Computer simulations showed that simple methods to obtain reciprocal cooperation based on spatial association or the sharing of similar features cannot work in the absence of memory of past interactions
Cooperative computerized agents that relied on such simple methods were regularly cheated by selfish agents that never returned cooperation. In evolving artificial populations of agents, cooperative agents always went extinct in a few generations. These studies show so called "symmetry-based reciprocity" cannot be evolutionary stable.
Computerized "agents" behave cooperatively
A set of computer simulations showed that artificial "agents" can reproduce several characteristics of animal cooperation if they can choose their partner based on cooperation received.
Recently, we developed agent-based models aimed at testing whether a strategy of pure partner choice could promote the evolution of cooperative behavior and reproduce significant aspects of cooperation in group living animals. Our models showed that agents equipped with a behavioral strategy of "choosing cooperation" can develop diffentiated social relationships and show a positive correlation between cooperation given and received. Furthermore, choosing cooperators won over selfish agents (that never cooperate) in an evolutionary model that included reproduction and selection along multiple generations. These computer simulations suggest partner choice is a significant aspect of cooperation in animals.
Researcher involved
Alle scimmie piace il rischio. Ma non l'azzardo. La Repubblica.it, January 23, 2015
Cebo dai cornetti: l’uso di stimoli simbolici facilita la risoluzione di un compito di inibizione. Le Scienze Web News, April 7, 2011.
Monkeys learn to do arithmetic for peanuts. New Scientist, August 15, 2007
Token values. Current Biology, 17:R737, September 4, 2007
Monkeys understand money. Nature, Published online June 11, 2008
Animal behaviour: Token symbolism. Nature, 453:960, June 19, 2008
Scambio di figurine tra cebi. Galileo, June 11, 2008
Le scimmie con il senso degli affari. Repubblica, June 12, 2008
Almonds or pizza? Capuchins are even smarter than we thought. Wellsphere, June 2, 2009
Monkeys choose variety for variety's sake. Science Daily, March 16, 2010
Personality differences in capuchin monkeys
No one is alike. This is true not only for us, but for monkeys as well. A new study published in the Journal of Research in Personality has demonstrated that the adult capuchins monkeys hosted at the ISTC-CNR of Rome exhibit pronounced individual differences in a broad range of behaviours . An international team of researchers from the ISTC-CNR and Free University Berlin, Germany —psychologist Jana Uher and her CNR-collaborators, the biologists Elisabetta Visalberghi and Elsa Addessi—applied a novel approach to measure personality traits in monkeys.
Most personality research on humans is done using self-reports in questionnaires. But these methodologies have many drawbacks because people usually do not have an accurate perception of their own and others' behaviours and their psychological underpinnings. Previous animal studies that adopted personality questionnaires developed for human individuals or that restricted the behavioural investigation of animal individuals to just a few personality traits. In contrast, this study used a complex procedure involving 15 behavioural tests in which each capuchin monkey was examined individually as well as methodical behavioural observations of each monkey when it was together with its group members. Moreover, this entire procedure was systematically repeated over time.
Striking individual differences in most of the capuchins' behaviors were demonstrated. The study measured 146 behavioural variables that led to identify 21 personality traits. For example, some individuals explored novel objects, novel foods, and changes in the environment quickly and intensely, whereas others were very cautious and reluctant to inspect them. When presented with a large bed sheet hanging between two horizontal poles, one much higher than the other, a male, Sandokan used it as a slide many times. In other words, he was ready to exploit the novelty of the situation to create a new game. In contrast, Vispo, another male of his group, tried all possible ways to avoid any contact with the sheet; in fact, he started walking bipedally when he moved over the poles! As is the case for humans, there were all possible intermediate variations of behaviour between these two extremes.
Very interestingly, individual differences in behavior were stable over time. Moreover, personality was differences are largely unrelated to sex and also to age. Despite the fact that capuchins ranged from 8 to 32 years, age played no role, since only impulsiveness declined with age!
How do these results compare to those in humans? According to Jana Uher "Stable individual differences are not uniquely humans but can be found in other species as well, including capuchins a species whose common ancestors with humans dates back to 35 million years ago. Our finding that capuchins, a group living species in which males are bigger than females, do not exhibit pronounced differences in personality traits, as frequently hypothesized for our human ancestors may shed new light on the causes and consequences of sex and gender differences in humans."
What is next?? The research team is now analysing how "capuchinologists", i.e. researchers and students who have spent months or years studying the monkeys of our Center and/or capuchin monkeys all around the world assess the personalities of their monkeys and how this matches with behavioral results.
For more info, please visit www.primate-personality.net
I cebi imitano Sherlock Holmes. La Stampa (15/10/2008)
Capuchins observe but do not learn novel behaviours by imitation
Cammello, an adult male of capuchin, learns how to use a stick to push a peanut out of a tube. Brahms, an adult female, carefully watches him using the stick over and over again. Although there are plenty of sticks, Brahms occasionally prefers stealing Cammello’s peanut instead of using a stick herself. Why? Has not she learnt how to get the peanut? No, she has not. Indeed, capuchins do not learn by imitation.
Are capuchins able to learn by imitation? We know that they are very innovative and when presented with a problem they are likely to find a solution. They are also tolerant enough, so that approaching and observing novel activities performed by other group members is the norm. Despite this, when capuchins’ ability to learn novel behaviours by observation (for example, the use of a tool) from other group members was experimentally investigated, the results did not evidence an ability to learn by imitation these novel behaviours.
Figure. Tube task. Left, a monkey using a stick to get a peanut out of the tube is closely observed by a member of its group. Right, afterwards, the same individual fails to use the stick correctly despite having watched the skillful monkey solving this problem many times. The unsuccessful capuchin continued to contact the tube with the stick in the same manner as it did before observing the many solutions performed by the others (photos by E. Visalberghi).
We found that naïve individuals do not learn by watching a model repeatedly solving a task that involves the use of the tool. Rather, the main effect of the social input provided by models is to make the observer more interested in manipulating the tool: watching another monkey solving a problem brings the observer’s to develop an interest in the place and objects where the other worked or is working. Eventually, from this increased motivation and the repeated trial-and-error attempts the success may arise. This “socially-biased learning” is possible because capuchins are quite tolerant of one another (they accept other group members inspecting at a close range their activities), and they also are socially oriented (they are pretty interested in other individuals).
img 4
Encountering durable artefacts
In tool using populations naïve individuals are likely to encounter durable artefacts used by expert group members. Fragaszy and colleagues (2013) investigated cases of habitual tool use in wild chimpanzees and capuchin monkeys to assess whether the presence of artefacts supports social learning.
Results indicate that enduring artefacts associated with tool use, such as previously used tools, partly processed food items and residual material from previous activity, aid youngsters to learn to use tools and to develop expertise in their use.
Interestingly, this occurs also when the skillful group members are not present and there is not ongoing behaviour to learn from. Therefore, the contribution of social context to learning a skill expands beyond the immediate presence of a model nearby. Thus, artefacts contribute to traditional technologies in non-humans.
Social learning and food choice: I learn with you, not from you
One capuchin is presented with two differently coloured foods: blue and yellow. He is trained to eat only the blue one while another monkey watches him. When the monkey is presented with the same two foods, does he make the same choice his companion did? No, he does not. He thinks for himself.
We investigated the effects of social influence on capuchins’ feeding behaviour. Overall, we found that the individual experience has a primary role on food acceptance and feeding behaviour: observing the others eating motivates the observer to eat, but does not drive his/her food choices. Independently of social context, capuchins are equipped with the behavioral and physiological tools necessary to select energy-rich food and avoid deleterious ones indipendently from the social context. Nevertheless, other social influences are still at work. Social facilitation (i.e., repeating a familiar behaviour when other individuals perform it) and stimulus enhancement (i.e., the phenomenon according to which the observation of another individual acting on an object draws the observer’s attention to that object, thus evoking an independent response) increase the chances that a naïve individual would feed at the same time and place its group members do, with the result that his/her food choices would be similar to those of the others. In summary, the individual learning of food selection is socially biased: capuchins learn with others rather than from others.
The same thing happens for the response to novel foods. Capuchins are neofobic and the acceptance of a novel food is socially facilitated (i.e., it is easier for a capuchin to eat novel food if its group members are eating it nearby). However, this does not necessary mean that capuchins learn whether to eat a food A or a food B from what other group members are eating. Nonetheless, social influences may speed the inclusion of novel foods into the individual's diet.
I like to be imitated!
Two experimenters stay in front of a capuchin monkey, who chooses to stand closer to one of them. Who is the lucky one? The one who imitates him! Capuchins show affiliation toward humans who imitate them, a propensity that may be a prerequisite for altruistic behaviour.
