Shulou(Shulou.com)11/24 Report--

Photo Source: pixabay {do not [do not (do not)]}, do you understand?

The crow is one of the smartest animals. They not only follow the rules to make decisions, make and use tools, but also seem to demonstrate an innate ability to perceive numbers. A new study shows that these smart birds can understand recursion-the process of nesting one structure in another similar structure, which has long been considered a unique human ability.

Recursion is a key feature of language. It enables us to build complex statements with simple statements. Take the phrase "the mouse pursued by the cat ran away" as an example, the phrase "the cat chased the mouse" is actually included in the sentence "the mouse ran away". For decades, psychologists have believed that recursion is a unique feature of human language. Some scientists believe that recursion is a key feature that distinguishes human language from other forms of animal communication. As for this inference, there have always been doubts in the academic circle. "people have always wondered whether other animals can also understand recursive sequences," said Diana Liao, a professor of animal physiology at University of T ü bingen in Germany and a postdoctoral student at Andreas Nieder's laboratory. (this is also a recursive statement)

In a study published in 2020, researchers surveyed adults, children and monkeys and found that the ability to generate recursive sequences may not be unique to humans. The researchers showed humans and monkeys a display screen with two pairs of randomly arranged parenthesis symbols. Volunteers (including monkeys) were asked to touch them in the order of a recursive sequence of "central embedding". For example, {()} or ({}) is a recursive sequence, while {(}) is not. If the correct answer is given, humans will receive oral feedback, while monkeys will be rewarded with a small amount of food or fruit juice. The researchers then showed the volunteers a new set of parentheses and observed how often they arranged them recursively. Of the three monkeys tested, two monkeys generated recursive sequences more frequently than non-recursive sequences, although they needed extra training to do so. One of the monkeys could arrange a recursive sequence after about half of the training. By contrast, children between the ages of three and four can arrange recursive sequences after about 40% of training.

The study made Liao and his colleagues wonder: do crows with excellent cognitive abilities also have recursive abilities? Referring to the methods used in the 2020 study, the team trained two crows to peck at pairs of parentheses with a recursive sequence embedded in the center. The researchers then tested their ability to spontaneously generate recursive sequences on a new set of symbols. The results showed that crows performed on a par with children, and they were able to arrange recursive sequences after about 40% of training, without the need for extra training like monkeys. The team's findings were recently published in the journal Scientific Progress (Science Advances).

George Valentigara, a professor of neuroscience at the University of Trento in Italy (University of Trento), who was not involved in the work, said it was "very interesting" that crows could identify structures embedded in the center and were even better at recursion than monkeys. He added that these findings raise a question: how do animals use recursion? "they don't seem to have any system similar to human language, so recursion may be related to other cognitive functions," he said. " One theory is that animals may use recursion to express relationships in their social groups.

In fact, after the 2020 study on the recursive ability of humans and monkeys was published, some scientists still could not believe that monkeys could understand recursion. Some people think that these animals may choose recursive sequences by learning the order in which parentheses are displayed. For example, if the training sequence is [()], but later shows the monkeys different pairs, such as () and {}, then they will first select the parentheses identified in the training, and then select new parenthesis pairs that they have never seen before. Finally, at the end of the sequence, they select the parentheses that match the training (because they have learned to put the matching parentheses at the end).

To overcome this limitation, Liao and his colleagues extended the training sequence from two pairs of parentheses to three pairs, such as {[()]}. Liao said that with three pairs of symbols, the probability of generating sequences without mastering the basic concept of recursion is much lower. As a result, the researchers found that birds are most likely to have the ability to choose embedded centers.

In the experiment, there is no border around the first parenthesis, and then the parenthesis is framed (picture source: original paper), but some scientists are still skeptical. Arnaud Rey, a senior researcher in psychology at the French National Centre for Scientific Research (French National Center for Scientific Research), said the findings can still be explained in terms of simple associative learning, that is, animals can learn to associate one symbol with the next, such as connecting one front parenthesis to the next. A key reason, he explains, is a feature of research methodology design: the researchers added a border around the parentheses, which they pointed out was necessary to help animals define the order of parentheses (the same border layout was used in the 2020 study). For Ray, this is a key point of the research limitation, because animals can only learn the order of display before and after parentheses by treating the border symbol at the end of the arrangement as a reward symbol.

In Ray's view, the concept of "recursive processing" as a unique cognitive form is flawed. Even in humans, he says, this ability is likely to be explained by associative learning mechanisms, which he and his colleagues suggested in a 2012 study of baboons. So far, there is no satisfactory explanation for how the human brain's ability to recognize and manipulate such sequences is encoded. According to Ray, researchers are currently divided into two main groups: some believe that human language is based on certain unique abilities, such as the ability to understand recursion; others believe that human language comes from simpler processes such as associative learning.

But Liao points out that even with the help of the border, the crow still needs to find out the order in which the center is embedded, that is, arranging parentheses and closing parentheses from the outside to the inside. In other words, if birds only know that the opening parenthesis is at the beginning of the sequence and the closing parenthesis is at the end of the sequence, then they will get the correct sequence in the same proportion and the wrong sequence such as "({)}". However, she and her colleagues found that even with more complex sequences of three pairs of parentheses, crows were more likely to choose the right sequence.

In Liao's view, the ancestors of birds parted ways with primates on the tree of life a long time ago, and birds also seem to be able to parse and generate recursive sequences, which means that this ability is "ancient and evolutionary." or it developed independently as a product of so-called convergence and evolution. Liao added that because birds' brains lack the neocortex of primates, this observation suggests that the brain structure of primates may not be necessary to show this cognitive ability.

The results of this new study show that birds have many of the same cognitive skills as primates, which is consistent with the results of many previous studies. Mathias Osvath, an associate professor in the department of cognitive science at Lund University University in Sweden, who was not involved in the new paper, said: "for me, this paper proves once again with strong data that birds are completely misunderstood. It is completely wrong for mammals to control the world at the cognitive level."

Original text link:

Https://www.scientificamerican.com/article/crows-perform-yet-another-skill-once-thought-distinctively-human/

This article comes from the official account of Wechat: global Science (ID:huanqiukexue), written by Diana Quan (Diana Kwon), translated by Peng Rong, revised: bad

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.