Animal intelligence is far richer, stranger, and more widespread than most people assume. Crows craft hooks from twigs, elephants comfort distressed companions, bees weigh their own uncertainty before making a choice, and a parrot named Alex could add small numbers using spoken English words. The deeper researchers look, the more they find that sophisticated mental abilities are not confined to a handful of brainy mammals. They pop up across the animal kingdom in species separated by hundreds of millions of years of evolution, suggesting that complex minds have arisen independently many times over.
Why There Is No Single Ladder of Animal Intelligence
For centuries, Western science arranged animals on a kind of mental ladder, from “simple” creatures at the bottom to humans at the top. That framework, sometimes called the scala naturae, has been thoroughly challenged. A critique in the Journal of Comparative Psychology argued that this approach is anthropocentric and that a sequence of ranked grades should not be taken to represent actual stages in the evolution of specific behaviors or brain structures.1PubMed. The Scala naturae revisited: evolutionary scales and anagenesis in comparative psychology In practice, different species evolved to solve different problems in different environments. An octopus hunting crabs across a reef faces nothing like the challenges of a crow extracting grubs from deadwood, yet both tasks demand flexible problem-solving.
Part of the difficulty is that humans can only perceive the world through human senses. Evolution sculpts each species’ brain around its own sensory toolkit, and those toolkits differ so dramatically that it is genuinely hard for us to grasp how another animal experiences its surroundings.2PubMed Central. The Visual Umwelt of primates and Hippocampal Representations of Space A bat navigating by echolocation, a shark sensing electrical fields, or a bee reading polarized light are all doing something cognitively demanding, but in sensory dimensions we barely have language for. When researchers design intelligence tests, they inevitably favor skills that make sense to a primate. That bias has led to decades of underestimating what non-primates can do.
Tools, Hooks, and Multi-Step Planning
Tool use was once considered uniquely human. Then Jane Goodall watched chimpanzees fish for termites with sticks, and the definition of what counts as “human” had to shift. Today, chimpanzees are known to use a wide range of tools, and those traditions are culturally transmitted. In one wild community, a novel tool behavior called moss-sponging appeared in 2011 and spread first through social proximity and then through mother-to-offspring learning, following a pattern of horizontal and then vertical cultural transmission.3PubMed Central. Kin-based cultural transmission of tool use in wild chimpanzees When female chimpanzees immigrate into a new community, they adopt that group’s tool preferences, even if they had different personal habits before.4Philosophical Transactions of the Royal Society B. Primate archaeology reveals cultural transmission in wild chimpanzees (Pan troglodytes verus) Social customs like the grooming handclasp show the same pattern: older and higher-ranking individuals influence their partners’ style, and over time individuals converge on the group norm, driven by something that looks a lot like conformity bias.5PubMed Central. Biased cultural transmission of a social custom in chimpanzees
But some of the most impressive tool users are not primates at all. New Caledonian crows in the wild craft hooked tools through a consistent three-step process: selecting raw material, trimming it, and then sculpting the hook shape over an extended period.6PubMed Central. The crafting of hook tools by wild New Caledonian crows In the lab, these crows go further. They can solve metatool problems, tasks that require using one tool to obtain a second tool to get food. In one experiment, crows had to use a stick to retrieve a stone from a tube while ignoring a distractor, then use that stone to release food from a separate apparatus. They could also do the reverse: use a stone to get a stick, then use the stick to reach food.7PubMed Central. New Caledonian Crows Use Mental Representations to Solve Metatool Problems This is not trial-and-error fumbling. The crows appeared to hold a mental representation of the goal and work backward through the steps needed to reach it.
Mirror Tests, Self-Awareness, and the Limits of Both
The mirror self-recognition test has been the go-to measure of self-awareness since the 1970s. The idea is simple: place a mark on an animal’s body where it can only be seen in a mirror, and see whether the animal uses the reflection to investigate the mark on itself. Great apes pass. Dolphins pass. And in a landmark study, an Asian elephant named Happy repeatedly touched a visible mark on her own head while looking in a large mirror, in a clear parallel to the responses seen in apes and dolphins.8PubMed Central. Self-recognition in an Asian elephant
More surprising entrants have joined the list. Cleaner fish, small reef-dwelling wrasses, not only pass the mark test but appear to build a mental image of their own body size afterward, adjusting their behavior in ways that suggest they know how big they are.9PubMed Central. Cleaner fish with mirror self-recognition capacity precisely realize their body size based on their mental image That result has stirred real debate. Either self-recognition is more widespread and ancient than anyone expected, or the mirror test is measuring something more basic than full-blown self-awareness. Either way, it complicates the tidy story that self-awareness belongs exclusively to big-brained mammals and birds.
