How Fear of Humans Can Ripple Through Food Webs and Reshape Landscapes

Predators like pumas cower in our presence. And these big cats aren’t the only ones

Even top predators like pumas flee in our presence. Worse, they might be disrupting entire food chains as they go. phil gould / Alamy

On a rainy night in the Santa Cruz Mountains, a mountain lion feasts on a deer carcass under cover of darkness. The lion dines alone, save for a chorus of tree frogs that start croaking just before he shears off another piece of meat with his powerful jaws. The big cat shakes the water from his head and looks around for a moment, as if searching for the source of the noise, but otherwise seems unfazed by the amphibian choir. Nearly an hour later, the lion is still working on the deer, but the frogs have gone silent. 

Suddenly, a man’s voice pierces the silence. In a flash, the lion is gone, leaving the remains of his kill. He does not return.

In reality, neither the frogs nor the man were real; both were audio recordings. The big cat, a six-year-old male named 66M, was part of a seven-month “playback” experiment on 17 mountain lions led by Justine Smith, as part of her doctoral research at the University of California at Santa Cruz. Just beyond the deer carcass was a motion-sensitive video cam-speaker system that Smith and her colleagues with the Santa Cruz Puma Project had set up whenever they found fresh kills. The team could usually tell when the mountain lions (also called pumas, cougars and scores of other regional names) had snagged a deer, because their GPS collars revealed that the roving animals had visited the same spot several times during the night.

Upon returning to its kill, a hungry puma triggered a recording of either a human pundit or the familiar, neutral calls of tree frogs, which don’t interact with pumas. Nearly all the cats responded like 66M, the team reported in Proceedings of the Royal Society B last month. Frogs didn’t bother them. But the mere sound of the human voice—in this case, Rush Limbaugh, speaking in an uncharacteristically calm tone—forced the animals to flee and abandon their hard-earned meal. The team concluded that the advent of the human “super predator” may be altering the ecological role of large carnivores—by disrupting the crucial link between a top predator and its prey.

Over the past 10 years, Puma Project research overseen by Chris Wilmers, an associate professor of environmental studies at UC Santa Cruz, has shown that human development affects where the cats move, feed, communicate with each other and stash their kittens. Last year, Smith showed that pumas spend less time feeding near neighborhoods, forcing them to kill more deer. And they’re not the only predator slinking away whenever humans are near: African lions, badgers and red foxes are all changing their behavior to avoid humans, with ecological consequences that scientists are just beginning to understand.

“We assumed from the beginning that mountain lions don’t like people,” Wilmers says. The evidence for that had been correlational, based largely on reading GPS data from radio-collared animals. This latest research, he says, “highly suggests” that the puma’s behavioral adaptations are driven by a particular mechanism: fear.

Smith, now a postdoctoral researcher at the University of California at Berkeley, had initially thought pumas living in such a developed landscape would be more habituated to people. “It was very dramatic to see that they fled almost every single time,” she says, “and often never returned at all.”

Fearsome predators afraid of us? Although we likely retain a primal fear of predators from the days when our forebears lived among giant ice age carnivores, today we overcompensate for that fear with a penchant for killing that is unknown in the wild. At a time when humans have become the dominant influence on the planet—leading many scientists to dub this epoch the Anthropocene, or Age of Humans—it’s perhaps not surprising that we distinguish ourselves as killers too.

We kill adult animals, the reproductive future of a species, at up to 14 times the rate seen in wild predators, Chris Darimont and his colleagues reported in a 2015 Science paper. We kill large carnivores at 9 times the rate they kill each other (mostly through intra-species battles). The wide-ranging ecological and evolutionary consequences of our extreme predatory behavior, the scientists argued, “uniquely define humans as a global ‘super predator.’” In the Anthropocene, Darimont told me, “humans have turned carnivores into prey.”

Only three people have died in mountain lion attacks in California since 1986, according to the California Department of Fish and Wildlife. Pumas, on the other hand, have a long history of dying at the hands of humans. Bounty hunters had largely eradicated the felids east of the Rockies by 1900, and hunted them for decades in California after they became confined to the West. Today, they’re typically killed by government officials after picking off someone’s pet or livestock. “The highest cause of mortality for pumas in our area is getting shot for eating goats,” Smith says. It’s no wonder the big cats bolt at the sound of a human voice.

“Understanding fear in the things that should be fearless is one of the coolest and newest [research] areas,” says Joel Brown, an evolutionary ecologist at the University of Illinois who was not involved in the puma research. Brown has long studied the larger ecological implications of being afraid, a phenomenon he calls “the ecology of fear.”