During social interactions, humans often unconsciously and unintentionally imitate the other's behaviour of others. This effect gives strenght to the relashionship, bonding and empathy between interaction partners, and it is thought to be an evolutionary adaptation that facilitates group living and may be shared with other primate species. An experiment conducted in the Primate Center showed that capuchins look longer at human imitators over the non-imitators, they spend more time in proximity of imitators, and choose to interact more frequently with imitators in a token exchange task than to non imitators. In other words, capuchins prefer imitators than non imitators. These results demonstrate that imitation can promote affiliation in non human primates. Behavior matching that leads to prosocial behaviours toward the others may have been one of the mechanisms at the basis of the prosocial behavioural tendencies in capuchins and in other primates, including humans.
Researchers involved
Elsa Addessi, Gloria Sabbatini and Elisabetta Visalberghi
Reciprocity in primates: do unto others as you would have them do unto you
Primates have a long memory. If someone does somebody else a favour, the latter will remember it, and return the favour over extended periods. This reciprocation works better if the monkeys are emotionally involved with their partners.
Apparently primates (and other animals) often behave for the benefit of another individual. For example, they groom and help each other during fights. The evolution of cooperation has been an important topic of research for the last 50 years, because it is not immediately obvious how natural selection could favour such cooperative behaviours. Scientists have devised several possible explanations for the evolution of cooperative behaviours. In this research area one of the most controversial topics is reciprocity, the basic idea being that apparently altruistic behaviours can be favoured by natural selection if animals can switch the roles and return the favour.
Although appealing, this hypothesis was criticized because of the lack of convincing examples and because reciprocity was believed to be too difficult (cognitively speaking) for animals. Over the last ten years, we have addressed these issues using a variety of different approaches. We relied on meta-analyses of published data to show that primates do in fact reciprocate one of their most common cooperative behaviour, grooming, and also exchange it for help during fights. We carried out naturalistic observations and experiments, followed by detailed analyses of the time frame of the reciprocation of various cooperative behaviours (grooming, coalitions, tolerance over food sources) to show that monkeys are able to reciprocate previously received favours over an extended period of time. Finally, we carried out several experiments exploring the emotional consequences of receiving cooperative and uncooperative interactions and testing the cognitive basis of reciprocity.
In primates, and probably in other animals, the choice of a preferred partner seems to be based on the emotional and social bond that monkeys develop with their group mates. Our studies showed that the cognitive requirements of reciprocity were probably overestimated and that reciprocity can well be based on an emotionally based account of cooperation received. Reciprocity appears to be a widely spread and well-developed phenomenon in primates and in many other animals.
You bully me, but is there someone who bullies you?
When a macaque is attacked by another group companion, he can ask for help using specific types of behaviour and vocalizations. But who should he ask? If the macaque is keen, he will ask for help neither a subordinate nor a kin of his opponent. That is it. Macaques, indeed, do understand the social relationships that link toghether other individuals.
We, human scientists, commonly describe the social organization of primates in terms of hierarchies of dominance (who bullies who) and of kinship (who is a relative of whom). Dominance and kinship are in fact believed to be two main pillars of primate societies. Although we have known for decades that dominance hierarchies and kinship are the main determinants of primate social relationships, there is still a question to answer: “what monkeys know about social relationships?”.
In principle, we can imagine two possibilities: first, monkeys could have a “self-centred” view of their social world, knowing only who is dominant (or subordinate) to them, or who are their own kin, but would have no idea of the social relationships that link the other individuals. Second, monkeys could have an “objective” view of their social world, and be aware that also the other individuals have their own kin and their own dominance relations. According to the first hypothesis, dominance hierarchies and kinship structures would be unknown to monkeys, which only understand their personal social relationships. According to the second hypothesis, monkeys would have some form of mental representation of the social structure of the group they live in. Given that we cannot ask the monkeys directly, these are hypotheses that are not easy to test. We took advantage of the recruitment behaviour of Japanese macaques, to try to understand what they know about social relationships.
When a macaque is attacked by a group companion, it can “ask for help” using specific types of behaviour and vocalizations. Who a macaque asks for help informs us about its ability to recognize the dominance and kin relations of its companions. In fact, if a macaque knows the dominance and kin relations of its group companions, then it should ask for help individuals that are dominant over their opponent (in contrast to being simply dominant to themselves) and individuals that are not the kin of their opponent. Japanese macaques did follow these rules in recruiting their allies, thus showing in this way, that they have an “objective” view of their social world and some understanding of the social relationships that link other individuals.
In later studies, we also showed that monkeys are very interested in observing the social interactions of their group mates, that they build their extensive social knowledge by observing others, and that those individuals having the best social knowledge also acquire a definite social advantage.
Cooperation in solving a task: I know you might help me, but I do not know precisely how
In order to get a food reward, two capuchins have to simultaneously pull two handles.Do they succeed in teamwork and do they learn to coordinate their action to reach the goal? Not at all. They learn to take into consideration the presence of the partner by the handle but not his behaviour. But with some help...
Figure. Pippi and Paquita solving a cooperation task (Photo by Elisabetta Visalberghi).
A study conducted in our Primate Center investigated capuchin monkeys’ cooperation abilities. A pair of capuchins faced a cooperative task: two monkeys that were free to move in a big cage had to jump on a platform and simultaneously pull two handles to obtain food. It turned out that, to some extent, they learn to take into consideration the presence of the partner by the handle but not the partner’s behaviour. When a monkey saw the other one coming on the platform, it pulled the handle more often when the other monkey was getting near the handle itself (although it didn't programmaticly pull the handle at the same time the other monkey did. It is important to underline that both of them were still willing to pull the handle even if their partner was no longer on the platform. This experiment suggested that capuchins’ success didn't rely on understanding the role of the partner.
When the experimental paradigm was slightly modified, capuchins started to achieve some success. In a study of Mendres and de Waal (2000), capuchins faced the same task but with some differences: they were housed in a smaller cage where they weren't free to move; when only one of them pulled the handl, it weighted more than when they pulled it toghether; if one individual pulled for some seconds and the other joined after a while, then they both were successful. These kinaesthetic information proved to be essential for capuchins’ understanding of each other role in a cooperative task. Probably then, the duration and the multimodal nature of the feedback that each actor gets from partecipating in a cooperative task, improve monkeys learning ability.
Conflict management: capuchin monkeys use grooming as a declaration of peace
Capuchin monkeys have a peaceful way to prevent conflicts that may arise in stressful situations such as feeding competition: they groom each other. Indeed, they make grooming, not war.
Figure. Grooming in a group of Sapajus libidinosus (left; photo Noemi Spagnoletti) and Sapajus nigritus (right; photo Brandon Wheeler)
Conflict management represents an important set of mechanisms that help group-living animals to cope with the negative effects of competition and to maintain the benefits of group life. We investigated conflict management in capuchin monkeys, both in captivity and in the wild, by combining the study of conflict prevention and of post-conflict resolution. In this way, we showed that these two sets of mechanisms are both part of the behavioral repertoire of primates.
These studies showed that capuchin monkeys are able to manage the negative consequences of impending feeding competition using grooming as an efficient instrument of conflict prevention. Nevertheless, when aggression did occur, capuchin monkeys coped with its negative consequences by reconciling with their former opponent, and such reconciliation successfully reduced post-conflict anxiety. Also, by experimentally manipulating the visibility between neighboring groups, we showed that, contrary to expectations, hostility between groups did not result in increased group cohesion (to deal with the external threat), but in contrast translated into increased within-group aggression and tension.
A poster showing results of the experiment on visibility (in italian)
Conflict management: the complex consequences of aggression in mandrills
In mandrills, aggression is not a dyadic affair. When two mandrills have a fight, the consequences can spread across the group and involve multiple individuals.
Other studies on conflict management focused on mandrills. After a fight, mandrills used different behavioral patterns to reconcile depending on the identity of their former opponent and the risk of approaching it. They used grooming to reconcile with their kin and visual signals exchanged from a distance to attain reconciliation with those opponents that most frequently renew their aggression. Mandrills seem therefore to be aware of the danger that the group mates represent for them and behave accordingly.
On receiving aggression, mandrills often "redirect" this aggression to another individual, that is, they attack others. We showed that group mates can approach and exchange friendly behaviors with victims of aggression in order to prevent being the target of this redirected aggression. We also discovered that redirected aggression is often targeted at the relatives of the original aggression. In this way, mandrills can "punish" their aggressors and prevent further aggression.
Overall, mandrills seem to have a rather complex set of behaviors used to manage and resolve their conflicts.
Reciprocity in primates: do unto others as you would have them do unto you
Primates have a long memory. If someone do somebody else a favour, the latter will remember it, and return the favour over extended periods.This reciprocation works better if the monkeys are emotionally involved with their partners.