Seeing Through Another’s Eyes
Social intelligence may be harder to test than tool use, but the evidence keeps accumulating. One of the building blocks of understanding other minds is visual perspective taking: grasping that another individual can or cannot see something. In a carefully controlled experiment, subordinate long-tailed macaques chose to grab food that was hidden from a dominant individual rather than food the dominant could see. A one-way mirror between the competitor and the food prevented the monkeys from simply reading body language, so their choice had to be based on an understanding of what the dominant could perceive. The researchers called it the first unequivocal evidence of visual perspective taking in a monkey species.10PubMed. Long-tailed macaques (Macaca fascicularis) understand what conspecifics can see in a competitive situation
A related capacity is a sense of fairness. Capuchin monkeys trained to exchange tokens for food rewards reacted negatively when they saw a partner receive a better reward for the same task. The reaction could not be explained by simple frustration or greed, because the controls ruled those out. The strongest protests came when the subject had to put in significant effort while watching its partner get the better deal.11PubMed Central. Inequity responses of monkeys modified by effort This kind of inequity aversion is widespread in cooperative species, especially those that cooperate outside the bonds of kinship and mating. Apes go even further, sometimes equalizing outcomes to their own disadvantage, apparently to keep partners satisfied and preserve future cooperation.12PubMed Central. Evolution of responses to (un)fairness
Empathy, too, is not a human monopoly. At an elephant camp in Thailand, researchers found that when one elephant became distressed, nearby elephants increased physical contact and vocalizations directed at the upset individual. Bystanders also matched the distressed elephant’s emotional state, a hallmark of emotional contagion. The researchers classified this behavior alongside the consolation responses documented in apes, suggesting that empathic capacities evolved independently in these two lineages.13PubMed Central. Asian elephants (Elephas maximus) reassure others in distress
Words, Alarms, and Numbers
Human language is unique in its open-ended grammar, but the components of meaningful communication show up across many species. The classic example is the vervet monkey. Vervets give acoustically distinct alarm calls for different predators, and each call triggers a different escape behavior. A leopard alarm sends them into trees; an eagle alarm makes them scan the sky; a snake alarm makes them look at the ground. When researchers played back recorded calls in the absence of any predator, the monkeys responded appropriately to the call type rather than to any contextual cue, confirming that the calls function as labels for different categories of danger.14Animal Behaviour. Vervet monkey alarm calls: Semantic communication in a free-ranging primate More recent quantitative analysis confirmed that these calls are acoustically separable with very high accuracy, correctly classified in over 93% of cases for males and nearly 99% for females.15Scientific Reports. Vervets revisited: A quantitative analysis of alarm call structure and context specificity
On the numerical side, an African grey parrot named Alex demonstrated abilities that compare favorably to those of chimpanzees and young children. Alex could label sets of up to six items using spoken English words, understood those labels as abstract representations of quantity, and had a concept analogous to zero.16PubMed. Grey parrot numerical competence: a review Later testing pushed these abilities further: Alex could add two Arabic numerals or three sequentially presented collections and answer “How many total?” vocally, in the absence of the previously viewed items. His accuracy was statistically significant and suggested addition abilities comparable to those of nonhuman primates.17PubMed. Further evidence for addition and numerical competence by a Grey parrot (Psittacus erithacus) Alex died before the full battery could be completed, but the data he left behind changed conversations about what a bird brain can do.
Remembering the Past, Anticipating the Future
Episodic memory, the ability to recall specific events from the past and use that information flexibly, was long assumed to require the kind of conscious experience only humans have. Western scrub-jays challenged that assumption with an elegant food-caching experiment. The jays cached both peanuts and waxworms in separate locations at different times. When the worms had been cached recently, the jays retrieved them first, since worms are preferred. But when enough time had passed for the worms to have degraded, the jays switched to retrieving peanuts instead. They remembered what they had cached, where they had put it, and how long ago the caching happened, all integrated into a single retrieval decision.18Current Biology. How Intelligent Are Animals? A Look at Their Minds – Section: Do Insects Have Cognitive Maps?
Navigation is another domain that demands sophisticated memory. Honey bees can set a straight course homeward from locations all around their hive, even from sites far outside the visual range of any familiar landmark. Researchers found that bees meet two criteria for a map-like spatial memory: they can orient from arbitrary locations and choose between at least two goals.19PubMed Central. Honey bees navigate according to a map-like spatial memory A more recent study of bumblebees showed that vectors learned during path integration can be stored in long-term memory, that multiple such vectors can be held in parallel, and that they can be recalled at a familiar location for homeward navigation.20PubMed Central. Parallel vector memories in the brain of a bee as foundation for flexible navigation This vector-based strategy may allow an insect brain with roughly a million neurons to achieve navigational feats that in mammals rely on far larger neural architecture. Whether this truly constitutes a cognitive map in the full mammalian sense remains debated, but the functional output is strikingly similar.
Small Brains, Big Surprises
Invertebrates have long been dismissed as simple automatons, but the evidence against that view has become overwhelming. Cephalopods, the group including octopuses, cuttlefish, and squid, are considered the most cognitively advanced invertebrates. Their nervous systems rival those of vertebrates in relative size and complexity, and their behavioral repertoires are remarkably flexible. A South African common octopus, for instance, hunts more than 35 different prey species, adjusting its strategy for crabs, fish, other cephalopods, and even members of its own species.21Trends in Ecology & Evolution. Cephalopods: An Evolution of Intelligence Octopuses also show intriguing sleep behavior: during active sleep states they display shifting camouflage patterns and changes in basal rhythms while remaining unresponsive to outside stimuli, leading some researchers to speculate about possible dream-like states analogous to those in mammals.22bioRxiv. Abnormal behavioral episodes associated with sleep and quiescence in Octopus insularis: Possible nightmares in a cephalopod?
