Social conventions are an integral part of our species, whether it is the way of greeting each other, the direction of automobile traffic or even language. This form of interaction benefits both partners (unlike altruism, for example, which only benefits one partner). For a behavior to be considered a social convention, it must have three characteristics: be effective, that is to say allow the resolution of a problem; be stable; and be arbitrary, i.e. another solution to the same problem could be just as effective. Although we expect to see social conventions in all social animals, especially other primates, their observation in non-humans is rare. Anthony Formaux, from the University of Aix-Marseille, and his colleagues have just observed the emergence of a social convention in Guinea baboons (Papio papio).
The researchers submitted a coordination problem to a group of 19 baboons from the Rousset primatology station, in the Bouches-du-Rhône. The monkeys had access to large bungalows adjacent to their park, in which there were pairs of touch screens. The experiment began when two monkeys placed themselves in front of two neighboring screens: on one of the two screens chosen at random, two symbols were displayed (among several possible ones), and the first baboon had to select a symbol. The two symbols then appeared on the screen of the second baboon. If he selected the same symbol, both monkeys were rewarded; otherwise, the task would fail and start again three seconds later.
Guinea baboons from the Rousset primatology station of the CNRS.
© Anthony Formaux/Nicolas Claidière/LPC/CNRS
Anthony Formaux and his colleagues used two indicators to study the emergence of a potential social convention over time: the individual score of each baboon in terms of success, and the hierarchy of stimuli, i.e. the probability that one stimulus is chosen preferentially over another.
In the first experiment, the stimuli were seven squares of different colors, and the monkeys could observe what their neighbor was doing. Very quickly, the score of the monkeys increased, and a hierarchy shared by the group was set up between the stimuli (for example, if the monkey had the choice between an orange square and a pink square, it had much more chances of choosing the orange square). Also, the further apart the two stimuli were in the hierarchy (eg, between the highly chosen orange square and the very poorly chosen dark blue square), the higher the probability of success.
These results can be explained in two ways: either the second monkey observes its partner and chooses the same stimulus as him, which would only be imitation; or the monkeys have learned a color hierarchy. The researchers therefore continued the experiment by placing an opaque wall between the two touch screens to prevent observation. The scores of the baboons remained high, which shows that it is indeed a learning process and not an imitation. However, the baboons with the best results in the first part of the experiment saw their results drop more than the others, which seems to indicate that they also used imitation in part, especially in the event of two close stimuli in the hierarchy (orange and yellow for example).
The experimental device with touch screens is in a building where the monkeys can access whenever they want. When two baboons occupy neighboring cabins, they must coordinate to earn rewards.
© Anthony Formaux/Nicolas Claidière/LPC/CNRS
To generalize their results, Anthony Formaux and his colleagues carried out the same experiment with stimuli of a different nature (five black and white images) with a transparent wall, then with five other images, starting directly with an opaque wall. In all cases, the monkeys’ scores increased over time to high levels: the nature of the visual stimulus does not matter, and the monkeys do not need to have visual access to their partner to learn the hierarchy of stimuli.
The hierarchy of stimuli learned by baboons is therefore stable and efficient. But is it arbitrary? To find out, the researchers taught the same monkeys a reverse pecking order for black-and-white images (for example, if the monkeys preferred the dog’s paw to the leaf, they taught them to prefer the leaf), then repeated the same experience. The results were once again similar despite the inversion of the hierarchy: this is quite arbitrary and can therefore be considered as a social convention.
Finally, the researchers were interested in the link between convention and the social structure of the group. “We took the three most socially integrated monkeys on one side, says Anthony Formaux, and the remaining eight on the other, and we taught them two inverse conventions. We had two hypotheses: either the most socially connected monkeys dictate the convention, or the majority wins. The social integration of the monkeys was estimated here from the number of times each monkey performed the experiment with each other monkey. The researchers observed that the convention that ends up being adopted by all is that of the majority. This result could be explained by a conformity of the minority, but also by the fact that the best integrated individuals are perhaps those who know how to adapt the most to others.
“The link between convention and social structure is precisely the subject of our next study,” says Anthony Formaux. We observed that some monkeys worked more with each other than others, and it would be interesting to understand how they organize themselves and how they choose their partners. »
–
–
–
–
