The experimental design was elegantly simple in concept but rich in the behavioral data it generated. Two groups of captive brown capuchin monkeys, one containing 22 individuals and the other 18, were given access to an apparatus featuring five distinct options. At the center stood the Stag box, a cooperative device that required two monkeys to press levers simultaneously to release high-value rewards. Surrounding it were four Hare boxes, individually operable devices that dispensed lower-value rewards to any single monkey willing to work alone. Because the animals could move freely among all five options and select any partner they wished across six testing sessions, the setup captured the genuine social dynamics of partner choice rather than forcing arbitrary pairings on unwilling participants.

The results were striking in their breadth. Most individuals in both groups engaged in successful cooperative bouts at the Stag box, demonstrating that the majority of capuchins in the study were both willing and able to coordinate their actions with a partner when the payoff justified the effort. At the same time, the monkeys did not neglect the safer alternative, regularly obtaining rewards on their own from the Hare boxes. This pattern mirrors the fundamental tension built into the Stag Hunt paradigm: cooperating with a partner promises a bigger prize but carries the risk of failure if the partner does not pull their weight, while working alone guarantees a modest but reliable return.

To disentangle the factors driving cooperative decisions, the researchers turned to generalized linear mixed models, a statistical framework well suited to behavioral data in which many observations are nested within individuals and dyads. The analysis revealed that dyadic tolerance, a measure of how comfortably two particular monkeys could share space and resources without conflict, significantly predicted the likelihood of cooperation in both groups. In other words, capuchins were not choosing partners at random. They were systematically more likely to attempt the demanding joint task with partners they tolerated well, suggesting that the quality of a social relationship acts as a gatekeeper for cooperative ventures in this species.

Individual characteristics mattered as well. The models showed that more exploratory monkeys, those that readily investigated novel objects and environments, participated in more cooperative interactions than their less adventurous groupmates. Dyads involving juveniles also engaged in more cooperative bouts, hinting that younger animals may be especially motivated to experiment with the apparatus or that their playful dispositions make them attractive or available partners. These findings underscore that cooperation is not solely a property of relationships; the personalities and life stages of the individuals involved shape how often cooperative opportunities are taken up in the first place.

Perhaps the most technically intriguing results came from a finer-grained analysis of pressing behavior, which examined how the mere presence of a partner influenced how hard monkeys worked at each type of box. Here the two groups diverged in unexpected ways. In the East Group, the presence of a partner increased pressing at the Stag box while simultaneously lowering pressing at the Hare boxes, a pattern consistent with genuine social facilitation of cooperation and a devaluation of the solo option when a collaborator was available. In the West Group, by contrast, Stag-box pressing increased steadily across sessions regardless of whether a partner was present or absent, suggesting that learning and familiarity with the apparatus, rather than partner presence, drove much of the cooperative effort in that group.

This group-level divergence carries important methodological implications for the field of comparative cognition. It demonstrates that ostensibly identical experimental conditions can produce different behavioral signatures in different social groups, and that conclusions about partner sensitivity drawn from a single group may not generalize. The authors argue that this variability is precisely why paradigms allowing free partner choice are so valuable: they reveal the social dynamics underlying cooperation rather than smoothing them away under the constraints of forced pairing. When animals can vote with their feet, researchers gain access to the decision processes that matter in the real world where cooperation evolved.

The study also speaks to a long-standing debate about the evolutionary foundations of mutualism. Mutualistic cooperation with free partner choice is widely considered a key pillar in the evolution of cooperative behavior, including in humans, because it allows individuals to shop for reliable partners and abandon unreliable ones, creating selection pressure for trustworthiness and coordination. Yet many experimental studies of animal cooperation have restricted partner choice or offered few alternatives, limiting insight into the factors that actually shape cooperative decisions. By combining a cooperative option with multiple solo alternatives in an open social setting, the Stag Hunt apparatus used here brings experimental designs closer to the ecological realities in which cooperative psychology took shape.