Apparentely primates (and other animals) often behave for the benefit of another individual. For example, they groom and help each other during fights. The evolution of cooperation has been an important topic of research for the last 50 years, because it is not immediately obvious how natural selection could favour such cooperative behaviours. Scientists have devised several possible explanations for the evolution of cooperative behaviours. In this research area one of the most controversial topics is reciprocity, the basic idea being that apparently altruistic behaviours can be favoured by natural selection if animals can switch the roles and return the favour.
Although appealing, this hypothesis was criticized because of the lack of convincing examples and because reciprocity was believed to be too difficult (cognitively speaking) for animals. Over the last ten years, we have addressed these issues using a variety of different approaches. We relied on meta-analyses of published data to show that primates do in fact reciprocate one of their most common cooperative behaviour, grooming, and also exchange it for help during fights. We carried out naturalistic observations and experiments, followed by detailed analyses of the time frame of the reciprocation of various cooperative behaviours (grooming, coalitions, tolerance over food sources) to show that monkeys are able to reciprocate previously received favours over an extended period of time. The choice of a preferred partner is probably based on the emotional and social bond that monkeys develop with their group mates. Our studies showed that the cognitive requirements of reciprocity were probably overestimated and that reciprocity can well be based on an emotionally based account of cooperation received. Reciprocity appears to be a widely spread and well developed phenomenon in primates and in many other animals.
You bully me, but is there someone who bullies you?
When a macaque is attacked by another group companion, he can ask for help using specific types of behaviour and vocalizations. But who should he ask? If the macaque is keen, he will ask for help neither a subordinate nor a kin of his opponent. That is it. Macaques, indeed, do understand the social relationships that link toghether other individuals.
We, human scientists, commonly describe the social organization of primates in terms of hierarchies of dominance (who bullies who) and of kinship (who is a relative of whom). Dominance and kinship are in fact believed to be two main pillars of primate societies. Although we have known for decades that dominance hierarchies and kinship are the main determinants of primate social relationships, there is still a question to answer: “what monkeys know about social relationships?”.
In principle, we can imagine two possibilities: first, monkeys could have a “self-centred” view of their social world, knowing only who is dominant (or subordinate) to them, or who are their own kin, but would have no idea of the social relationships that link the other individuals. Second, monkeys could have an “objective” view of their social world, and be aware that also the other individuals have their own kin and their own dominance relations. According to the first hypothesis, dominance hierarchies and kinship structures would be unknown to monkeys, which only understand their personal social relationships. According to the second hypothesis, monkeys would have some form of mental representation of the social structure of the group they live in. Given that we cannot ask the monkeys directly, these are hypotheses that can't be easily tested. We took advantage of the recruitment behaviour of Japanese macaques, to try to understand what they know about social relationships.
When a macaque is attacked by a group companion, it can “ask for help” using specific types of behaviour and vocalizations. Who a macaque asks for help informs us about its ability to recognize the dominance and kin relations of its companions. In fact, if a macaque knows the dominance and kin relations of its group companions, then it should ask for help individuals that are dominant over their opponent (in contrast to being simply dominant to themselves) and individuals that are not the kin of their opponent. Japanese macaques did follow these rules in recruiting their allies, thus showing in this way, that they have an “objective” view of their social world and some understanding of the social relationships that link other individuals.
If you care, yawn me back!
Contagious yawning is a signal of empathy. Look for “contagion” in humans and gelada baboons and you will find that it is not a matter of spatial proximity, age or sex…what matters is social bonding. Friends yawn together more than strangers!
Figure. (From Left to right) Yawn in ring-tailed lemur (photo Elisabetta Palagi), sifaka (photo Ivan Norcia) and in gelada baboon (photo Elisabetta Palagi)
The emotions, or the empathy for other individuals, is crucial for complex social interactions. It was suggested that yawn contagion in humans (homo sapiens) is related to our emphatic capacities. We run a study to find out the effect of different variables (e.g., country of origin, sex, yawn characteristics) on yawn contagion. Overall, the results demonstrated that yawn contagion is affected primarily by the emotional closeness between individuals (the rate of contagion was greatest in response to a kin, then friends, then acquaintances, and at last strangers), while it is not affected by other variables, such as gender and nationality.
It is presently unclear whether this capacity is uniquely human or if we share it with other primates, especially monkeys. Recently, we discovered that in gelada baboons (Theropithecus gelada) yawning is contagious between individuals, especially those that are socially close, as it occurs in humans. In particular, we found that the contagiousness of yawning is correlated to the level of grooming. This indicaties that, it is the emotional proximity rather the spatial proximity, that mainly affects yawn contagion in this particular species of baboons. Adult females, which represent the core of the geladas’ groups, have an enhanced sensitivity and emotional tuning toward companions. In fact, adult females showed a precise matching of different yawning types, suggesting the existence of a mirror mechanism that activates shared representations at a neural level. All these findings support the idea that contagious yawning reveals an emotional connection between individuals. This phenomenon, here demonstrated in monkeys, could be a building block for a full-blown empathy.
In play we trust!
Playing and grooming: such different types of behaviour for the same purpose. As for grooming in capuchin monkeys, in fact, bonobos, gorillas and chimpanzees play to prevent conflict and favour a peaceful resolution of stressful situations.
Figure. Sequence of play in captive gorilla (photo Elisabetta Palagi)
Playing behaviour is an ontogenetic trait in many mammalian species and it is widely believed to have an important role in the assembly of adult behaviour. However, play may be at work also during some peculiar situations that favour cohesion and social manipulation. We demonstrated that in bonobos, chimpanzees, and gorillas social play increases around feeding time and, as it occurs for grooming in capuchin monkeys, play is a good tool to limit the probability of aggression and to favour a peaceful co-feeding.
Another topic of research focuses on the study of the ontogenic playing trajectories of chimpanzees in comparison with humans. Chimpanzee play, both solitary and social, shows many changes from infancy to juvenility. While solitary play peaks during infancy, social play does not show any quantitative variation between infancy and juvenility but shows a strong qualitative variation regarding the complexity, asymmetry, and playmate choice. The playful expressions in chimpanzees (that vary troughout the different age phases) seem to have a role in advertising cooperative dispositions and intentions (like laughter in humans) thus increasing the possibility of engaging in solid social relationships. In chimpanzees, as in humans, both playing behaviour and the signals that accompany it serve multiple functions according to the different age phases.
Capuchins’ facial displays
Robinia, a female capuchin, looks in your eyes raising her eyebrows and forehead. What is she doing? She is trying to court you while displaying her most gorgeous scalp lift! Capuchins, indeed, are among the monkey species with the richest repertoire of facial displays.
Figure. (Left) Relaxed open-mouth, in which the upper tooth rows remained covered by the upper lips, is performed during play interactions. (Right) Silent bared-teeth is used by capuchins during affiliative contexts as well as to signal submission to the dominant (drawings by Arianna De Marco).
Facial displays are important for communication, and their ontogeny has been studied primarily in humans, chimpanzees and macaques. We investigated the ontogeny, communicative function and target of facial displays in two species of capuchin monkeys: the tufted capuchins (Cebus apella) and the white-faced capuchins (Cebus capucinus). In Cebus apella facial displays are absent at birth and develop as infants grow older. Lip-smacking (which is associated with an affiliative type of behaviour both when performed and received by a young capuchin) appears at about 1 month of age, followed by scalp-lifting (affiliative behaviour), relaxed open-mouth (affiliative and playing behaviours), silent bared-teeth (affiliative behaviour), open-mouth silent bared-teeth displays (playing, affiliative and submissive behaviours) and finally the open-mouth threat face (agonistic behaviour). All these facial displays are reciprocated except the agonistic one. Infants of tufted capuchin perform most facial displays in the same contexts of the adults, with an exception for the silent bared-teeth display that youngsters use primarily, or exclusively, in affiliative contexts. Interestingly, facial displays are exchanged very often with peers, less frequently with the adults and almost never with the mother.
Research on the white-faced capuchins, was carried out in the Primate Centre of the Louis Pasteur University of Strasbourg, in France. We recorded both facial displays commonly used by the tufted capuchin (such as relaxed open-mouth, lip-smacking, open-mouth threat-face, silent bared teeth and open-mouth silent bared-teeth) and new ones as the protruded-lip face and the tongue-out. Surprisingly, researchers never observed the scalp-lifting display, one of the most common displays characterizing tufted capuchins. Generally, facial displays have the same communicative functions in both species. But there are some differences. For example, while in the tufted capuchin the silent bared-teeth indicates submission as well as affiliation, in the with-faced capuchin conveys only a positive message. Indeed, white-faced capuchins seem to lack of ritualized signals of submission. Morphological and functional variations in the facial displays of different species, are related to the different social-dynamics of the tufted and white-faced capuchins.
