other two. Through a citizen science initiative, researchers gathered 16 new observations of wrasses deliberately picking up hard-shelled prey and smashing them against rocks, corals, and other hard surfaces. The findings extended the known range of fish tool use from the Indo-Pacific to the Atlantic and from a handful of isolated observations to a pattern distributed across an entire fish family spanning 50 million years of evolution.

    Culum Brown, head of the Fish Lab at Macquarie University and one of the foremost researchers on fish cognition, suggested that wrasses may be fishes’ answer to primates among mammals and corvids among birds—a lineage with a disproportionate number of examples of cognitive complexity relative to the broader group. Researchers at the Paris-Saclay Institute of Neuroscience found that wrasses have a larger telencephalon and forebrain region compared to other teleost fish, including a substantially enlarged inferior lobe—a brain structure with no direct analog in mammals or birds—that shows unique connectivity to the pallium, a region already linked to higher-order cognition in other animals.

    The physics problem fish solved

    The reason tool use is rare in fish isn’t necessarily cognitive. It’s physical. Water is 800 times denser than air. Try swinging a hammer underwater and you’ll understand the constraint immediately. The momentum required to crack a shell with an object held in your mouth, while suspended in a fluid medium that resists rapid movement in every direction, is orders of magnitude harder to generate than doing the same thing on land. A chimpanzee cracking a nut with a rock is operating in an environment that cooperates with the physics of impact. A fish is operating in an environment that actively resists it.

    The tuskfish solved this by inverting the relationship: instead of swinging a tool against a stationary target, it swings the target against a stationary tool. The rock is the anvil, fixed in the substrate. The shell is the projectile, gripped in the fish’s jaws and slammed against the anvil through rapid body rotation. This isn’t just tool use. It’s tool use adapted to an environment where the conventional approach—wielding a hammer—is physically impossible. The fish engineered a workaround.

    The sixbar wrasse took the same approach in captivity. Given food pellets too large to swallow and too hard to break with its jaws, the wrasse carried the pellets to a rock in its aquarium and smashed them. The researcher who observed it, Łukasz Paśko at the University of Wrocław, watched the wrasse perform the behavior 15 times and described it as “remarkably consistent” and “nearly always successful.” The behavior only appeared after many weeks in captivity, suggesting the fish learned it through individual experience rather than instinct—it tried other approaches first, found them inadequate, and developed a new strategy.

    Anvils, middens, and long-term site fidelity

    A 2023 study on graphic tuskfish in New Caledonia found that specific anvils showed evidence of being used by one or more tool-using fish for years. The anvils accumulated debris. Other fish species learned to recognize the visual and auditory cues of tool use in progress—the body movements, sand clouds, and the “clack” sound of shell hitting rock—and gathered as scavengers. In 94 percent of observed tool-use events, attendant fish from six different families showed up to pick up fragments: surgeonfishes, triggerfishes, butterflyfishes, wrasses, angelfishes, and damselfishes. The tuskfish’s tool use had created a micro-ecosystem around its feeding station—a social and ecological structure generated by a fish banging a clam on a rock.

    The wrasses also showed flexibility in their tool use, selecting different types of anvils for different prey and sometimes switching anvils mid-session when the first choice wasn’t working. This isn’t stereotyped behavior—the kind of fixed action pattern that “instinct” describes. It’s decision-making under uncertainty, adapted in real time to the properties of the specific prey item and the available tools.

    The archerfish problem

    The wrasses aren’t the only fish that complicate the tool-use question. Archerfish—four-inch tropical marksmen from estuaries and mangroves between India and the Philippines—hunt by shooting precisely aimed jets of water at insects sitting on vegetation above the water’s surface, knocking them into the water where they can be eaten. The archerfish accounts for refraction at the water’s surface, adjusts for the target’s distance and position, and can hit prey up to three meters above the waterline. Researchers have demonstrated that archerfish can learn to recognize human faces and can be trained to hit specific targets, showing a capacity for visual discrimination and precision that wouldn’t be out of place in a primate cognition lab.

    Whether the water jet constitutes a “tool” depends on how strictly you define the term. The archerfish isn’t wielding an external object—it’s producing a projectile from its own body, more analogous to a spider’s web than a chimpanzee’s stick. But the functional outcome is the same: an organism using a mechanism beyond its own body to obtain food that would otherwise be inaccessible. The boundary between tool and technique blurs when the organism in question can’t hold anything in its hands, because it doesn’t have hands.

    What 600 species of wrasse haven’t told us yet

    There are over 600 species of wrasses worldwide. The Macquarie University team’s citizen science initiative is explicitly calling for divers and snorkelers to report observations of anvil use, acknowledging that the documented cases almost certainly represent a fraction of the actual prevalence. Brown put it directly: “For a long time, tool use was thought to be exclusive to primates and birds. We are still far from knowing how many species of wrasses use tools.” The field of fish cognition itself is young—69 percent of published studies used captive-reared subjects, only 9 percent conducted experiments on wild fish in their natural environment—meaning we’ve been studying fish cognition primarily by watching captive fish in artificial environments and then drawing conclusions about what fish can’t do.

    The tuskfish cracking a cockle on a rock doesn’t prove that fish are as smart as chimps. It proves that the cognitive hierarchy we built—mammals on top, birds below them, everything else at the bottom—was a projection of our anatomy onto our definition of intelligence. An animal that solves the same problem a primate solves, in a medium 800 times denser than air, without hands or arms, using a body plan that hasn’t shared a common ancestor with primates in over 400 million years, isn’t failing to be smart. It’s being smart in a way we weren’t looking for.

    We cover fish cognition alongside dolphin communication, elephant memory, and primate social intelligence across our Animal Culture & Knowledge course—including why the most important discoveries in comparative cognition keep coming from the species we assumed had nothing to teach us.