Crows Have Another Skill That Was Once Thought To Be Unique To Humans

Crows Have Another Skill That Was Once Thought To Be Unique To Humans

In A New Study, Researchers Have Shown That Crows Have A Special Ability That Is One Of The Key Features Of Grammar; Of Course, Not All Scientists Are Convinced.
Crows are among the most intelligent animals. They can make rule-based decisions and have the ability to build and use tools. Now researchers report that these smart birds can understand recursion (the process of nesting structures inside other similar systems), a capability long thought to be unique to humans. They also seem to have an innate sense of the nature of numbers.
Recursion is one of the critical features of language. This feature enables us to make complex sentences using simple sentences. Consider the sentence, “The mouse ran, chased by the cat.” Here the semi-sentence “the cat that chased” is included in the semi-sentence “the mouse ran.”
For decades, psychologists thought that rebounding was unique to humans. Some considered it a critical feature that distinguishes human language from other forms of communication between animals, But there were questions about this assumption. “There has always been interesting in whether nonhuman animals can also understand recurrent sequences,” says Diana Liao, a researcher in the Andreas Nieder lab at the University of Tübingen in Germany.
In 2020, a group of researchers reported in a study of young and adult monkeys and humans that the ability to produce repetitive sequences may not be unique to our species.
Two pairs of bracket symbols were shown to humans and monkeys in random order on the screen; Humans and monkeys were then trained to touch them in the order of a centrally embedded recurrence sequence, such as { () } or ({ }). The researchers then showed them a new set of brackets and checked how often they reordered the symbols. After a correct answer, the humans received verbal feedback, and the monkeys were rewarded with a small amount of food or juice.
Two of the three monkeys in the experiment produced recurrent sequences with greater frequency than non-recurrent sequences such as { (}). However, they required an additional training session—one of the animals made return sequences in about half of the trials. In contrast, three- to four-year-olds produced recursive lines in approximately 40% of problems. The paper prompted Liao and his colleagues to investigate whether crows, with their famous cognitive skills, might have a similar ability.
Modifying the protocol used in the paper published in 2020, the researchers trained two crows to peck at pairs of brackets presented in the form of a recurrent sequence embedded in the center. The researchers then tested the birds’ ability to spontaneously generate such recursive sequences on a new set of symbols.
The performance of the crows was at the level of children. Without the extra training the monkeys needed, the birds produced the return sequences in about 40 percent of the trials. The researchers published the results of their experiments in the journal Science Advances.
Giorgio Valortigara, a professor of neuroscience at the University of Trento in Italy, who was not involved in the new study, said the discovery that crows can perceive centrally embedded structures and are better at doing so than monkeys is surprising.
These findings raise the question of what nonhuman animals might use the ability to perceive embedded structures for, Valortigara adds. “They don’t seem to have anything like human language, so the reversal probably has to do with other cognitive functions,” he says. One of the scientists’ guesses is that animals may use backbiting to indicate relationships within their social groups.
When a 2020 study of reversal capabilities in humans and monkeys was published, some experts weren’t convinced that monkeys understood reversal. Instead, some have argued that animals select the recurrent sequences by learning the order in which the brackets are presented. For example, if the trained line was [ ( ) ] and the monkeys were later shown a different pair such as ( ) and { }, they would first choose the bracket they had learned in training and then select the new bracket pair they had previously known. Had not seen. Finally, they would choose the bracket corresponding to the training session at the end of the sequence (because they had learned that the bracket would come at the end).
To overcome this limitation, Liao and colleagues extended the sequences from two pairs to three pairs, such as { [ () ] }. By having three teams of symbols, Liao says, it is much less likely to generate sequences without understanding the concept of recursion. Here, too, the researchers found that birds were more likely to choose lines embedded in the center.
Some scientists are still skeptical about this. These findings can still be interpreted from the perspective of associative learning, in which the animal learns to associate one symbol with another (for example, an open bracket with a closed bracket), says Arnaud Rey, a senior psychologist at the French Center for Scientific Research.
Another critical reason, Ray explains, lies in a feature of the study’s design: the researchers included closed brackets in the frame, which they say was necessary to help the animals recognize the order of the shelves. (The same design with a box was used in the 2020 study).
Understand that open and closed brackets were displayed. According to Ray, this is one of the main limitations of the study because the animals were able to understand that the boxed symbols (which always appeared at the end of the reversal sequences) were the ones that were associated with the reward. Hence, the boxes helped them sort out the series.
According to Ray, the concept of recursive processing as a unique form of cognition is incomplete. He says associative learning mechanisms can explain this capacity even in humans. He and his colleagues suggested the same in a 2012 study of baboons. Furthermore, no convincing explanation has been provided as to how the ability to recognize and manipulate such sequences is encoded in the human brain.
According to Ray, researchers are currently divided into two groups: the group who believe that human language is built on unique capacities such as the ability to understand repetition, and the other group who thinks this phenomenon results from a more straightforward process such as associative learning.
Liao notes that even with the frames, the crows still had to find the arrangement embedded in the center where the open and closed brackets were paired from outside to outside. In other words, if the birds had only learned that available racks were at the beginning of the sequence and closed frames were at the end of the line, you would expect the proportion of ({) } responses to be equal to correct answers. Still, the researchers found that the crows chose the latter more often, even with more complex arrangements that included three pairs of brackets.
For Liao, seeing that birds, whose ancestors have diverged much further down the evolutionary tree than primates, also appear to be able to parse and produce recursive sequences suggests that this ability is either evolutionarily ancient or the result of convergent evolution. And it was created independently.
Liao adds that since bird brains lack the neocortex found in primates, this observation suggests that a second brain architecture may not be necessary to display this cognitive ability.
According to Matthias Oswalt, an associate professor of cognitive science at Lund University in Sweden, who did not contribute to the new paper, the findings of the new study are among the many studies that show that birds have many of the cognitive skills found in primates. “To me, the new study adds to the data that show we don’t understand birds very well,” says Swat. “It is not true to say that mammals are cognitively superior to other creatures.”