Researchers involved
Dr. Arianna De Marco, Dr. Luciana Massaro, Dr. Elisabetta Palagi, Dr. Eugenia Polizzi di Sorrentino, Dr. Gabriele Schino and Dr. Elisabetta Visalberghi
How to spend a token? Capuchins know it very well
Sandokan, a young male capuchin, is presented with a blue plastic poker chip and a metal nut. He has learned that the poker chip stands for one piece of peanut, and that the metal nut stands for three pieces of the same food. Which one does he choose to return to the experimenter in order to obtain food? And what happens if he has to choose between four plastic chips and one metal nut? We discovered that Sandokan knows how to maximize his choices by using tokens as symbols to flexibly combine quantities.
We investigated capuchins’ comprehension of symbols, which could be defined as “something that someone intends to represent something other than theirselves”. We focused on capuchins’ use of tokens, which can be considered symbols since they arbitrarily stand for something else without having any iconic relation to their referent. In a typical task, subjects had to choose between various combinations of tokens, each standing for one or more pieces of the same food. Pay-off maximization required the assessment of the value of each offer by estimating token numerousness, representing what each token stands for and making simple computations. Capuchins managed to maximize their choices, demonstrating they can use tokens as symbols to flexibly combine quantities. But are they able to reason about symbols in ways similar to how they reason about real objects?
Figure. (Left) Gal, an adult male, has a choice between two brass plugs, on the right, corresponding to three pieces of food each and four grey PVC cylinders, on the left, corresponding to one piece of food each. (Right) Gal has selected the better option by pulling the tray with the two brass plugs and exchange them one at a time with the experimenter, thus obtaining a total of six pieces of food (not shown).
In another study, we assessed whether, in a symbolic context, capuchins’ preferences satisfy transitivity, a fundamental trait of rational decision-making. Monkeys were trained to exchange three types of tokens for three different foods (options were A, B and C, where A was preferred over B, B over C and A over C), then we compared choices monkeys made between different types of tokens with choices monkeys made between the foods. Qualitatively, capuchins’ preferences revealed with tokens were similar to those measured with the foods. Quantitatively, however, values measured by the way of tokens differed from those measured with the actual foods: for any pair of foods, the relative value of the preferred food increased when monkeys chose between the corresponding tokens. Thus, capuchins are capable of treating tokens as symbols but they experience the cognitive burdens imposed by symbolic representation.
Capuchins know objects and pictures are not the same
In front of a capuchin there is a little blue cork. Beneath the object two pictures are displayed: one representing the cork, the other an apple. Is the monkey able to match the object with its two dimensional representation? Yes! And be sure that he is not confusing between the 3D object and the 2D image.
Although pictures are frequently used in place of real objects to investigate various aspects of cognition in different non-human species, there is little evidence that animals treat pictorial stimuli as representations of the real objects. We carried out a study to investigate if capuchin monkeys are able to match objects with their colour photographs and vice versa, without confusing them. They do, as our results demonstrated. Moreover, capuchins could solve the task with a high level of accuracy even when less realistic images, such as black-and-white photographs, silhouettes and line drawings, were employed as bi-dimensional stimuli.
In the above video Pippi, a female capuchin monkey, is dealing with a matching-to-sample task. To solve it, she is required to touch the picture (bottom right) which depicts the object presented in the sample (top).
Where to search? Watch a video, it might help you (even if you are a capuchin)
Robot, a male capuchin, stands in front of a television screen. He is not watching a movie, but a video with the experimenter hiding a food reward under one of two different cups. Then, the screen is turned off and the monkey is presented with two real cups, the same ones he has just watched in the video. Is he able to use the information acquired trough television to choose the right cup and obtain the food? Yes, he is.
A vital function of symbols is to allow to acquire information without direct experience. The life of modern humans is strongly influenced by the use of symbols, but what about nonhuman primates? We examined capuchins’ use of videos and scale models as symbolic artefacts to obtain information about reality.
In our experiments, we assessed if and to what extent capuchins are able to use videos and scale models to solve a search task. In the first study, capuchins watched video clips showing an experimenter hiding a food reward under one of two different cups, whose positions were randomly changed. Then, they were asked to choose between two real cups (the same ones shown in the video) in order to obtain the reward. After training, one subject was able to solve the task. This demonstrates that he established a correspondence between the video and its referent.
Another study on capuchins’ use of scale models, which is a miniature representation of a larger environment, led to similar results. Capuchins used the information about the location of hidden food obtained via a scale model to find the food in the real situation.
Researchers involved
Dr. Elsa Addessi, Dr. Patrizia Potì, Dr. Valentina Truppa, Dr. Serena Gastaldi and Dr. Elisabetta Visalberghi
Spatial relations: capuchins are their own frame of reference
Consider an imaginary square delimited by four identical landmarks. Gal, a male capuchin, is trained to search in the middle to obtain a food reward. Now the square is stretched by moving away the landmarks. Where will he search? Not in the middle, but near one of the landmarks, thus preserving the relative distances and orientations from each one. With such a complex configuration, capuchins are not able to taking into account the spatial relations between all landmarks. But with simpler configurations.
We investigated capuchins’ use of different spatial reference frames to locate objects in space. Researchers usually recognize two basic frames of spatial reference humans use to find objects: an egocentric frame, which relies on the body (e.g., on my left, in front of me, below me), and an allocentric frame, which is based on the relation between objects (e.g., the object is near the wall, in a tree, under a bush). Which one do capuchins rely on? Up to now, the results suggest that capuchins mainly rely on a self-referenced spatial frame to code and remember positions of objects in space. In particular, our experiments results suggest that capuchins search next to individual landmarks to find hidden food, but they do not exploit landmarks’ configurations. However, they can use landmarks’ configurations given a series of conditions: that is if only two landmarks are present in a small scale space and their position can vary but the line connecting them has always the same orientation (e.g. vertical) with respect to the subject’s point of view.
Figure. Configuration 1 (left): Gal has learned that in the central hole of the imaginary square delimited by four red sticks (references) is a hidden reward. Configuration 2 (right): if the sticks are moved away, Gal is no longer seeking the center hole of the new configuration but close to the reference, taking from them the same distance as he learned during training (a hole).
Inter-objects spatial relations: capuchins are not so good architects
Which is the best way to investigate primates’ comprehension of spatial relations between objects? Give them some blocks to play with and observe the constructions they made...
We investigated primates’ constructing inter-object spatial relations as indicated by spontaneous play with blocks. Findings indicate that species and age influence the degree to which primates rely on their own action and body to place objects in space. In particular, capuchins and young chimpanzees move and place objects together with the same part of the body and/or using their body as a support surface, so that inter-object relations are coincidental to forming relations between objects and their body. Older chimpanzees produce more complex constructions but, compared to human children, they too are relatively impaired in constructing object relations that cannot be directly supported by action coordination or functional relations (i.e. containment and support relations).
Video. Researchers try to understand how chimpanzees realize on spatial relationships between objects, from the way they combine theme. The video shows how Panzi builds a tower by placing the bowls on top of each other.
Researcher involved
We analyze behavior to gain insight into the intelligence behind it. In particular, we study how human and nonhuman primates grasp and use objects and we do this to learn about action planning and control. We believe that a useful way to study action planning is to observe changes in behavior as a function of the behavior that follows. If an action differs depending on the subsequent action, the anticipatory effect can be said to reflect planning.
Altering an object manipulation not only on the basis of the immediate task demands but also of the next action to be performed requires second-order motor planning abilities. Thus, the form of an organism’ response allows measuring anticipatory abilities. Grasping an object seems so simple that we rarely think about it. Yet even casual observation of very young children shows that the skill takes time to develop. When does grasp planning begin to appear, both in individuals (ontogenetically) and over the course of evolution (phylogenetically)?
Second-order motor planning in humans develops over years and what makes it so difficult for young children to comfortably and differentially adjust their initial grasps in several manipulation tasks is a matter of ongoing research and debate. Studies on human and nonhuman primates have demonstrated that the rudimentary motor planning abilities appear to be shared across species. These studies have shown that subjects grasp objects to be moved from one location to another in a way that affords a “comfortable” final posture and a good control on the objects, even if this involves an “uncomfortable” initial posture. This effect has been termed the end-state comfort effect by Prof. David Rosenbaum in 1990.
We investigated second-order planning abilities in capuchin monkeys both in captivity and in the wild. Capuchins were first tested in a task requiring individuals to grasp a dowel inserted into a vertical tube and to bring the baited tip at the bottom to the mouth. They spontaneously performed actions in which the forearm was in pronation or radially rotated to turn over the dowel and eat the food from its baited tip. In general, these comfortable actions, in which the initial orientation of the forearm afforded a more comfortable final position in bringing the dowel to the mouth, were performed as frequently as the uncomfortable actions. It is possible that capuchins’ higher manual dexterity may elicit less consistent second-order motor planning in this task, as species that show a low rate of second-order planning have more means of compensating for inefficient initial postures.
