Predators create a "landscape of fear" for prey, and now we see that humans create a "landscape of fear" for top predators.
What is a landscape of fear? In short, it describes the sum changes to an ecosystem due to the presence of a predator. The impact of a single wolf on an ecosystem is far greater than how many deer it kills, because prey species change their behaviour in large ways in the presence of predators. In a landscape of fear, deer survival depends on being able to detect and avoid wolves, so the presence of predators could, for example, lead to more careful and dispersed foraging. Fear-induced changes in foraging preference -- which vegetation assemblages to forage more in-- and foraging efficiency-- how much the vegetation is exploited per time -- have been observed in many species, such as the Nubian Ibex. This is one reason why the reintroduction of top predators is a conservation priority; predators have lasting (and healthy!) effects on ecosystems, often promoting more sustainable foraging practices among prey species.
In this study, researchers wondered whether humans-- as a perceived top predator of pumas-- would invoke a similar "landscape of fear" response in these feline predators. Indeed, they found that female pumas increased their kill rates and spent less time feeding at each kill site. This study demonstrates the profound (and difficult to predict) effect of humans and housing developments on species behaviour.
The fear induced by predators on their prey is well known to cause behavioural adjustments by prey that can ripple through food webs. Little is known, however, about the analogous impacts of humans as perceived top predators on the foraging behaviour of carnivores. ... We observed strong behavioural responses by female pumas to human development, whereby their fidelity to kill sites and overall consumption time of prey declined with increasing housing density by 36 and 42%, respectively. Females responded to this decline in prey consumption time by increasing the number of deer they killed in high housing density areas by 36% ... our study emphasizes that knowledge of the indirect effects of human activity on animal behaviour is a necessary component in understanding anthropogenic impacts on community dynamics and food web function.