These New Zealand parrots master the concept of statistics

The kéas, parrots present in New Zealand, would be able to use statistics to choose the best option. Until now, this ability has only been observed in great apes.

A study published in Nature communications just brought down a myth. Great apes, including humans, are not the only ones who can assess the chances of success of the various possible choices. Simply put, use the odds to get what they want. The kea parrot, a mountain dweller endemic from New Zealand, is also capable.

Indeed, using jars filled with clothespins, researchers from the University of Auckland (New Zealand) have shown that Nestor notabilis would master statistical inference. This means that he would be able to draw a conclusion about a situation thanks to a statistical analysis of a sample of this situation. In this case, the parrots received food by choosing the black tongs, and nothing by choosing the orange tongs. Very quickly, they only chose the black pliers. But the experiment was just beginning.

Which jar to choose?

Subsequently, the experimenter came into play. He presented several jars to the six kéas studied in six separate tests. Each time, two jars were shown simultaneously. The first containing more black clips, and the second more orange clips, thanks to different configurations: jars of the same size, jars of different sizes, jars with an inaccessible bottom, etc.

The experimenter then drew pliers from each jar. While keeping his hands closed, he let the parrot choose the hand from which he wanted to retrieve the clothespin. Quickly, parrots learned to designate the hand from the jar with either the most black pliers or the most black pliers on the surface, since the hand only caught a pliers on the surface of the jar.

Children and primates are subject to comparable experiences. At first glance, this may seem simple. Black forceps = food. More black pliers = more chance of gaining food. In reality, this means that these parrots can observe a situation, realize that a pattern is repeating, and use the results previously obtained to predict (with a certain margin of error) the following results, taking into account variables environmental.

Thus, from a situation sample, they induce a probability that the sample is representative of the whole situation, without forgetting that a degree of error must be taken into account. In summary, they would master inference statistical.

This will also interest you

Did you like this article ? Don't hesitate to share it with your friends and help us to promote ABSMARTHEALTH :)! The editorial team thanks you.