In a following study, in a task in which capuchins had to choose how to grasp a horizontal baited dowel containing food in its right or left end, they used a radial grasp (i.e., with the thumb-side oriented towards the baited end) with the forearm in pronation to bring it to the mouth. Similarly to 19 months old human infants tested in an analogous task, capuchins switched the hands between trials to use a radial grip with the forearm in pronation. This suggests that object orientation is an important constraint guiding action selection in capuchin monkeys.
Video 1 from Sabbatini et al 2016 Behavioural Brain Research
Cammello, a male capuchin, uses a radial grasp with the forearm in pronation to bring the baited right end of the dowel to the mouth.
Age-related differences in nonhuman primates have been poorly investigated. We investigated age-related differences in motor planning of wild capuchin monkeys in the dry forest habitat of Piauí (Brazil) by recording their strategies to bring a baited stick to the mouth. In particular, we investigated anticipatory motor planning of infant, juvenile and adult wild capuchin monkeys grasping a horizontally-positioned stick baited to the left or right side. We recorded the grasps capuchins used to bring the baited end of the stick to the mouth. We found that motor planning abilities improved with age, as in humans, and adult capuchins used efficient grips significantly more frequently than infants.
Video 2 from Truppa et al 2020 Developmental Science
Cacau, an infant male capuchin, uses a radial grasp to bring the baited end of the stick to the mouth.
We study also what capuchins do when they have to use a dowel as a tool to interact with an out-of-reach food target. Capuchin monkeys showed action-selection planning when using tools to interact with distally located targets: they consistently used a radial grip (with the thumb-side oriented towards the centre of the dowel) to grasp a dowel that was positioned horizontally at different orientations and to dislodge an out-of-reach food reward.
Video 3 from Sabbatini et al 2016 Behavioural Brain Research
Sandokan, a male capuchin, uses a radial grasp with the forearm in pronation to dislodge the out-of-reach food reward.
An intriguing issue to consider is that capuchins showed more pronounced planning abilities when the principal axis of the dowel is aligned with the horizontal plane compared to the vertical plane. How the spatial orientation of the object can affect monkeys’ motor planning abilities in grasping actions seems to be a crucial point that deserves further investigations.
Researchers involved
Dr. Gloria Sabbatini, Dr. Valentina Truppa and Dr. Elisabetta Visalberghi
- Scientific publications
- Truppa, V., Spinozzi, G., Laganà, T., Piano Mortari, E., Sabbatini, G. (2016). Versatile grasping ability in power grip actions by tufted capuchin monkeys (Sapajus spp.). American Journal of Physical Anthropology, 159: 63-72. https://doi.org/10.1002/ajpa.22836
- Truppa V., Spinozzi G., Laganà T., Piano Mortari E., Sabbatini G. (2016). Flessibilità nell'uso delle prese di forza nei cebi dai cornetti (Sapajus spp.), dallo studio della manipolazione degli oggetti al legame tra cognizione e azione. Sistemi Intelligenti, XXVIII: 69 - 82. DOI: 10.1422/83836
- Sabbatini G., Meglio G., Truppa V. (2016). Motor planning in different grasping tasks by capuchin monkeys (Sapajus spp.). Behavioural Brain Research, 312: 201-211. https://doi.org/10.1016/j.bbr.2016.06.010
- Truppa V., Sabbatini G., Izar P., Fragaszy D., & Visalberghi E. (2020). Anticipating future actions: Motor planning improves with age in wild bearded capuchin monkeys (Sapajus libidinosus). Developmental Science https://doi.org/10.1111/desc.13077
Capuchins are erotic artists
“I had always considered myself free from prejudices. But when I saw a capuchin female courting two males - both rather reluctant - in turn I could not believe my eyes. She filtered, she insisted, and after much frustration, she harassed one and then the other. The amazing variety of facial expressions, vocalizations, and charming gesture that she directed at him had little apparent effect; he just ignored her or turned away”. Elisabetta Visalberghi from “The Complete Capuchin. The Biology of the Genus Cebus” (Cambridge University Press, 2004).
Figure: Courtship behaviour and ventro-ventral mating. Drawings from videos by Andy de Paoli
Even if males are often as enthusiastic as females about mating, the vignette reflects what researchers studying Cebus apella often witnessed: an astonishing richness of sexual behaviour, especially in females. In fact, in tufted capuchin females the ovarian cycle is not signalled by external morphological changes or scent marking behavior. Instead, the females courts the males with an extremely rich repertoire of proceptive behaviours.
Figure: The female (left) and the male (right) are courting each other's by performing eye-brow raising and grin. Drawings from videos by Andy de Paoli
We carried out a parallel study between females' behaviour and their ovarian hormones, discovering that several behaviours had a cyclicity (21.3 days) marking that of urinary progesterone (21.9 days). Moreover, there was a set of behaviours that showed a dramatic increase during the periovulatory phase. These findings demonstrated that female behavior is a good indicator of the periovulatory phase, and that it enhances female attractivity. Each female made a different use of the behavioural repertoire by performing some behaviours significantly more than others. This variability during courtship calls for further research regarding how much do these kind of behaviours affect mating success. Males seemed to be sensitive to the behavioural/hormonal state of the female. Although they mounted females both during the periovulatory and the non-periovulatory phase, there was a significant periovulatory phase effect for mounts not associated with play, and ejaculations always occurred within proceptive periods.
Video "Wild capuchin monkeys (Sapajus libidinosus) courtship. Female's solicitation"
The video was filmed at Fazenda Boa Vista, in the North-East of Brazil. Piassava, a wild Sapajus libidinosusfemale, shows sexual interest toward Jatobà, the alfa male. The male but he does not reciprocate behaviorally (see also pictures in the gallery). This phase of courtship is characterized by the female visually monitoring the male, vocalizing, seeking his proximity and displaying persistent eyebrow raising in his direction. Some females trigger the male's attention by using stones and/or branches. Visalberghi and colleagues studied. Visalberghi and colleagues (2017) investigated how FBV females use objects to solicit males and showed that affiliative behaviors are occur immediately before or after this use the objects. Thus throwing or pounding stones and pushing or dropping branches by females performed in the sexual context have a clear affiliative meaning. It would be important to determine whether the use of objects is present in capuchin populations that do not use tools.
Filmed and edited by Noemi Spagnoletti/Ethocebus project
Fazenda Boa Vista, Piauí, Brazil 2010
Researcher involved
Video "Sexual behaviour in the South American primate, Cebus apella"
A cura di Elisabetta Visalberghi & Monica Carosi, Unità di Primatologia Cognitiva, Istituto di Scienze e Tecnologie della Cognizione, CNR, Roma.
Capuchins are natural born tool users
“Some of those cats (monkeys) are so astute that many things they see man do, they imitate and also do. In particular, there are many that when they see how to smash a nut or a pine nut with a stone, they do it in the same way and, when leaving a stone where the cat (the monkeys) can take it, smash all that are given to them” (Sumario de la Natural Historia de las Indias, de Oviedo 1526/1996). This is the first report of tool use in Cebus, dating about 500 years ago. After hundreds of years of research, we know that capuchins are very skilled at manipulating objects in captivity.
We investigate the behavioural and cognitive ingredients of capuchins’ success in tool use and the extent to which naive individuals benefit from expert models. Overall, our results show that capuchins use tools in a very flexible way, generalizing how to use an object from one context to another. They can also solve the same task using different objects and they can also modify an object in relation to the task to be performed. Moreover, capuchins demonstrate a good appreciation of the relation between the tool properties and the task requirements. For example, they are able to efficiently select a tool on the basis of a non-visual property such as rigidity. When give them a chance to manipulate new stick-like tools differing in flexibility, they were able to choose the right one to insert into a tube to obtain food both manipulating the tools theirselves or observing a human demonstrator repeatedly bending them.
Video. The experimenter introduced the three tools inside the compartments so that Gal, an adult male, could freely manipulate them. Only one of the three tools, the flexible one, is useful to reach the yogurt in the baited apparatus, a tube presenting an angle of approximately 90°. Once Gal stopped manipulating the tools, they were retrieved and placed on their preassigned location on the sliding platform. After Gal chose one of the three tools, the experimenter gave the selected tool to him and allowed him to move to the baited tube.
Video. The experimenter manipulated the three tools in front of Pedro, an adult male. Only one of the three tools, the rigid one, is useful to rake in a food reward (a 1-cm-thick banana slice) located on a rectangular wooden platform beyond the cage mesh. Once the experimenter stopped manipulating the tools, Pedro could select the desired tool by inserting a finger into the corresponding hole. After the experimenter gave the selected tool to him, Pedro could go to retrieve the banana from the table.
