Tuesday, March 5, 2013

Monkeys Reject Food From People Who Are Mean To Fellow Humans

When we observe an interaction between two other human beings (Person A and Person B), we sometimes draw conclusions about the personality traits or character of these two individuals. For example, if we see that Person A is being rude to Person B, we may be less likely to trust Person A, even though we are merely "third-party" evaluators. i.e. not directly involved in the interaction. Multiple studies with humans have already documented such third-party social evaluation, which can even occur among children. A study published in 2010 showed that 3-year old children were less likely to help adults who had previously acted in a harmful manner in front of the kids, i.e. torn up a picture drawn by another adult in a staged experiment.

Do animals who observe humans also conduct such third-party social evaluations of humans? The recent study "Third-party social evaluation of humans by monkeys" published in Nature Communications by James Anderson and colleagues staged interactions with human actors in front of tufted capuchin monkeys (Cebus apella). The researchers found that the monkeys indeed evaluate humans after witnessing third-party interactions involving either helpful interventions or a failure to help fellow humans.
In front of each monkey, two actors performed either "helper" sessions or "non-helper" sessions. In the "helper" sessions, Actor A tried to get a toy out of a container and requested help from Actor B, who complied and helped out Actor A. In the "non-helper" sessions, Actor B refused to help. After the sessions, both actors offered a piece of food to the monkey. In the helper sessions, monkeys readily accepted food from both actors. On the other hand, monkeys in the non-helper sessions accepted food more frequently from actor A (the requester of help) than Actor B (the non-helper).
The researchers also added an interesting twist to the experiment by creating a situation in which both actors had their own containers. The researchers then created an "occupied non-helper" condition in which Actor B did not even acknowledge Actor A's request because Actor B was pre-occupied by their own container. In this "occupied non-helper" situation, the monkeys accepted food from both actors equally. In an "explicit non-helper" condition, Actor B acknowledged the request for help from Actor A but explicitly rejected the request. In this latter situation, the monkeys were less likely to accept food from Actor B.
This study is not the first study to evaluate third party social evaluations of humans by non-human primates, but its strength lies in its meticulous design. Both actors offered the same type and amount of food to the monkeys, so that the most likely explanation for the monkeys' choices was indeed the interactions of the humans with each other.
The research presented in this study gives us a fascinating insight into how third party social evaluations by non-human primates. The fact that the monkeys discriminated between occupied non-helpers (i.e people who were too busy to notice the request for help) and explicit non-helpers (i.e. people who were just plain mean - they noticed the call for help but rejected it) shows a very fine-tuned analysis of human interactions. It is a good reminder of how human interactions can leave lasting impressions on fellow beings - humans and non-humans.

Image credit: Adult Tufted Capuchin by Charles J Sharp, via Wikimedia Commons (Creative Commons License).

ResearchBlogging.org Anderson, J., Kuroshima, H., Takimoto, A., & Fujita, K. (2013). Third-party social evaluation of humans by monkeys Nature Communications, 4 DOI: 10.1038/ncomms2495


  1. These monkeys are more forgiving than my kids.

  2. Yeah I don't. Forget being mistreated either

  3. Sure, three year old kids can figure it out.

    Monkeys can figure it out.

    But adults can't figure out that if dude screws with so-in-so and does him dirt, he's probably gonna do it to you, too.

    Pally or not.

    Thanks so much for this knowledge. Ahm interested.