Animal Behaviour — Cooperative Hunting
The Hunt That Requires a Mind-Reader
Some predators don't just coordinate their bodies — they model each other's intentions, and that changes everything we thought we knew about animal intelligence.
The Idea
Cooperative hunting is easy to misread. When wolves fan out around an elk or pelicans herd fish into shallow water, it can look like teamwork — but much of it can be explained by simple rules: follow the prey, mirror your neighbour, respond to movement. No theory of mind required. What's genuinely startling is when animals go further: when they don't just react to what their partner is doing, but anticipate what their partner is about to do, and adjust their own role accordingly. This distinction — reactive coordination versus role-based cooperation — turns out to be a meaningful cognitive line. Role-based hunting requires an animal to represent itself as one agent among several, each with a distinct function. That demands something like a mental model of another individual's future behaviour. Chimpanzees show flashes of this during hunts for red colobus monkeys in Taï Forest, where individuals reliably adopt positions — blocker, driver, ambusher — that only make sense in relation to what others are doing. Crucially, experienced hunters don't just grab the nearest role; they read the group's current formation and fill the gap. Orcas take this further still, coordinating attacks across family lines with what researchers describe as cultural transmission — hunt strategies passed down through generations, not encoded in genes. The cognitive scaffolding required to hunt cooperatively at this level begins to look less like instinct and more like something we'd hesitate to name in an animal.
In the World
In the 1990s, primatologist Christophe Boesch spent years following chimpanzees through Taï Forest in Côte d'Ivoire, trying to understand what was actually happening during their colobus monkey hunts. What he documented was uncomfortable for the field at the time. The chimps weren't just chasing. Boesch identified four distinct roles — driver (pushing the prey forward), blocker (cutting off escape routes), ambusher (waiting ahead of the prey's predicted path), and chaser (keeping pressure on). Individual chimps shifted between roles depending on what the group already looked like. An experienced male arriving late to a hunt would scan the formation and slot himself into whichever position was missing — not the most convenient one, but the strategically necessary one. More striking: when a chimp played ambusher — sprinting ahead to a position the prey hadn't yet reached — it was betting on where the prey would be driven, not where it currently was. That's a projection of future states based on what other hunters are doing. Boesch argued this constituted genuine collaboration, not just parallel action. His conclusions were contested — they still are, in corners of the field — but subsequent research, including studies on cooperative hunting in bottlenose dolphins off the coast of Brazil and coordinated orca attacks in Antarctic waters, has kept the question alive. The cognitive bar for "true" cooperation keeps getting tested, and certain animals keep clearing it.
Why It Matters
The standard story of intelligence tends to place it on a ladder: simple animals at the bottom, humans at the top, with most of the interesting cognition concentrated near the peak. Cooperative hunting quietly destabilises that picture. When a chimpanzee takes the ambusher role — a position that only makes sense if you can model another individual's intentions and project a future state — it is doing something that brushes against capabilities we usually reserve for ourselves. It's not language, but it's not nothing, either. This matters for how we think about the evolution of social intelligence. The pressure to read minds, plan jointly, and maintain roles across a moving, chaotic hunt may be precisely the kind of pressure that sharpens cognition over generations. In other words, the hunt might not just reveal intelligence — it might help explain how intelligence got here at all. For the reader, the invitation is a shift in default assumption: the next time you observe animal behaviour that looks mechanical or instinctive, it's worth pausing to ask what cognitive machinery would actually be required to produce it. The answer is sometimes more than we expect.
A Question to Ponder
If an animal can model another's intentions well enough to coordinate a hunt, what exactly is the remaining difference between that capacity and what we call 'understanding another person'?
Get a new one of these every morning.
Start learning with Thinkable