In 2005, the members of the EthoCebus Project observed wild bearded capuchin monkeys (Cebus libidinosus, now Sapajus libidinosus) living in the dry woodland habitat of Piaui (Brazil) routinely using stones to crack palm and other hard fruits they have previously placed on stone or wood anvils. As in captivity, capuchins proved to be very expert users. For example, when faced with stones differing in functional features such as friability and weight, they choose the functional tool on the basis of material and weight, even when weight could not be judged by vision. Moreover, prior to pounding, the monkeys achieve a stable position of the nut on the anvil. Monkeys' strategic placement of the nut reveals their capacity to improve their tool-use skills based on an understanding of shape.
Stone tool use was the first to be observed systematically in a New World monkey species. Apart from capuchins, in fact, the only non human primates who use stones and anvils to process food in the wild were chimpanzees of Western Africa. Nowadays, the EthoCebus project is going on studying all the ecological, developmental, social, physical and historical implications of this behaviour (more on http://www.ip.usp.br/ethocebus/)
Figure. Wild capuchin monkeys using hammer stone on a log anvil (left) and a sandstone anvil (right) to crack open resistante palm nuts (photo Elisabetta Visalberghi).
Why does it take such a long time to discover that capuchins are so good at using tools in the wild? Because they are essentially arboreal: capuchins spend most of the time on trees, where it is more difficult to use tools.In contrast, capuchins living in Fazenda Boa Vista (Piauí) have terrestrial habits, an ecological condition which has favoured the emergence of this demanding behaviour. In fact, when Wright and colleagues (2019) studied the positional behavior and substrate use behaviors of these monkeys and found that they spend 27% of time on terrestrial substrates.
Figure. In Piauí capuchin monkeys spend some time on the ground, even playing and traveling (photo Noemi Spagnoletti)
Researchers involved
Dr. Elsa Addessi, Dr. Gloria Sabbatini, Dr. Noemi Spagnoletti, Dr. Valentina Truppa and Dr. Elisabetta Visalberghi
Capuchins’ manual skills: precision and asymmetry
Those who carefully observe capuchins, certainly appreciate their precision in moving objects. Although they fail the tip-to-tip grip seen in humans (and not for not being able but for anatomical reasons), capuchins possess functional precision grips, more than those showed by other New World species. Moreover, they have a favorite hand to rely on, when performing complex motor tasks.
Capuchin monkeys have the greatest degree of manual dexterity of all New World monkeys. Although their prehensile hand has a pseudo-opposable thumb, they possess both a precise and powerful grip and frequently make use of unimanual and bimanual forms of precision handling. In our Primate Center, we studied different aspects regarding hand use. We found that to perform grasping actions, capuchins use a wide variety of grips, which often involve distal areas of the hand. In particular, the precision grip most frequently used to grasp small food items, involves the distal lateral areas of the thumb and the index finger.
We believe that wild capuchins acquire their manual dexterity by acting on substrates, and by combining an object with the natural substrate, activities that they perform on a daily basis during extractive foraging. On the other hand, captive monkeys can earn the accolade “dexterous”, if trained with tasks involving the fine use of hands. Moreover, we found that both simple and complex grasping actions, induce the hand preference at an individual level, whereas only tasks involving more complex motor patterns are likely to induce an asymmetry in the distribution of lateral bias at group level. But we are still far from understanding the origins or the functional significance of motor asymmetry in non-human primates; nor if these asymmetries are similar to those found in humans.
A video on wild capuchins’ manual skills during nut cracking and food processing was presented at the temporary exhibition "La Mano – Arto, arte, artefatti” (July 27, 2013 – January 20, 2014) organized by MUSE (Museo delle Scienze di Trento). A contribution was also realized for the official catalogue of the exhibition (pdf) presented during the evening opening of MUSE, the day October 23, 2013.
Filmed by Alessandro Albani and edited by Elisabetta Visalberghi and Valentina Truppa/Ethocebus project
Researcher involved
Capuchins like junk food but remain cautious about it
In the Parque Nacional de Brasilia capuchins became smart-aleck. They learned that tourists could give them delicacies. It is much more easier than looking for food in the forest, especially during the dry season when fruit is scarce. But how does the human presence affect capuchins’ diet and behavior toward novel food? Well, thanks to the visitors they taste foods they would have never eaten in the forest, but they still remain cautious.
In the Parque Nacional de Brasilia there are some groups of wild capuchins, that are used to visitors’ presence. Capuchin monkeys habitually eat the the food that visitors give them, take food from the garbage or steal it from the visitors. In collaboration with the Parque and the Brasilia University, we investigated the interactions between capuchin monkeys and visitors in the Park, by means of interviews to the visitors and direct observations of the behaviour of capuchins and visitors. We also examined whether seasonal differences in the availability of wild food (that monkeys find in the forest), and whether the presence and abundance of human food (brought by visitors to the Park) affected capuchins’ activity budgets and diet.
In general, we found that capuchins living in the Park, spent less time foraging for wild foods than other groups living in similar habitats. Moreover, the capuchins living in the Park relied more on human food during the dry season, when pulpy fruits were less available, than in the wet season. Nearly 80% of the interactions observed involved the presence of food. There were some differences emerged between what the visitors reported in the interviews and what we observed. In the interviews most respondents (76.1%) reported that interactions were started by monkeys, while the analysis of direct interactions showed that 47.3% were initiated by visitors and only 39.6% by capuchins. Moreover, 83.9% of the visitors affirmed they didn’t feed capuchins, while 70.2% of them reported having seen other visitors feeding them.
These findings confirm other studies conducted on different monkey species, that demonstrate that access to human food decreases in relation to the time spent foraging for wild food and to the home range size. Moreover these findings demonstrate that capuchins are able to modify their diet, to exploit alternative food sources and to change their activity budget, in response to new food opportunities and to seasonal food availability. On the basis of the results listed above, it would be useful to establish an educational program that provides information regarding capuchins' behavior; underlining the consequences that feeding them could have on their ways of behaving and their interaction with visitors.
Video. In the National Park of Brasilia (Agua Mineral), monkeys usually eat human food from garbage cans. Despite that giving food to animals is forbidden, visitors of the park interact with monkeys giving them food. Video by Gloria Sabbatini.
In another study, we investigated the response of capuchins living in the Parque toward novel food. In fact given their daily exposure to human foods, we expected them to be more explorative toward novel food compared with capuchins that are not habituated used to visitors. However, since the safety and palatability of potential foods is to be learnt we expected capuchins to remain cautious about eating novel food, as it was reported for wild and captive populations. We gave capuchins familiar and novel foods on a platform, observing their behavior. Latency to contact the platform, and ingest food did not differ between novel and familiar stimuli. However, capuchins were less interested in manipulating the novel food than the familiar one, and consequently they ate only small pieces of the former while and they ingested big quantities of the latter. These results show that, despite capuchins used to humans’ presence become more explorative towards novelty, they still remain cautious about unfamiliar food, thus demonstrating that neophilia and neophobia are motivationally independent responses.
Wild capuchins use complex processing techniques to avoid caustic chemicals
Complex and flexible food processing enlarges the range of exploitable foods. Only a few primate species crack open encased food by using percussive tools and/or avoid physical contact with irritant compounds by removing the structures containing them.
Sirianni and Visalberghi discovered that the population of bearded capuchin monkeys (Sapajus libidinosus) living in Fazenda Boa Vista (www.ethocebus.net) accesses the nutritious kernel of cashew fruit (Anacardium spp.) avoiding the caustic chemical protecting it (present in the green mesocarp) (a, in the figure), or in the brown mesocarp when the fruit is dry.
Figure. The kidney-shaped fresh cashew nut is hanging below the bell-shaped pseudofruit (a). An adult male rubbing a fresh cashew nut (b). An adult female extracting the kernel with the index finger (c). Fresh nut with the hole (produced by rubbing) from which the kernel has been extracted; the caustic chemicals are contained inside the green mesocarp (d).
To access the very nutritious kernel capuchins use different processing strategies. When the fruit is not dry they rub it on a surface (b, in the figure) until a hole is produced and then use their index finger to extract the seed (c, in the figure). Later, when the fruit is dry they crack the shell with a tool and then extract the seed.
At both levels of ripeness, capuchins can easily break the fruit with their teeth. But in order to avoid the caustic liquid (when the nut is fresh) or the caustic resin (when the nut is dry) capuchins do not do use teeth. Infants and juveniles are by far less skilled than adults. This same set of processing strategies appears to be absent in other capuchin populations, making cashew nuts processing an excellent candidate for social transmission.
Researchers involved
Gloria Sabbatini and Elisabetta Visalberghi
What do capuchin monkeys feel when they take risks?
