Intelligence Is Not a Ladder

01Stop Asking for One Number

The Kurzgesagt video begins with a useful provocation: intelligence is not a glowing essence that sits inside a skull. It is a toolbox. An animal gathers information, remembers what mattered, predicts what might happen, and changes its behavior when the old solution stops working. That definition is broad enough to include a raven hiding food, a bee navigating by landmarks, an octopus opening a container and a human composing a symphony—without pretending those are the same achievement.

That matters because “smartest animal” lists quietly turn a multidimensional capability into a vertical scoreboard. A creature can be excellent at navigation and poor at imitation; a social species can read relationships while a solitary predator excels at spatial problem-solving. Comparative psychologists therefore test particular capacities: attention, memory, categorization, timing, planning, tool use, communication and self-recognition.

02Brains Scale With Bodies—But Not Neatly

A large brain is expensive tissue, and a large animal generally needs more neural machinery simply to operate a larger body. That is why raw brain mass is a blunt comparison. The encephalization quotient, or EQ, tries to improve the comparison by measuring brain mass against the amount expected for an animal of that body size. It is still a proxy, not a verdict; the formula was developed for mammals and does not transfer cleanly across all animal groups.

A newer line of work focuses on neuron counts in the forebrain or pallium—the sensory-associative machinery more directly involved in flexible behavior. Even there, the comparison is not a universal leaderboard. The structures counted in a honey bee, a bird and a mammal are not identical, and methods differ. The honest conclusion is narrower: neural architecture constrains what an animal can do, but behavior reveals what it actually does with that architecture.

03The Crow Problem

Corvids make the “intelligence is not a ladder” argument impossible to ignore. Crows, ravens, rooks and magpies have demonstrated tool use, flexible problem-solving and—depending on species and test design—behaviors interpreted as self-recognition. Their brains are tiny beside ours, yet their pallium is densely populated with neurons. A raven’s estimated pallial count is about 1.204 billion; a Eurasian magpie’s is about 443 million. Those are not human-like brains in miniature. They are different solutions to an information problem.

Tool use is particularly revealing because it requires a relationship between present action and future outcome. A New Caledonian crow bending or selecting a probe is not merely moving an object; it is exploiting an object’s properties to reach a goal. But even this needs care. A behavior may be inherited, individually learned, socially transmitted or some mixture of all three. Calling it “reasoning” too quickly can hide the question researchers actually need to answer: what information does the animal represent, and how does that representation guide the next move?

04Octopus, the Convergent Outsider

If corvids are a reminder that a small bird brain can be sophisticated, the octopus is a reminder that sophistication does not require a vertebrate brain at all. Wikipedia’s review of cephalopod intelligence places octopuses among the most capable invertebrates and estimates roughly 500 million neurons for an octopus. About two-thirds of those neurons are in the arms, where local circuits can coordinate complex actions without every detail being routed through the central brain.

That arrangement changes the question. An octopus is not a soft-bodied mammal with eight extra hands; it is a distributed nervous system that senses and acts through a body capable of camouflage, exploration and manipulation. Its intelligence is embodied. The animal’s skin can change color and texture, its arms sample the environment, and its hunting tactics are shaped by a life in which every crevice may contain prey or danger.

05Memory Is More Than a Trick

Animal cognition research moved beyond anecdotes partly by moving into controlled experiments: delayed responses, mazes, novel-object recognition, categorization tasks and field observations of food caching or navigation. The point is not to make animals perform party tricks. It is to separate a one-off response from a reusable internal model.

Memory becomes especially interesting when it changes attention. A bird that has learned where food is likely to appear does not sample every part of the world equally. A bee that has learned a visual cue can prioritize one feature over another. In laboratory terms, previous learning can “block” attention to information that adds nothing new. In ecological terms, a mind is always deciding what not to process.

This is one of the video’s strongest themes: intelligence is not only solving a problem. It is selecting the problem worth solving. Every nervous system is a filter, and every filter is shaped by the animal’s body, habitat and history.

06The Mirror Is a Narrow Window

The mirror test is a famous attempt to probe visual self-recognition. Researchers place a mark somewhere the animal cannot normally see, then provide a mirror. If the animal uses the reflection to investigate the mark on its own body, that is interpreted as evidence that it understands the reflected image as self rather than as another individual.

The result is compelling when it works, but it is not a universal consciousness detector. Dogs may rely more heavily on smell than sight. A dolphin may experience its body acoustically. An animal may understand its own body without caring about an artificial sticker. The literature itself records failures, mixed findings and critiques, including the possibility that mirror performance depends on experience and training.

The better conclusion is modest: some animals can use a visual reflection as information about their own bodies. That is remarkable. It is also not the same as proving that they possess a human-like narrative self.

07A Better Question for the Future

Animal cognition is most interesting when it refuses the easy metaphor. A crow is not a defective human. An octopus is not a wet robot. An elephant’s large brain is not a guarantee of every cognitive skill, and a bee’s small nervous system is not evidence of an empty inner world. Each animal is an engineered compromise between energy, senses, body plan, social life and ecological pressure.

The practical payoff of this view reaches beyond clever facts. It changes how we design enrichment for captive animals, how we interpret welfare signals, how we protect habitats and how we decide which experiments are ethically justified. If cognition is a collection of evolved capacities rather than a human-shaped ladder, then the burden is on us to measure animals on terms that their lives make meaningful.