Insects are even more startling because their brains are so tiny. Honey bees trained on a discrimination task opted out of difficult trials more often than easy ones, and opting out improved their overall success rate. They could also transfer this opt-out strategy to a completely new task, suggesting a general rule rather than rote conditioning. The researchers noted that the bees’ performance was comparable to that of primates in a similar experimental setup.23PubMed Central. Honey bees selectively avoid difficult choices Bumblebees show the same pattern: they dynamically adjust their opt-out behavior based on how hard a decision is, settle for a smaller guaranteed reward to avoid errors, and even pay a reward cost to seek additional information when uncertain. This strategy transferred across sensory modalities and task types without any retraining.24bioRxiv. Uncertainty-Guided Decision-Making in Bumble Bees A brain of about a million neurons appears to support what the researchers called a domain-general uncertainty-monitoring policy.
And then there is play. Bumblebees given access to small wooden balls rolled them around repeatedly with no food reward, no training incentive, and no obvious survival benefit. The researchers argued that this behavior meets established criteria for play and has actual hedonic value, lending support to the idea of positive emotional states in insects.25Animal Behaviour. Do bumble bees play? If a bumblebee can experience something like enjoyment, the question of sentience stretches much further across the tree of life than most people are comfortable with.
Why Such Different Brains Produce Similar Abilities
One of the most striking findings in animal cognition research is that very different brain architectures arrive at comparable cognitive solutions. Corvids and apes last shared a common ancestor over 300 million years ago, yet both groups show causal reasoning, tool use, and flexible social behavior. Researchers have argued that these shared cognitive tools evolved independently to solve similar ecological and social problems.26PubMed. The mentality of crows: convergent evolution of intelligence in corvids and apes A broad review of intelligence across the animal kingdom found that complex brains and high-level cognition have evolved multiple times: in certain insect groups, in octopuses, in cichlid fish, in corvid and parrot birds, and in cetaceans, elephants, and primates. In every case, the cognitive heavy lifting happens in highly ordered associative networks, whether those networks sit in insect mushroom bodies, an octopus vertical lobe, or a primate cerebral cortex.27PubMed Central. Convergent evolution of complex brains and high intelligence
Part of what makes birds so cognitively capable despite their small skulls is neuron density. Parrot and songbird brains pack roughly twice as many neurons per gram as primate brains of the same mass. Corvids and parrots also devote a higher fraction of their total brain neurons to the pallium, the region analogous to the mammalian cortex that handles higher-order processing.28PubMed Central. Birds have primate-like numbers of neurons in the forebrain So when someone dismisses a clever animal by saying “it only has a bird brain,” the joke is on us. The number of neurons and how they are connected matters far more than overall brain size.
City Birds and the Flexibility Factor
Intelligence in the wild is not a fixed trait stamped by species membership. It responds to the environment an individual actually lives in. A study of great tits, a common European songbird, compared problem-solving performance between urban and forest populations. Urban pairs were significantly faster at solving novel tasks.29PubMed. Problem-solving performance and reproductive success of great tits in urban and forest habitats City life, with its unpredictable food sources, unfamiliar objects, and constantly changing landscape, appears to select for or encourage behavioral flexibility. This lines up with a broader idea in cognitive ecology: intelligence is not about doing one thing brilliantly, it is about adjusting your behavior when the world throws you something new.
The same principle likely applies across many species now living alongside humans. Raccoons opening trash cans, crows dropping nuts onto crosswalks to let cars crack them, macaques stealing tourists’ phones and bartering them back for food. These are not pre-programmed routines. They are flexible responses to environments that did not exist a few generations ago. Animals that thrive in human-altered landscapes tend to be the ones with the cognitive toolkit to innovate on the fly.
When Sentience Meets Law
The growing recognition of animal intelligence has started to reshape how societies treat animals legally. The formal acknowledgment of animal sentience has influenced welfare legislation in multiple countries, changing standards of care and strengthening the justification required before animals can be used in research or industry.30PubMed Central. The sentience shift in animal research Legal scholars have argued that once sentience is formally recognized, it becomes difficult to deny animals legal protection simply because they are not charismatic, emotionally close to us, or economically useful.31Journal of Animal Ethics. The Recognition of Animal Sentience by the Law
This matters well beyond the courtroom. If bees can experience something like pleasure, if octopuses may dream, if elephants console each other and monkeys protest unfairness, then the moral circle most cultures draw around “beings that matter” is almost certainly too small. The science does not tell us exactly where to draw that line. But it keeps pushing the line outward, and faster than most ethical frameworks have been able to keep up with.