Emotion and cognition have traditionally been viewed as separate entities. Only recently has empirical research examined the role of emotions in decision making. In human beings, all decisions have affective consequences that can, in turn, impact subsequent decisions. In nonhuman animals, ,still little is known about the role of emotional responses on decision making. In a new study of the Unit of Cognitive Primatology, published on Animal Behaviour, researchers analyzed behavioral indexes of negative emotions, like scratching or hiccup vocalizations, in capuchin monkeys during a risky choice task. In this task, monkeys have to choose between a safe option (always constant) and a risky option that corresponds to a better or a worse prize. They found that capuchins show more negative emotions after choosing the risky option and receiving the worst prize than when they receive the better one. Moreover, capuchins try to switch their choice more often after receiving the worse prize than the good prize, a phenomenon known as "regret" in humans.
Video clip 1. Neutral condition. Robin hood, a male capuchin, is presented with the choice between a safe option (four food items covered by the white bowl, on the experimenter’s right) and a risky option (in this case, one food item covered by the red bowl, on the experimenter’s left). He chooses the risky option by inserting his finger in the hole of the corresponding transparent box, the experimenter provides him with the corresponding food reward (one food item) and Robin hood shows scratching behaviour.
Video clip 2. Disadvantageous condition. Gal, a male capuchin, is presented with the choice between a safe option (four food items covered by the yellow bowl, on the experimenter’s left) and a risky option (in this case, one food item covered by the pale blue bowl, on the experimenter’s right). He chooses the risky option by inserting his finger in the hole of the corresponding transparent box, the experimenter provides him with the corresponding food reward (one food item) and Gal shows switching behaviour (i.e., he inserts his finger in the hole of the previously non-selected box).
Capuchins prefer risky gambles over a safe reward
Pathological gambling affects 0.2% to 5.3% of adults in western societies, is highly comorbid with a range of other psychiatric disorders and with substance abuse, and is associated with increased suicidal ideation and attempts compared to the general population. In view of the growing incidence of pathological gambling and its severe mental and social consequences, it is urgent to further deepen our understanding of the neuronal and psychological underpinnings of this condition. Laboratory studies on nonhuman primates can inform the research on human pathological gambling in several ways.
The comparison of risk preferences between phylogenetically closely related nonhuman primate species with different ecologies can shed light on the selective pressures that shaped decision-making under risk in the course of the evolution. We presented 10 capuchin monkeys with choices between a "safe" option (always four food items) and a "risky" option (either one or seven food items) in three conditions differing for the probability of receiving the larger outcome when selecting the risky option (66%, 50% and 33%, respectively). When the probabilities of obtaining the larger outcome were 66% and 50% capuchins were risk prone, but when the probability was lowered to 33% they flexibly modified their preferences becoming indifferent to risk. Capuchins’ decision-making under risk mirrors their risk-prone behaviour in the wild, where they often rely on unpredictable and/or hazardous food sources.
The video shows Robin hood, a male capuchin, presented with the choice between a "safe" option (four food items covered by the white bowl) and a "risky" option (either one or seven food items with 50% probability, covered by the red bowl; in this case, one food item). He chooses the "safe" option by inserting his finger in the hole of the corresponding transparent box and the experimenter provides him with the food.
Sometimes, to be patient pays off
Is a bird in the hand worthier than two in the bush? Ask a capuchin and he might probably prefer two birds in the bush. When choosing between two rewards, giving that the favorite one is not immediately available, capuchins are willing to wait to obtain their preferred recompense. Indeed, they show a high tolerance to delay.
Both human and non-human animals often face decisions between options available at different times, in the animal world the capacity of delaying gratification has usually been considered one of the features distinguishing humans from other animals. However, this characteristic can vary across individuals, species, types of task, and it is still unclear whether it is accounted for by phylogenetic relatedness, feeding ecology, social structure, or metabolic rate.
To disentangle these hypotheses, we carried out an experiment to evaluate temporal preferences in capuchin monkeys. We gave capuchins the opportunity to choose between two food quantities (two versus six pieces of peanuts), the smallest was immediately available, the other delayed. It came out that subjects could wait up to two minutes in order to obtain the bigger reward. Overall, capuchins possess a significantely higher delay tolerance than closely related species, such as marmosets and tamarins, and that is likely to be compared to that shown by great apes. Capuchins’ tool use abilities might explain their comparatively high preference for delayed options in inter-temporal choices. Moreover, as in humans, capuchin females showed a greater delay tolerance than males, possibly because of their different fraging style. Males are more opportunistic than females: they spend more time on the ground foraging for exposed large invertebrates and small vertebrates, while females remain a few meters above the ground searching for small and embedded invertebrates. Thus, it is possible to explain the evolutionary origins of delay tolerance (or self-control) in terms of feeding ecology.
Video above. Paprica, a female capuchin, has to choose between a smaller immediate reward (two pieces of food, on the left) and a larger delayed reward (six pieces of food, on the right). When Paprica selects the larger delayed reward by inserting her finger in the hole of the box containing the food, the experimenter removes the non-chosen smaller reward, and – after a delay of 10 seconds – she pushes the sliding panel with the six pieces of food towards the capuchin.
To be self-controlled is much more easier when you do are not dealing with real objects, but with their symbolic representations. We carried out an experiment to see if symbolic artifacts such as tokens would inhibit capuchins’ strong behavioral predisposition, as occurs in children and chimpanzees. In our experiment, monkeys had to choose between two food quantities or two tokens, representing respectively the smaller and the larger food quantity. In all conditions, subjects had to select the smaller quantity (or the low-value token) to obtain the larger reward (or the amount of food corresponding to the high-value token). Symbolic artifacts did improve performance: most of the subjects succeeded with tokens, while only one succeeded with food. Thus, tokens allowed capuchins to achieve psychological distancing from the incentive features of food, leading them to avoid impulsive choices in favour of more advantageous alternatives.
Videos from Addessi & Rossi 2010 Proc. R. Soc. B-Biological Science:
Experimental phase - FOOD-condition.
Sandokan, a male capuchin, has to choose between five units of food (on the right) and two units of food (on the left). He makes the wrong choice, since he selects the larger amount of food and thus receiving in this way the non-chosen smaller amount of food as reward.
Experimental phase - LSDT-condition. Robinia, a female capuchin, has to choose between five low-symbolic distance tokens (on the left) and two low-symbolic distance tokens (on the right). Each low-symbolic distance token is worth one unit of food. Robinia correctly chooses the smaller amount of low-symbolic distance tokens, and receiving five units of food as reward, i.e. the amount of food corresponding to the non-chosen five low-symbolic distance tokens.
Experimental phase - HSDT-condition.
Sandokan, a male capuchin, has to choose between two high-symbolic distance tokens: a blue chip, worth two units of food (on the left) and a metal strip, worth five units of food (on the right). He correctely chooses the blue chip and thus receiving five units of food as reward, i.e. the amount of food corresponding to the non-chosen metal strip.
Are capuchins indeed so patient?
In the Delay choice task subjects are presented with a choice between a smaller immediate option and a larger delayed option. This task is frequently used to assess delay tolerance, interpreting a preference for the larger delayed option as willingness to wait. However, recent data show that this might not be always the case.
In the Delay choice task subjects face a dilemma between two preferred responses: “go for more” (i.e., selecting the larger, but delayed, option) vs. “go for sooner” (i.e., selecting the immediate, but smaller, option). When the options consist of visible food amounts, at least some of the choices of the larger delayed option might be due to a failure to inhibit a prepotent response towards the larger option rather than to a sustained delay tolerance. To disentangle this issue, we tested 10 capuchin monkeys, 101 preschool children, and 88 adult humans in a Delay choice task with food, low-symbolic tokens (objects that can be exchanged with food and have a one-to-one correspondence with food items), and high-symbolic tokens (objects that can be exchanged with food and have a one-to-many correspondence with food items). This allows evaluating how different methods of representing rewards modulate the relative contribution of the “go for more” and “go for sooner” responses. Consistently with the idea that choices for the delayed option are sometimes due to a failure at inhibiting the prepotent response for the larger quantity, capuchins and children chose the larger delayed option more with food than with high symbolic tokens, whereas low-symbolic tokens were ineffective in decreasing the salience of the larger option. Conversely, the sophisticated symbolic skills of adult humans prevented the distancing effect of high-symbolic stimuli in this population. Thus, at least in capuchins and children, opting for the larger delayed option in the visible-food version of the Delay choice task seems to partially result from an impulsive preference for quantity, rather than from a sustained delay tolerance. Our data extend the knowledge concerning the influence of symbols on both human and non-human primate behavior and invite greater caution in interpreting the results obtained with the visible-food version of the Delay choice task, which may overestimate delay tolerance. The above findings are supported by the lack of correlation between the performance in the Delay choice task and in other tasks measuring delay tolerance in non-human primates, children, and adults.
Capuchins set the right value on things
If capuchins were to choose between one piece of good food and two pieces of a low-quality one, it is not so sure they will take the larger quantity. As humans, in fact, monkeys understand the difference between quantity and quality, and their choices depend mainly on the relative value they assign to things.
