Crows Decide Fast. Intelligence Is More Than Speed
A crow can make one kind of visual decision in about 235 milliseconds—roughly twice as fast as the human and monkey results used for comparison. That is a remarkable finding. It is not proof that crows are generally smarter than people, nor that reaction speed is intelligence. The real discovery is more interesting: a bird brain organized without a mammalian neocortex appears to accumulate evidence, approach a decision threshold, and commit to a choice using neural dynamics strikingly similar to those measured in primates. Evolution may have built comparable computations out of very different biological architecture.
The study was published in Science on October 22, 2026, then returned to the news cycle on November 12 under the irresistible headline that crows decide twice as fast as humans. To understand what survives that headline, we need to separate the task, the neural mechanism, the comparison, and the philosophical claim.
What the experiment actually measured
Researchers Philipp Schmidbauer and Andreas Nieder trained two carrion crows to watch a cloud of moving dots on a touchscreen. Some dots drifted coherently left or right while the rest supplied visual noise. The birds reported the dominant direction by pecking the corresponding target. According to the peer-reviewed paper, the crows responded in approximately 235 milliseconds on average while maintaining accuracy.
This is a classic random-dot-motion task, designed to study perceptual decision-making rather than memory, planning, language, social judgment, or tool use. When the signal was clear, the birds approached perfect accuracy. Even when only a small fraction of dots moved together, their choices remained better than chance. The important point is not merely that the peck was fast. The birds extracted a weak pattern from noise and committed to an answer.
Evidence accumulation without a mammalian cortex
The researchers recorded activity from neurons in the nidopallium caudolaterale, or NCL, an association region of the avian forebrain. It is not a miniature prefrontal cortex, but it performs some comparable executive functions. About 70 milliseconds after the dots appeared, neural activity already began distinguishing the direction the crow would eventually choose—roughly 170 milliseconds before the physical response.
Those signals did not simply flash ‘left’ or ‘right.’ Activity ramped as evidence accumulated, varied with the strength of the visual signal, and converged near the moment of commitment. The University of Tübingen’s account describes dynamics compatible with established decision models in humans and nonhuman primates. Different anatomy, similar computational logic: gather evidence until the system has enough to act.
That pattern matters because intelligence is often described as though one brain plan won evolution and everything else is an approximation. Crow cognition keeps breaking that hierarchy. The result joins a broader lesson also visible in animal communication research: behavior should be measured on the animal’s terms, not ranked by resemblance to us. Our earlier piece on why cats meow differently for human audiences examines the same trap from another direction.
The headline’s hidden comparison
“Twice as fast as humans” sounds like a direct race. It was not. The crow results were compared with previously published human and monkey performance on comparable self-paced visual-motion tasks. The basic paradigm is well established, but different studies can vary in displays, training, response methods, motivation, and analysis. Humans may press a control or move their eyes; crows peck. The numerical comparison is meaningful, but it is not the same as placing birds and people in one laboratory under perfectly identical conditions.
The sample also consisted of two extensively trained crows. That is common in invasive single-neuron research, where each animal produces many trials and detailed recordings, but it limits claims about all crows, all corvids, or all birds. The study identifies an extraordinary capability in these subjects and a plausible species-level mechanism. It does not supply a population survey.
Speed itself is not a universal measure of intelligence. A chess player who waits before moving is not necessarily less intelligent than a player who moves instantly. Deliberation can be adaptive when consequences are uncertain, social, or irreversible. The experiment measures fast perceptual evidence accumulation in a constrained task—an important cognitive component, not the whole mind.
Three claims, only one is firmly established
Strong claim: these crows solved this task extraordinarily fast
The behavioral result is direct: the birds selected motion direction in around a quarter of a second without an obvious speed–accuracy tradeoff. The neural recordings independently support that behavior by showing decision-related activity emerging before the peck.
Plausible claim: avian biology helps compress the decision
Shorter neural wiring, densely packed neurons, high body temperature, and evolutionary pressure for rapid visuomotor control may all contribute. Flying animals must process movement, avoid obstacles, guide a beak, and correct a trajectory quickly. These explanations are biologically sensible, but the study did not isolate one cause. As a detailed technical report notes, the advantage may be a cascade of small speed gains rather than a single secret circuit.
Unsupported leap: crows are twice as intelligent as humans
No accepted unit converts milliseconds into general intelligence. Minds trade among speed, accuracy, flexibility, inhibition, memory, social inference, and the ability to revise a choice. Calling the result proof of crow superiority would reproduce the same crude ladder that once placed humans automatically at the top. The better conclusion is plural: different nervous systems can be optimized for different cognitive economies.
What crows complicate about the idea of intelligence
The philosophical shock is not that a crow can beat us at one stopwatch test. It is that recognizable decision dynamics can appear in a brain whose large-scale organization diverged from ours hundreds of millions of years ago. Similar problems may pull evolution toward similar computations without requiring similar anatomy. Intelligence may be less like one sacred organ and more like a family of workable solutions.
That should also change how science headlines are read. A good headline opens a door; it should not replace the room. The most useful response is to ask what was measured, against which baseline, through which mechanism, and with what limits. We used the same discipline when examining new biomarkers for feline kidney disease: an exciting signal is not yet a universal diagnosis, and an impressive task is not yet a complete theory of mind.
A faster answer is not the final answer
The crow study does not shrink human intelligence. It enlarges the map of possible minds. Two birds showed that rapid evidence accumulation does not require a layered mammalian cortex, and their neurons revealed a familiar decision process running on unfamiliar biological hardware. The headline gives us a race. The science gives us convergence, specialization, and a warning against confusing one measurable advantage with an entire hierarchy of consciousness.
Which ability should count most when comparing intelligence across species—speed, flexibility, memory, social understanding, self-control, tool use, or something else—and how would you test it fairly?
Explore the Cat-Coded collection for strange feline signals, then join the Claw & Riot Salon to continue the argument.




Comments