Lessons from Behavioral Science for the Next Era of Climate Action
As a professor of environmental social sciences at the Stanford Doerr School of Sustainability who researches the psychology of climate solutions, my work immerses me in both the urgency of the climate crisis as well as the systemic societal changes necessary to make tangible progress against a problem of this magnitude. My time at Climate Week NYC last month reinforced just how closely these two are connected.
The week started with chaos, as hundreds of flights were canceled, including my own. It felt like a miracle when I finally made it onto a red-eye. From JFK, I went directly to my first event, a panel organized by the Environmental Voter Project in collaboration with the Climate Museum. That conversation with more than 100 climate leaders, and the many that followed over the next few days, left me thinking about what behavioral science can contribute to the next phase of the climate movement.
Polling and science both suggest that talking about climate change builds momentum
My first panel opened with a closer look at new polling from eight battleground states with competitive U.S. Senate races this November. One of the implications was particularly relevant to the phenomenon of “climate hushing” (the tendency of candidates to avoid mentioning climate). The data suggested that talking about climate change could help, rather than hurt, candidates. This aligns with research from my own lab and the broader literature in political science and communication, which shows that talking about climate change is critical to keeping the issue within the realm of political possibility and building momentum for moments when ambitious climate policy becomes possible. For example, social movements including the Sunrise Movement helped build momentum ahead of the passage of the Inflation Reduction Act. Silence does not create political opportunity. Talking about climate change can.
Evidence says: build efficacy and belonging
During that panel, I shared research from Stanford’s Climate Cognition Lab on the psychological pathways that can catalyze collective climate action and increase participation in the political process. One lesson from this research is that distrust in institutions and in one another can undermine collective action, cooperation, coordination, and civic engagement. Conversely, increasing people’s trust in one another and their belief in the efficacy of institutions is the most powerful lever for collective action. Giving people the sense that they have agency over their futures and the sense that they belong to the broader climate movement is key to motivation. People need to understand the severity of the problem while believing that something can be done about it.
Evidence says: prioritize high-impact climate actions
People can act on climate change in many ways, but these actions vary substantially in their potential impact. Research comparing household climate behaviors shows that actions such as avoiding beef, driving electric vehicles, and electrifying homes can produce much larger emissions reductions than commonly promoted behaviors such as recycling or turning off lights. People's actions can also extend beyond consumption behaviors to collective behaviors such as joining organizations, volunteering, and voting, which can contribute to political support for climate policy and broader systemic change. Accelerating the climate movement requires increasing participation in the behaviors with the greatest potential for impact, and behavioral science can help identify what motivates people to take these consequential steps and how to make them easier, more appealing, and more widely adopted.
One size doesn’t fit all: actions depend on motivations
Our research finds that different motivational pathways work for different actions. To increase climate policy support, thinking about future generations is especially powerful, as is helping people connect climate change to concrete consequences in their lives. To increase advocacy, efficacy beliefs paired with a sense of belonging to a movement are particularly important. For civic engagement, social norms can be powerful. And our latest research suggests that combining these elements can be even more impactful, and more efficient, than relying on any one strategy alone.
The media ecosystem matters
This brings me to a larger challenge. We are getting better at understanding what motivates climate action. But we operate within a media environment that does not necessarily transmit those messages.
The media helps people connect the dots between the climate events they experience and the larger forces causing them. Yet those connections are often missing. For example, a Media Matters analysis found that only 2–6% of broadcast-network segments about extreme weather events caused by climate change mentioned climate change. And connecting those dots matters. In a global study across 68 countries, we found that simply being exposed to most forms of extreme weather did not predict greater support for climate policy. What did predict policy support was whether people attributed extreme weather to climate change. In other words, experiencing the consequences of climate change does not necessarily translate into support for addressing it. People have to understand the connection.
This positions the information environment itself as a key player in addressing the climate problem. If we care about climate progress, democratic participation, and societal welfare, we also have to care about who controls the information environment, which voices it amplifies, and whether citizens have access to information that allows them to make those connections. More and more, advancing climate action by engaging participation requires a pluralistic and democratic media ecosystem.
Climate change and AI can no longer be treated as separate risks
My second Climate Week panel focused on AI and sustainability. During the event, I discussed a growing tension between rapid AI development and climate goals. AI has the potential to accelerate scientific discovery, economic productivity, and climate solutions, but realizing those benefits comes with growing energy demands at a moment when the world needs to rapidly decarbonize. In many places, that new demand is being met with methane gas generation rather than clean energy.
The risks posed by AI extend beyond its climate footprint. Increasingly, powerful AI systems can expand the capacity for mass surveillance, enable more sophisticated cyberattacks on critical infrastructure, lower barriers to developing biological and chemical weapons, and create or amplify risks to financial systems and political freedom. AI is advancing faster than the institutions designed to govern its consequences, making it more urgent to build the safeguards, democratic oversight, and collective capacity needed to ensure that its potential benefits do not come at unacceptable costs.
New evidence suggests that, without regulation, the rapid expansion of AI reinforces fossil fuel incumbency and increases emissions. My own research suggests that AI has the potential to help people develop more accurate climate beliefs, support climate policy, and take action. Which trends will prevail remains to be seen.
Many attendees approached me after the panel discussion to thank me for raising these issues in a conversation often dominated by AI acceleration and techno-optimism. It reminded me of a phenomenon we study in behavioral science: pluralistic ignorance, when people mistakenly believe their private concerns put them in the minority, even when those concerns are widely shared. This misperception can trigger a spiral of silence, as people remain quiet because they assume others do not share their views.
Later that night, I took the train to Princeton University, where I received my PhD years ago. In the morning, I gave the cognitive science seminar and shared two theoretical puzzles I have been working on for years: to what extent are our beliefs formed rationally versus motivated by our social identities, and to what extent do those beliefs actually influence our behavior? The next day, I was back on NJ Transit, headed to the final Climate Week event focused on climate mitigation and adaptation in the Global South. Before I knew it, I was back at JFK en route to California. I left New York with a clearer view into the next era of climate action, and an awareness that behavioral science has a role to play in igniting shared purpose among all who inhabit this planet.