The ethical design of the research deserves note as well. The study was entirely non-invasive and relied exclusively on voluntary participation: no animal was food-deprived, socially separated, restrained, or physically handled for the purposes of the experiment. Multiple apparatuses were installed simultaneously to reduce monopolization by dominant individuals, and zoo staff implemented scattered feeding during testing to maintain naturalistic foraging dispersion and minimize competition. All rewards formed part of the animals’ regular diet, and no violent or injurious interactions were observed in connection with the procedures. The work was approved by the School of Psychology and Neuroscience Ethics Committee at the University of St Andrews and conducted in close collaboration with zoo authorities at Living Links, Edinburgh Zoo.

Taken together, the findings highlight the importance of integrating social relationships and individual traits into any account of cooperative decision-making in primates. Tolerance between partners determines whether cooperation is attempted at all; exploratory tendencies and youth shape how often individuals engage; and the influence of a partner’s presence on effort can vary from group to group. For capuchins, at least, the road to the big prize runs through social tolerance, and the willingness to take that road depends on who is standing beside you. As researchers continue to probe the roots of cooperation across species, studies like this one demonstrate that the richest answers emerge not from forcing animals into prescribed partnerships, but from letting them choose for themselves.

Subject of Research: Cooperative behavior and partner choice in brown capuchin monkeys tested with a Stag Hunt cooperation task

Article Title: Cooperation in brown capuchin monkeys: social tolerance, individual traits, and partner sensitivity in a Stag Hunt task

Article References: Baruah, K. M., Call, J., & Seed, A. M. (2026). Cooperation in brown capuchin monkeys: social tolerance, individual traits, and partner sensitivity in a Stag Hunt task. Animal Cognition. https://doi.org/10.1007/s10071-026-02107-8

Image Credits: AI Generated

DOI: 10.1007/s10071-026-02107-8

Keywords: capuchin monkeys, cooperation, Stag Hunt, partner choice, social tolerance, animal cognition, primates, mutualism, exploration, behavioral ecology, comparative psychology, social behavior

The observations began in early 2022 when Zoë Goldsborough, a doctoral researcher at the Max Planck Institute of Animal Behavior (MPI-AB), noticed an astonishing sight while sifting through thousands of images and hours of footage: a young male capuchin carrying a baby howler monkey on his back. This behavior had never been recorded on Jicarón despite continuous monitoring since 2017, marking a startling deviation from known primate interaction patterns. The initial surprise quickly gave way to curiosity, setting the stage for a detailed investigation into this unusual form of cross-species infant carrying.

The primary carrier was identified as a subadult male capuchin, affectionately named "Joker" by the research team. Over time, Joker was seen carrying not just one but multiple howler infants, sparking questions regarding the origin, motivation, and implications of this behavior. The phenomenon was initially hypothesized to be a form of interspecies adoption, a behavior very rarely documented, typically involving females caring for younger infants as part of a nurturing role. However, the fact that it was exclusively young males engaging in this carrying activity challenged this assumption, hinting at a more complex social dynamic at play.

Following the early spike of observations, the behavior ceased temporarily, leading researchers to speculate whether Joker’s actions were a singular experimental behavior typical of juvenile capuchins’ exploratory nature. Capuchins are renowned for their cognitive curiosity and problem-solving abilities, traits that often lead to novel behavioral innovations within their social groups. The gap in observations also prompted questions regarding the stability and potential spread of this behavior throughout the population.

A breakthrough came when, approximately five months later, new recordings revealed that several other young male capuchins had begun adopting this behavior, carrying howler infants with increasing frequency. This rapidly expanding pattern suggested a form of social transmission, a hallmark of cultural behaviors among animals that are learned not genetically but through observation and imitation. The researchers dubbed this unfolding sequence “a social tradition or cultural fad,” emphasizing the parallels between animal and human culture in the propagation of behaviors that may not necessarily confer obvious benefits to the practitioners.

