Look at any kids’ playground – or adults’ corporate boardroom-- and the fact becomes obvious: Hierarchy and status-striving are part of primate living.
Not all primate species are arranged into hierarchies of dominance, but most are. High-status individuals may have great access to resources such as nutrient-rich food items and choice mates. It’s fascinating to consider how monkey and ape individuals come to understand their own relative rank among their groupmates. Televised nature documentaries sometimes give the idea that this understanding emerges from a literal battle: two gorilla males are shown locked in combat, or two monkey females vie for the same limited good.
Research on wild monkeys shows, though, that they are smart enough to learn about dominance by observation – that is, by watching the “social dance” of what other monkeys are doing around them. Japanese macaques (or snow monkeys) make careful choices of alliance partners, for instance, in ways that show how closely they assess others’ social connectedness. When Japanese monkey A is ready to aggress against monkey B, A chooses for an alliance another monkey that is both unrelated to B and higher in status than B.
This example is noted by scientists Paxton, Basile, Adachi, Wilson, Suzuki and Hampton in a new paper published in Journal of Comparative Psychology. Its title is “Rhesus monkeys rapidly learn to select dominant individuals in videos of artificial social interactions between unfamiliar conspecifics.” In it, Paxton et al.also review a captive study in which researchers had tried to understand just how monkeys come to be able to know what they know. In that case, rhesus macaques were shown videotapes of unfamiliar monkeys interacting in real time, in dominance-related contexts.
After first learning how to discriminate dominant individuals early on, the viewer monkeys were able to generalize their learning to novel videos. That is, in experimental trials they reliably picked out the dominant individual on screen.
Paxton et al. believe, however, that those monkeys might have made their choice based on physical markers that correlate with dominance, such as posture, health, and body size. They wondered whether a different type of experiment could avoid that confound, and tell once and for all whether observation of behavior alone is sufficient for dominance learning.
So, Paxton’s team created artificial video tapes and played them for rhesus monkeys. First the researchers filmed individual rhesus macaques each carrying out some activity relating to threats and high-status behaviors. Then, via editing, they paired up monkeys who had not interacted in real time. As the authors note, “This random assignment allowed us to control for any physical cues that might identify a dominant or subordinate individual. The artificial hierarchy was created by showing animal A as the dominant individual in all videos in which it appeared, animal B as the dominant individual when paired with all animals except A, and so on through C, D, and E.”
Next came a twist: The researchers also made what they call identity probe videos. In these, no dominance information was made available. Monkeys (the same ones already used) were filmed eating (or alternatively, jumping) then paired on film; the viewer monkeys could thus only pick out a dominant if they remembered its status from an earlier video.
On both tests the monkeys performed significantly above chance. This result in itself was pretty cool, but scientists being scientists, the team wanted to be sure the monkeys weren’t just memorizing which of the two animals was correct for each pair, instead of generalizing their understanding of dominance. (They worried about this because the monkeys had been reinforced for selecting the dominant animal in early trials).
The researchers decided to present monkeys with videos of two novel individuals interacting with the already-viewed monkeys. In this experiment, no reinforcement was offered. The results allowed the researchers to feel satisfied that the rhesus were not merely memorizing as they went along. However, in the identity-probe trials in this section of the work, the monkeys did less well. They did not make correct choices over multiple trials, suggesting that longer exposure to novel monkeys is needed. To me this finding isn’t a surprise. In real life, rhesus monkeys encounter each other again and again—they don’t glimpse a new individual for only a brief period.
While it is important to seek monkeys’ cognitive limits as well as their successes, I think these ingenious experiments show clearly that monkeys know about others’ relative positions in a hierarchy because they are so well-adapted to acquire social information by observation.
It’s no wonder we humans love to people-watch. Like other primates, we observe closely those around us-- and we draw vital conclusions from what we see.