Apart from humans, monkeys are able to make rational and economically advantageous choices based on the relative value of things. We studied economic choice behavior in capuchin monkeys by offering them to choose between two different foods available in variable amounts. When monkeys select between familiar foods, their choice patterns are well-described in terms the relative value of the two foods. In other words, they choose not only to take into consideration the quantity, but also the quality of the resources.
The same results emerged when capuchins had to select foods they have never tasted before: they learned to choose between them gradually, assigning each food a specific value. This result demonstrates that capuchins’ choices are not based on a stimulus-response association acquired through experience, as the leading view in economics and biology uphold. We propose instead a cognitive model in which the economic choice is the result from a two-stage mental process characterized by two different stages: of value-assignment and decision-making.
In the video Robin hood, an adult capuchin male, presented with a choice between two small raisins (on the left) and one large piece of banana (on the right), selects the two raisins.
Decision making and tool use
Wild bearded capuchins (Sapajus libidinosus) are small bodied primates. Adult females weight about 2 kg while adult males weigh about 3.5 kg (the dominant male being more than 4 kg). The most resistant palm nuts they break with tools are about 20 times harder than walnuts, whereas the resistance of other palm nuts is only about 10 times that of walnuts.Therefore, no doubt that the use of heavy stones is especially advantageous when the capuchin has to break hard nuts. However, if the heavy stone needs to be transported to the anvil then ... the monkey faces a high cost of transport.
What capuchins decide to do when faced with a heavy stone far from the anvil and a light stone close to the anvil? is theur decisional process affected by the resistance of the nut they need to crack open?
To assess the role of stone mass and transport distance in capuchins' tool selection, Massaro and colleagues (2012) carried out a series of field experiments. Overall, they demonstrated that all things being equal capuchins minimize transport distance choosing the closest stone to the anvil. However, when the nut is very resistant and there are a light stone is on the anvil and a heavy stone farther away, capuchins transport to the anvil and use only the latter. In contrast, when the nut is less resistant capuchins use both stones.
Overall, individuals vary in their sensitivity to distance of transport, a few meters are perceived as a substantive cost by some monkeys, and individuals' body mass affects decision making.
Researcher involved
Dr. Elsa Addessi, Dr. Elisabetta Visalberghi
Perceptual grouping in capuchins: when local wins over global
Human perceptual system processes the global features of objects to identify them, proceeding then to a more fine-grained analysis of local details. Capuchins have a visual system similar to that of humans, but they are faster at identifying the local elements of a visual scene and then its global configuration. This local superiority emerges also in other monkeys: possibly it evolved as a response to different ecological pressures experienced by the different species.
A fundamental aspect of visual perception, concerns the ability to process the global and the local aspects of the stimuli. In order to visually identify the objects and segregate them from the background, we must be able to group their component into a coherent perceptual whole. Many studies on visual grouping focused on the ability of the subjects to process hierarchical stimuli, i.e. small (local) shapes arranged to form a larger (global) shape. Human beings identify more quickly the global configuration than the local elements.
A problem that remains unresolved in visual cognition is why the well-established advantage shown by humans in the processing of global properties of visual patterns, is not mimicked by other primates species which nevertheless share a very similar visual system. In fact, our studies, similarly to other studies regarding non-human primates, demonstrate that capuchins discriminate the local components of the hierarchical stimuli more accurately than their global structure, in particular when the local elements are relatively distant from one another. According to some authors, these results could underline a greater difficulty in monkeys, than in humans, in grouping the local elements into a coherent whole rather than a local advantage per se. It is still unclear if such a cross-species difference, may reflect the existence of different evolutionary patterns on perceptual grouping capabilities in different species.
Researcher involved
Electronic versions of these papers are provided as a professional courtesy to ensure timely dissemination of academic work for individual, noncommercial purposes. Copyright and all rights therein resides with the respective copyright holders, as stated within each paper. These files may not be reposted without permission.
Sabbatini, G., Truppa, V., Hribar, A., Gambetta, B., Call, J. & Visalberghi, E. (2012). Understanding the functional properties of tools: Chimpanzees (Pan troglodytes) and capuchin monkeys (Cebus apella) attend to tool features differently. Animal Cognition, 15, 577–590
Visalberghi, E., Truppa, V. & Sabbatini, G. (2011). Analogical reasoning: What capuchin monkeys can tell us? In B. Kokinov, A. Karmiloff-Smith & N.J. Nersessian (Eds.) European Perspectives on Cognitive Science. New Bulgarian University Press
Truppa, V., Piano Mortari, E., Garofoli, D., Privitera, S. & Visalberghi, E. (2011). Same/different concept learning by capuchin monkeys in matching-to-sample tasks. PloS ONE, 6(8): e23809. doi:10.1371/journal.pone.0023809
Truppa, V., Garofoli, D., Castorina, G., Piano Mortari, E., Natale, F. & Visalberghi, E. (2010). Identity concept learning in matching-to-sample tasks by tufted capuchin monkeys (Cebus apella). Animal Cognition, 13, 835-848
Sabbatini, G. & Visalberghi, E. (2008). Inferences about the location of food in capuchin monkeys (Cebus apella) in two sensory modalities. Journal of Comparative Psychology, 122(2), 156-166
Spinozzi, G, Lubrano, G. & Truppa, V. (2004). Categorization of above and below spatial relations by tufted capuchin monkeys (Cebus apella). Journal of Comparative Psychology, 118(4), 403-412
Videos from Sabbatini & Visalberghi 2008 Journal of Comparative Pshycology
About nuts and abstract concepts
When it comes to opening nuts, capuchins know very well which features a stone must possess to be functional. Indeed, they use the experience acquired in nature with a variety of stones, to identify the functional features in tools they had never used before. And what about concept learning? Well, if properly trained, even capuchins are able to recognize abstract concepts of same/different between items, and use it to categorize new sets of stimuli.
The ability to understand similarities and analogies is a fundamental aspect of human cognition and humans are well known for their analogy-making capabilities. According to some authors non-human animals lack true analogical reasoning that involves the ability to figure out similarities between relations. However, it is possible that analogy-making did not emerge abruptly in our species, and that its precursors are present also in monkeys. To disentangle this issue, we explored the behavior of capuchin monkeys in tool using and matching-to-sample tasks requiring different levels of abstraction.
Tool use
Wild capuchin monkeys living in the dry forest habitat of Piauí, in Brazil, use hammers and anvils to open nuts. An experiment was designed to find out if capuchins’ stone selection was accidental, or based on the ability to identify the functional features of the tool. Monkeys were given artificial and natural stones differing in features such as friability, size and weight. In this experiment, capuchins always chose the functional tool on the basis of material and weight, even when weight could not be judged by sight. These results demonstrate that capuchins are able to use the experience acquired in nature with a variety of stones, to select among tools they have never encountered before, those with features that are functional features to open the nuts.
We investigated how capuchin monkeys solve tool problems in captivity. This kind of problem solving requires an evaluation of the tool length and of the tubes containing a reward. In particular, capuchins had to select the longest tool from different sets of sticks varying in length and in the shape or color of the handle. Our findings show that, when trained to concentrate on the functional feature (the length), some capuchins learnt to select the right stick, thus demonstrating the acquisition of a relational rule. The same experiment was carried out with chimpanzees, who solved the task applying a relational rule more quickly than capuchins.
Videos from Sabbatini et al 2012 Animal Cognition
Matching to sample
This study was aimed to the evaluation of the capacity of capuchin monkeys to acquire abstract concepts of same/difference and use them to solve matching-to-sample (MTS) tasks, involving relations of an increasing level of abstraction. Capuchins were first trained to recognize the relation (either same or different) between two sample items. Then, they were tested in selecting, between two pairs of new items, those with the same relation of the sample stimuli. Moreover, we examined the ability of capuchins to solve this kind of relational MTS task on the basis of the number of items composing the stimuli. We found that under specific training conditions (i.e. the number and the kind of stimuli used during the training), capuchins accurately match novel stimuli, showing in this way the first evidence of same/different relational matching-to-sample abilities in a New World monkey.
Videos from Truppa et al 2011 PlosOne
Researchers involved
Dr. Gloria Sabbatini, Dr. Valentina Truppa and Dr. Elisabetta Visalberghi
The comparative study of behavior and cognition in human and non-human primates enables to better understand developmental, cognitive and adaptive processes in different species. A comparative study of cognition and behavior, also allows further insight into major switching points in the phylogeny of primates. Our research group puts special emphasis on examining the processes by which primates acquire and process information coming from physical and social stimuli. In particular, we examine what kind of information is picked up, which factors facilitate the acquisition of new types of behavior, and the cognitive and social processes underlying the transmission of new behavior in non-human primates.