The interspecies infant carrying typically lasted for periods up to nine days, as documented in extended video sequences. During these times, the capuchins engaged in their daily routines—including extensive tool use to crack open food sources such as nuts and shellfish—while the howler infants clung to their backs or bellies. The astonishing tolerance of prolonged physical contact between species, combined with the lack of aggressive interaction, provides valuable insight into primate social flexibility and raises questions about the cognitive underpinnings of such behaviors.

Yet, this social tradition carries a darker undertone. The howler infants, all less than four weeks old, were effectively abducted from their parents, who were observed vocalizing distress calls from nearby trees. Tragically, at least four of these infants were documented to have died, likely due to deprivation of maternal care and essential nutrition such as milk, which the capuchins cannot provide. The phenomenon thus highlights a rare case where social learning produces behaviors with destructive consequences for a vulnerable species, a situation that opens new discourse on how culture in non-human animals can generate maladaptive outcomes.

This leads to a broader and more philosophical inquiry regarding the drivers of such behaviors. Unlike human cultural evolution, where many traditions evolve to serve some functional benefit, these capuchin behaviors appear to have no clear advantage. The males do not gain increased social standing, mating opportunities, or enhanced access to resources through infant carrying. Nor do they engage in play or other interactions that could suggest a practical purpose. This enigmatic behavioral pattern presents a compelling example of social learning decoupled from immediate fitness benefits.

In addressing why this behavior emerged at all, researchers point toward environmental and ecological factors unique to Jicarón Island. The absence of predators and a scarcity of competitors may have created an environment of low external pressures, effectively producing a form of ecological luxury. Under such conditions, boredom and excess free time among intelligent and inquisitive capuchins could act as the impetus for innovation and experimentation. Such an interpretation challenges the classical notion that necessity is the mother of invention, offering instead that cognitive play and social experimentation are important drivers of cultural evolution in animals.

Moreover, the fact that only male capuchins both carry howler infants and participate in the island’s established tool use behaviors suggests a potential connection between these social traditions. Both could stem from the same underlying cognitive and social state—perhaps a form of restlessness or boredom in a relatively stress-free environment, leading to the birth of novel and unexpected behaviors that ripple through the social fabric of the group.

The social dynamics observed underscore the intricate ways in which culture can evolve in animal societies, revealing a capacity for innovation, transmission, and even the adoption of behaviors that might, paradoxically, harm other species. This discovery challenges rigid frameworks that equate animal culture strictly with adaptive benefits and underscores the richness and variety of social learning mechanisms across taxa.

Looking forward, ongoing analysis of the extensive camera trap data, accumulated between January 2022 and July 2023, will shed light on the persistence and potential ecological ramifications of this behavior. If the infant-carrying tradition continues to spread to other capuchin groups or increasingly impacts the endangered howler monkey population on Jicarón, the phenomenon could evolve into a critical conservation concern, necessitating management interventions within Coiba National Park.

Witnessing such a bizarre and unprecedented social phenomenon has profoundly impacted the researchers involved. As Meg Crofoot, managing director at MPI-AB, eloquently puts it, this natural population of capuchins provides a unique live model for understanding animal culture’s surprising and sometimes unsettling trajectories. The team feels a deepened responsibility to continue monitoring, studying, and learning from these intelligent monkeys who have, unintentionally or not, crafted a social tradition bridging species boundaries in a way unseen before.

The findings were recently published in the journal Current Biology, marking a seminal contribution to the study of animal behavior, culture, and interspecies relationships. As we unravel more about the cognitive and social complexities of capuchins on Jicarón, this story reminds us that the natural world still harbors enigmas as strange and fascinating as those in human societies, warranting our ongoing curiosity and respect.

Subject of Research: Animals
Article Title: Rise and spread of a social tradition of interspecies abduction
News Publication Date: 19-May-2025
Web References: Max Planck Institute Interactive Website
References: Current Biology, DOI: 10.1016/j.cub.2025.03.056
Image Credits: Brendan Barrett / Max Planck Institute of Animal Behavior
Keywords: Capuchin monkey, howler monkey, social tradition, animal culture, interspecies behavior, primate tool use, social learning, behavioral ecology, infant carrying, cultural transmission