Scientists used to think mostly about predators’ ecological effects in terms of the direct impacts of killing, Brown says. “We now know that fear responses are often more important than the direct killing effect,” he says. The mere presence of a predator—signaled by a scent, sudden movement or an approaching shadow—triggers a range of responses in prey species as they try to avoid becoming food. “The mere risk of predation dictates where they forage, when they forage, how much they’re willing to forage and how vigilant [they are],” says Brown.


Theoretical models from the 1970s assumed that risk of predation influenced how animals foraged. This assumption was tested a decade later in pikas, small mountain-dwelling rodents that nest among boulders and also happen to be the inspiration for the Pokemon Pikachu. Nancy Huntly, now an ecologist at Utah State University, created experimental enclosures for the skittish herbivores by carrying boulders out to meadows, far from their dens. Pikas took advantage of these new refugia and promptly moved down the meadow.

In a now classic experiment from 1997, Oswald Schmitz, an ecologist at Yale University, showed that fear can ripple through trophic levels in the food web. Schmitz glued together the mouthparts of grasshopper-eating spiders, to see how grasshoppers would respond to predators that couldn’t kill them. The grasshoppers didn’t distinguish between the intact and incapacitated spiders, he found. They changed their feeding behavior when either spider was present, which in turn affected the biomass of the grasses they ate.

Fear can ripple not just through a food web but through future generations. In 2011, Liana Zanette, an expert on predator-induced fear who helped Smith design her puma study, showed that simply hearing the sounds of predators lowers breeding success in songbirds. Zanette used the same type of setup on songbirds in Vancouver’s Gulf Islands. Her team removed real predation risk by protecting nests with electric fences to zap hungry raccoons and fishing nets to thwart raptors. Then they manipulated the birds’ perception of risk by alternating recordings of raccoons, hawks and other predators—which typically eat half the songbirds’ offspring every year—with those of nonthreatening animals like hummingbirds and loons.  

“The fear effect was extremely costly for these animals,” says Zanette, who is at Western University in Ontario. Females ate less, and so laid fewer eggs. They spent most of their time looking for predators instead of foraging for their nestlings. As a result, these songbird parents produced 40 percent fewer offspring over the breeding season compared to animals that heard nonthreatening sounds.

Last year, Zanette’s team used this experimental setup in the same ecosystem to test the idea that fear of large carnivores can ripple through the food web. They focused on raccoons, opportunistic omnivores which their songbird experiments revealed were particularly fond of songbird eggs. It turns out they also love intertidal crabs and fish. With top predators long gone on the Gulf Islands, the fearless coons are free to chow down 24 hours a day, Zanette says.

So she and her student Justin Suraci tried to put the fear of predators back into the gluttonous bandits. They set up speakers and cameras along the shoreline, then played recordings of either dogs (which occasionally kill raccoons) or seals and sea lions (which don’t). “When raccoons heard the sounds of barking dogs, they fed 66 percent less than when they heard the sounds of barking seals,” Zanette says. “And there was a massive increase in the intertidal fishes and crabs, all the stuff raccoons loved to eat.”

If fear produces such dramatic effects through a mesopredator like a raccoon, what might it produce through a top predator like a puma? “We would expect these fear effects to be a common pattern across every single species in the animal world, because being killed by a predator immediately in an attack is such an extremely powerful evolutionary force,” says Zanette. Perhaps stating the obvious, she adds: “If you die instantly in a predator attack, your fitness falls to zero.”

If people are frightening a top predator to such an extent that it’s eating less of its cache, she says, that’s clearly going to affect the predator population. But altering the behavior of a large carnivore and how it moves through the landscape will also affect the fear responses of animals in the middle of the food chain and how much they can eat, she says: “And that’s going to cause a trophic cascade.”

On the positive side, the fact that a top predator fears us enough to avoid us when we’re out and about means they can coexist with us, says Smith. But it’s a balance. If they become too fearful to traipse through human landscapes, their habitat and hunting grounds will become even more fragmented, drastically reducing their chances of long-term survival.

Smith tries to understand what it’s like to live with people from the puma’s point of view. “Imagine a zombie apocalypse where there are these dangerous things that they can’t comprehend, and they have to hide and slink around like in a zombie movie to find food and navigate the landscape,” she says. “We have all these weird sounds and technology, and kill them all the time, but probably in ways they can’t predict or perceive. They’re kind of living in this postapocalyptic world, trying to escape us.”

Get the latest Science stories in your inbox.