Crows may be bird-brains, but the feathered creatures can understand the highly abstract concept of zero, new research suggests.
The concept of zero, as used in a number system, fully developed in human society around the fifth century A.D., or potentially a few centuries earlier, Live Science previously reported. For instance, the notion of multiplying 8 by 0, or adding 0 to 10, didn't emerge until then. The concept of "none," or the absence of any quantity, likely emerged earlier, but this differs from using zero as a distinct "quantity," in and of itself.
That idea may sound obvious, but following the conception of zero as a numerical value, the field of mathematics underwent a dramatic transformation.
"If you ask mathematicians, most of them will probably tell you that the discovery of zero was a mind-blowing achievement," said Andreas Nieder, a professor of animal physiology in the Institute of Neurobiology at University of Tübingen in Germany. "The special thing about zero is that it doesn't fit into a routine of counting real objects, as with the actual integers." In other words, someone can count three apples placed in a basket — one, two, three — but when the basket is empty, there are no apples to count.
Related: The 12 weirdest animal discoveries
Zero represents that emptiness, the absence of apples, and "that obviously requires very abstract thinking ... thinking that is detached from empirical reality," Nieder said. And now, by peering into the brains of crows, Nieder and his colleagues have discovered that the birds' nerve cells, or neurons, encode "zero" as they do other numbers. The birds' brain activity patterns also support the idea that zero falls before "1" on crows' mental number line, so to speak.
In the new study, published June 2 in The Journal of Neuroscience, the team ran experiments with two male carrion crows (Corvus corone), during which the birds sat on a wooden perch and interacted with a computer monitor in front of them. In each trial, a grey screen containing zero to four black dots popped up in front of the crows; this "sample" image was followed by a "test" image containing either the same or a different number of dots.
The crows were trained to peck at the screen or move their heads if the two images matched one another, and to remain still if they did not match.
In a previous study using the same setup, the group showed that crows could successfully identify the matched and unmatched pairs of images about 75% of the time after undergoing extensive training for the experiment, according to a report published in 2015 in the journal Proceedings of the National Academy of Sciences. This previous study did not include an empty screen, standing in for zero, but it did demonstrate that the crows could differentiate an image containing three dots from a screen containing five, for instance.
The greater the difference between the two sets of dots, the more accurately the birds responded; in other words, the birds mixed up closer quantities, such as two and three, more often than more divergent quantities, such as one and four. This phenomenon is known as the "numerical distance effect," which can also be observed in monkeys and humans during similar tests, Nieder told Live Science.
In the more recent study, which included a blank screen, "what we found is that the crows, after this training, could discriminate zero from the other countable numerosities," Nieder said. However, importantly, the birds still demonstrated the numerical distance effect in trials that included the empty screen.
That means that the birds mixed up the zero-dot image with the one-dot image more often than with two-, three- or four-dot images, Nieder explained. "This is an indication that they treat the empty set, not just as 'nothing' versus 'something,' but really as a numerical quantity," in that they perceive zero dots as proximal to one dot.
To better understand the brain activity behind these behaviors, the team implanted tiny, glass-coated wires into the birds' brains to record electrical activity while the crows repeated the behavioral tests. The chosen neurons sat within a region known as the pallium, which is located toward the back of the bird brain and handles high-level cognitive functions.
The avian pallium belongs to a larger brain region called the telencephalon; humans also have a telencephalon, of which the cerebral cortex, the wrinkled outer layer of the human brain, is one part. But although both the pallium and cortex lie in the telencephalon, there's where many similarities end between the two structures. While the cerebral cortex contains six distinct layers of brain tissue, connected by crisscrossing wires, the avian pallium contains no layers and instead arranges neurons in nuclear clusters, Nieder said.
Related: 7 ways animals are like humans
In the prior 2015 study, the team also gathered recordings from the pallium and specifically zoomed in on one key region, known as the nidopallium caudolaterale (NCL). The NCL receives sensory information, including that from the eyes, processes that data and sends it to areas of the brain related to motor functions, to coordinate physical behaviors. (In primates, the prefrontal cortex plays the same role.)
In the NCL, the team found that certain subgroups of neurons went wild when specific numbers of dots appeared on the screen. Some would begin rapidly firing in response to two dots, while others kicked off for four, for example. These neurons appeared "tuned" to a specific number. And interestingly, the greater the distance between that preferred number and the number of on-screen dots, the less active those neurons became.
These patterns of brain activity hinted at how the crows perceive numerical values in relation to each other, Nieder said. "They are inherently representing this ordinality aspect of numbers, this ordering along a number line, so that after one comes two and after two comes three, and so on," he said.
In the new study, the team repeated this experiment with the addition of the zero-dot screen. In all, they took recordings from more than 500 neurons, 233 in one crow and 268 in the other. As before, they found that different subsets of NCL neurons lit up in response to different numbers of dots, but in addition, another subset fired in response to the blank screen. These neurons became less and less active the more dots popped up on-screen — or the further from zero the number grew.
In combination, the observed patterns of behavior and brain activity suggest that, yes, crows indeed grasp the concept of zero, the authors concluded. What utility this holds for the animals, if any, remains unclear, Nieder told Live Science. While being able to distinguish one piece of fruit from two can be useful for survival, for instance, "I don't see an immediate advantage for these animals to understand nothing as a quantity," he said.
Other behavioral studies have shown that rhesus macaques and honeybees also demonstrate an understanding of zero. In terms of brain activity linked to zero, multiple studies have shown that monkeys carry specially tuned neurons for the number zero, just like crows. And more recently, Nieder and his colleagues demonstrated the same in humans, as described in a 2018 report in the journal Neuron.
"I think that initially it sounds a little crazy to ask whether animals understand zero, because zero is a very, very special, almost magical, number that we have," Nieder said. But now, growing evidence hints that more animals may understand the concept of zero than scientists originally realized.
Even so, Nieder said he'd be surprised if animals like amphibians or reptiles could do mathematical calculations that rely on an understanding of zero, since their learning capabilities don't match that of mammals and birds. But given that birds and mammals split off from their common ancestor well before the extinction of the dinosaurs, the fact that they share overlapping cognitive abilities is also remarkable, Nieder said.
"That's the fascinating aspect, that evolution obviously found different anatomical ways, independently, to equip those birds and mammals with high-level cognitive functions."
Originally published on Live Science.
