Spider monkeys crowdsource best places to eat in forest

Spider monkeys don’t forage at random, but instead pool knowledge about where to find the best fruit in the forest, according to new research.

A study based on seven years of field observations in Mexico’s Yucatán Peninsula shows that spider monkeys, known for their long limbs and tails, have developed a kind of natural crowdsourcing system for finding food.

Geoffroy's spider monkey, considered an endangered species, lives in societies that contain between 20 and 42 members. Their diet consists primarily of ripe fruit and they require large tracts of forest to survive.

Scientists from the University of Edinburgh, Heriot-Watt University and the National Autonomous University of Mexico, wanted to use the field data to help explain a feature of spider monkey social behaviour: group members split into small subgroups, then rejoin in different combinations.

Spider monkey

The same subgroup might never forage together twice. 

The team discovered that the monkeys social mingling was not random, but a clever system for sharing insider knowledge about where the best fruit trees are located across their forest.

Complex mathematics

Traditional ecological models can only analyse pair relationships, how one monkey meets another, but spider monkeys often travel in subgroups of three or more and their ranges also overlap in sets of three or more. 

Ross Walker, a PhD student at Heriot-Watt University, supervised by Matthew Silk at the University of Edinburgh, developed a new method based on abstract mathematical theory rather than rainforest ecology. 

The team used data collected by experienced observers of a group of Geoffroy’s spider monkeys (Ateles geoffroyi) living around the Punta Laguna lake in the Otoch Ma’ax yetel Kooh protected area, in the Yucatan peninsula, Mexico. 

They had tracked individual monkeys’ movements and mapped out their core ranges, or the areas each monkey knows well. 

The team found that some parts of the forest are known by multiple monkeys, like a town’s most popular restaurant, while others are known by only one or two monkeys, like a hidden gem.

The findings revealed that there is enough overlap for monkeys to meet and exchange tips, but enough separation that each monkey scouts different parts of the forest. This maximises the whole group's collective coverage of the best feeding spots.

Optimal Middle Ground

The fact that these subgroups regularly meet and mix means that information can potentially be shared across the whole group. 

Importantly, the researchers found that these patterns aren’t explained by factors like age, sex or immigration status, which suggests they’re linked to how the monkeys gather and exchange information rather than who they are.

This demonstrates there’s an optimal middle ground between the monkeys sticking together and spreading out too far. 

It’s not helpful if every monkey knows exactly the same thing, and it’s not helpful if no one ever meets. It’s best when individuals explore different areas, but still reconnect often enough to pool what they’ve learned, researchers say.

Future Work

In future, the team would like to use the same mathematical techniques to explore other interactions between more than two individuals, which are very common and understudied.

For example, they could study many aspects of social dynamics in spider monkeys and other species that form temporary subgroups.

The study, published in the journal npj Complexity, was funded by the Engineering and Physical Sciences Research Council.

 

It isn’t random social mingling; it's a clever system for sharing insider knowledge about where the best fruit trees are located across their forest home. By constantly changing their subgroups, monkeys who know different parts of the forest can share information about where fruit is available.
 

Dr Matthew Silk Royal Society University Research Fellow, Institute of Ecology and Evolution

As you can imagine, if you’re trying to map how subgroups of monkeys move, when those subgroups change several times a day, any graph can start to look very messy. I developed a new approach based on simplicial complexes, which allowed me to capture how whole subgroups interact. This approach helped identify where monkeys’ knowledge about the forest overlaps, and where there are gaps. Those gaps matter, because they suggest that some subsets of individuals know things about parts of the environment that others don’t. 
 

Ross Walker PhD student, Heriot-Watt University

We have shown that the fluid social dynamics of spider monkeys has an important consequence for their foraging success: by exploring their environment in a distributed fashion and then coming together to share their uniquely obtained information, the group as a whole can know the forest better than a single individual could on its own. This foraging strategy is a good example of collective intelligence in natural conditions.
 

Professor Gabriel Ramos-Fernandez National Autonomous University of Mexico
Publication date