“People believe all sorts of things about how the world’s going to end,” said Matthew Billet, social psychologist at University of California, Irvine, in conversation with Science unscripted hosts Conor Dillon and Gabriel Borrud.
“Some people mean human extinction,” Billet said. “Some people mean the collapse of civilization or some sort of transformation of civilization as we know it, sometimes leading to a utopia, or a revitalization of humanity. And some people mean the complete destruction of Earth [by] a comet or a solar flare or something.”
Most of us have had thoughts about the end of the world. Billet and his colleagues set out to understand how those beliefs shape people’s attitudes and responses to global risks.
They found that attitudes toward global risks depend on four factors:
– How soon you think the world will end
– How you think it will end
– What personal role you believe you have in causing or preventing it
– What you expect will happen afterward
For example, if you believe the end is in God’s hands—a fulfillment of supernatural prophecy—you are less likely to support costly actions to prevent global risks like climate change. Similarly, if you feel there is no collective future, you’re less likely to back community-focused policies such as higher taxes to fund decarbonization.
Studying the psychology of end-of-world thinking
The researchers ran six pilot studies with 2,079 participants in the US and Canada, and a pre-registered study of 1,409 people. Pre-registration increases transparency and reduces biases like HARKing (hypothesizing after results are known).
Participants came from a range of religious backgrounds—Catholic, Mainline Protestant, Evangelical Protestant, Jewish, Muslim, and non-religious—with a mean age of 50. Genders were roughly balanced; most identified as white, while about a quarter identified as Black/African American, Hispanic/Latino, or Asian. Economic backgrounds varied.
They presented participants with five categories of global risk:
– Economic (supply chain collapse, debt crises)
– Environmental (natural disasters, failure to mitigate climate change)
– Geopolitical (nuclear war, collapse of nation states)
– Societal (global pandemic, erosion of social cohesion)
– Technological (artificial intelligence, disinformation)
Participants answered questions designed to measure how imminent the end of the world felt, whether humans or divine/cosmic forces would cause it, whether they felt personal control over events leading to the end, and their emotional valence about an apocalypse—whether they viewed it as good or bad.
Key findings
The majority see the end of the world as distant and abstract, but about one in three contemporary Americans consider an apocalypse personal and imminent. Those who think the world will end within their lifetime perceive global risks—climate change, pandemics, AI—as more severe, fear them more, and are more willing to take costly actions to prevent them, Billet said.
How end-of-world thinking helps cope with uncertainty
Beliefs about what comes after the end matter. Some people believe nothing good follows—if nuclear war occurs, everyone dies. Others expect a utopia or restoration of the Earth, especially for the righteous.
End-of-world thinking is both personal and collective. Because it is a shared process, these beliefs can help people cope with the fragility of groups, communities, and civilizations. “If you believe that you have a personal role to play in the apocalypse, like your actions matter, and you believe that there’ll be a utopia afterwards, then you can tolerate exposure to [global] threats that are distressing for [other] people,” Billet said. There is evidence these beliefs help people manage uncertainty.
Edited by: Zulfikar Abbany
Editor’s note: Matthew Billet (University of California, Irvine), Cindel J.M. White (York University, Canada), Azim Shariff and Ara Norenzayan (University of British Columbia, Canada) published their study “End of world beliefs are common, diverse, and predict how people perceive and respond to global risks” as a pre-print in January 2026 (not peer-reviewed at the time of the interview).