Crows can evaluate visual info twice as fast as mammals

When two hand‑raised carrion crows were pitted against the classic random‑dot‑motion test that has long been the workhorse of primate decision‑making research, they didn’t just hold their own—they outpaced their mammalian counterparts by a wide margin. The study, led by neurobiologist Andreas Nieder of the University of Tübingen, shows that crows can evaluate visual information roughly twice as fast as monkeys or humans while preserving the same level of accuracy. The findings force a rethink of how brain architecture, not just sheer neuron count, shapes cognition.

The classic random‑dot‑motion paradigm

The random‑dot‑motion task presents a cloud of flickering specks on a screen, a fraction of which drift coherently in a single direction. Subjects must judge the overall motion direction, a decision that in primates is known to involve a gradual buildup of neuronal firing rates in the neocortex until a threshold is crossed. This “evidence accumulation” process has been mapped extensively in monkeys, where reaction times typically stretch from four to six hundred milliseconds depending on task difficulty.

By reproducing the exact same visual stimulus and response demands for crows—except that the birds peck a target instead of making an eye movement—Nieder and colleague Philipp Schmidbauer created a direct cross‑species comparison. The crows’ performance, measured from stimulus onset to the first head turn toward the pecking button, averaged 235 milliseconds, a figure that sits well below the 410‑600 millisecond range reported for monkeys in comparable studies.

Speed without sacrificing accuracy

Speed alone would be a hollow victory if it came at the cost of precision, but the crows proved otherwise. Even when only three percent of the dots moved coherently—a condition barely distinguishable from random noise—the birds performed above chance, matching the accuracy levels of primates and humans in the same task. The researchers deliberately allowed the crows to take as much time as they needed; the birds were not forced into a rapid‑response regime, yet they still responded roughly twice as fast as any primate subject.

Reaction‑time analysis revealed only a modest 34‑millisecond spread between the easiest (81 percent coherence) and hardest (3 percent coherence) conditions. While this suggests a less pronounced difficulty effect than seen in mammals, Nieder emphasizes that the crows still showed a statistically significant increase in latency as coherence dropped, indicating that they too experience a trade‑off between evidence strength and decision time.

Neuronal evidence accumulation in a bird brain

To probe the neural underpinnings, the team recorded from 248 neurons in the nidopallium caudolaterale (NCL), a region widely regarded as the avian analogue of the primate prefrontal cortex. Of these, 58 neurons displayed direction‑selective firing that predicted the bird’s eventual choice. Crucially, these decision‑related signals emerged about 70 milliseconds after stimulus onset—well before the bird’s head began to move—mirroring the early evidence‑accumulation dynamics observed in primate neocortical neurons.

The NCL’s lack of layered cortical architecture, replaced instead by dense, non‑laminar clusters, did not prevent it from generating a decision variable comparable to that seen in the mammalian neocortex. Nieder interprets this as a striking example of convergent evolution: two lineages separated by more than 360 million years have arrived at essentially the same computational solution for visual decision‑making, albeit built on very different anatomical scaffolds.

Why birds are faster: size and wiring

One obvious factor is sheer scale. Birds, constrained by the physics of flight, pack an extraordinary number of tiny neurons into compact brains. Shorter axonal pathways mean that electrical signals travel over reduced distances, shaving off precious milliseconds. Nieder likens this to “short lines speed everything up,” underscoring how reduced wiring length can accelerate the entire sensorimotor cascade.

Temperature also plays a role. While the average human body temperature hovers around 36.6 °C, many birds maintain core temperatures of 41 °C or higher. Elevated temperature accelerates ion‑channel kinetics and other biochemical processes, effectively turning up the speed of neural computation across the board. The combination of compact wiring and a hotter internal environment creates a physiological milieu where every step—from photon capture to motor output—runs faster than in mammals.

Implications for neuroscience and AI

The crow study forces a reassessment of the notion that the neocortical sheet is the sole architecture capable of sophisticated decision‑making. If avian brains can implement the same evidence‑accumulation algorithm within a radically different structural framework, then the computational principles underlying cognition may be more abstract than previously thought. This opens the door for bio‑inspired AI designs that eschew layered networks in favor of densely interconnected, high‑speed modules.

Moreover, the finding that speed does not inherently erode accuracy challenges a long‑standing trade‑off assumption in both neuroscience and engineering. Understanding how crows preserve precision while compressing decision latency could inform the development of rapid‑response systems—ranging from autonomous vehicles to real‑time medical diagnostics—where split‑second judgments are critical.

Future directions and unanswered questions

While the current work highlights a cascade of faster processes in birds, it does not pinpoint a single bottleneck responsible for the speed advantage. Nieder suggests that the acceleration likely arises from a combination of visual processing, evidence accumulation, and motor execution, each operating at a higher tempo. Further experiments that isolate each stage—perhaps by varying visual contrast, imposing forced‑choice deadlines, or recording from downstream motor nuclei—could clarify which links in the chain contribute most to the overall gain.

Another avenue lies in comparative studies across avian species. Carrion crows are known for their problem‑solving prowess, but do other birds, such as pigeons or parrots, exhibit comparable decision speeds? Mapping the spectrum of avian decision latency could reveal whether the observed speed is a general feature of bird brains or tied to the ecological demands of highly intelligent corvids.

What the rapid crow tells us about brain evolution

The crow’s performance underscores a broader evolutionary lesson: distinct neural architectures can converge on similar computational solutions when faced with comparable ecological pressures. Birds and mammals diverged over 360 million years ago, yet both have evolved mechanisms for accumulating noisy sensory evidence and committing to a choice. The avian brain’s clustered organization, devoid of cortical layers, achieves this with a speed advantage that likely reflects the demands of flight, foraging, and predator evasion.

In the end, the study reminds us that brain function cannot be reduced to a single structural motif. Instead, evolution tinkers with size, wiring length, temperature, and cellular composition to meet the same computational challenges. As we continue to decode the neural basis of decision‑making across species, the crow’s swift, accurate judgments will stand as a benchmark for how diverse brains can solve the same problem—fast, and without cutting corners.

This article was produced with AI-assisted research and editorial support. Reporting is based on the source material cited below. Sources: Ars Technica; arstechnica.com; Global1.News (23 October 2026).

By Jessica Ali, Staff Writer