Climate change is one of the most pressing challenges of our time, with impacts already being felt globally. The burning of fossil fuels for the past century has caused a remarkable increase in the planet’s average global temperature, destabilizing the Earth system and already leading to more frequent and severe weather events, including heatwaves, wildfires, hurricanes, and floods [1]. Alarming projections for the upcoming decades under current policies [2] range from decreased crop yields [3] and increased water insecurity [4] to billions of people potentially becoming displaced [5]. However, a narrow window of opportunity to maintain safe planetary boundaries for human civilization still exists, and it entails decarbonizing our society, transitioning to sustainable energy sources [1].

There are many ways in which individuals can contribute to this effort, from lifestyle choices to advocacy, and civic engagement [6]. However, despite increasing levels of belief in climate change and climate concern around the world [7], many are still not taking the climate actions necessary to trigger meaningful emissions reduction [8]. This collective failure to act has been attributed to several psychological barriers to action, at the cognitive, affective, motivational, social, political, or behavioral levels [9]. Accordingly, a rapidly growing body of research across the behavioral sciences has proposed interventions aimed at stimulating proclimate behaviors by overcoming these barriers 10, 11••.

Here, we propose a framework for conceptualizing this emergent body of work at three levels of analysis: the individual, the collective, and the system (Figure 1). Noting these levels are highly permeable, we review recent interventions at each level focusing on their contributions and interactions and offer a roadmap for future integrative climate action research and practice to better position behavioral scientists to contribute to the climate change response. Of note, we offer these categories as helpful conceptual guides rather than strict divisions. Our aim is to organize the diversity of interventions in this domain and illustrate how this structure can guide decisions about where and how to intervene to maximize positive impact, although we encourage future work to include more systematic meta-analyses of interventions across these levels to complement the approach taken here. We argue that such a multilevel framework has the potential to guide more integrated, impactful research.

At the individual level, climate action interventions tend to leverage cognitive, affective, or motivational processes to change conceptual representations of climate change (e.g. attitudes and beliefs) and their corresponding behavioral signatures (e.g. lifestyle changes).

Notable interventions at the individual level include decreasing the psychological distance of climate change by emphasizing proximal climate change effects [12], inducing negative emotions by highlighting damaging climate change consequences [11], or correcting pluralistic ignorance by changing norm perceptions [13]. Norm-based interventions using neighborhood or social comparisons were found particularly successful at reducing water [14], electricity [15], and meat consumption [16], as well as at increasing residential recycling [17].

In a recent megastudy aggregating this literature in a single large experiment, researchers tested 11 expert-crowdsourced theoretically driven interventions aimed at stimulating climate beliefs and policy support in a head-to-head comparison [11]. The results suggested that decreasing the spatial [11], temporal [18], or social [19] psychological distance of climate change are among the most effective psychological pathways for increasing belief in climate change and catalyzing public support for climate mitigation policies, such as implementing carbon taxes or transitioning to a clean energy model [11]. Taken together, these interventions provide policymakers and practitioners with a rich psychological toolbox to begin empirically addressing the growing climate crisis through behavioral interventions.

Future work at the individual level would benefit from incorporating longitudinal designs to capture the long-term effects of interventions 20•, 21••, 22, 23, employing mixed methods to provide a more comprehensive understanding of intervention impacts, and leveraging digital technology and big data to enhance the precision and scalability of impact measurement [24]. Moreover, recent theoretical developments have proposed prioritizing more impact-oriented research in environmental psychology, recommending the development of interventions focusing on pro-environmental behaviors with higher climate mitigation potential 21••, 25, 26. High-impact behaviors include increasing investments in renewable energy, increasing the use of public transportation and of electric vehicles, reducing air travel, reducing meat consumption, shifting to renewable electricity [27], as well as participating in collective action, such as advocacy and civic engagement 6, 28, 29, 30, 31. Future work would therefore benefit from developing behavioral interventions targeting such high-impact behaviors. Implementing such methodological suggestions stands to increase the impact of behavioral sciences on climate mitigation efforts.

At the collective level, behavioral interventions mainly target social and political barriers to climate mitigation, aiming to stimulate collective climate action such as advocacy, activism, or civic engagement.

While most climate action interventions to date have targeted private mitigation efforts based on individual decision-making 32••, 33, recent theorizing has suggested that focusing solely on the individual can miss essential processes that occur when individuals interact in their social networks, giving rise to emergent phenomena 34, 35. The dynamics of such macro level processes are critical to understanding how climate action can scale in collectives. Indeed, adopting a collective mindset has been suggested to mobilize collective action [36]. Accordingly, recent work has started to investigate collective solutions to climate change 37•, 38.

Recent notable interventions at the collective level include emphasizing social norms [39], increasing efficacy beliefs [8], or eliciting emotional responses [40]. For instance, a conjoint experiment found that participants expressed stronger support for messages that combined information about climate-related health risks, proposed climate solutions, and emphasized descriptive social norms [39]. However, while this design suggests that these message elements are associated with higher levels of advocacy support, it does not necessarily establish causality in this context.

Similarly, norms and efficacy beliefs were found to be strong determinants of collective action, such as participating in protests, contacting representatives, voting, donating, or volunteering 8, 41. Of note, while much of this work examines efficacy in general terms, efficacy can be further divided into personal (individual-level) efficacy (i.e. the belief in one’s own ability to make a difference) and collective efficacy (i.e. the belief that a group acting together can successfully achieve a desired goal) [41]. While strengthening personal efficacy is often associated with motivating private-sphere climate actions (e.g. reducing energy use, shifting to a plant-rich diet), enhancing collective efficacy is especially effective in galvanizing public-sphere actions (e.g. protests, advocacy campaigns, or civic engagement). Consequently, interventions that explicitly target both dimensions of efficacy may be more successful at driving meaningful climate mitigation by reinforcing individuals’ confidence in their personal actions and their belief in the power of collective efforts.

Moreover, dynamic anger elicitation was also found to increase support for climate policy and expectations for collective climate action [40]. And decreasing the social psychological distance of climate change by writing a letter to a future generation member was found to increase the support for climate mitigation policies across the political spectrum, for people identifying as liberals and conservatives alike [42], overcoming ubiquitous political barriers to climate action. This example of an individual-level cognitive intervention impacting a collective-level outcome showcases the permeability of the levels introduced here, pointing to their interconnectedness and complexity.

Moreover, political identity can serve as a powerful barrier or motivator of collective action; for instance, individuals may resist climate policies if they perceive them as clashing with their partisan or cultural group identity [43]. Addressing these identity-based barriers often requires tailored messaging that resonates with specific group norms and values [42].

In a recent megastudy aggregating this literature in a single large experiment, researchers tested 17 expert-crowdsourced behavioral interventions aimed at stimulating collective climate actions such as political or financial advocacy; they found that emphasizing the efficacy and emotional benefits of engaging in collective action to be the most effective strategies of catalyzing climate advocacy in the United States [10]. Other effective interventions targeting advocacy were the writing a letter to a future generation member explaining one’s current climate actions or appealing to moral values such as preserving the purity and sanctity of the environment [10].

However, focusing solely on motivating pro-environmental behaviors by targeting individuals and collectives has been described as “ahistorical and ignorant of how institutions shape the choices we can make” [30]. Instead, adopting a systems perspective was proposed to catalyze climate mitigation by considering the broader context in which behaviors occur. Empirical evidence for this argument shows that only 21% of the variance in people’s self-reported pro-environmental behaviors can be explained by their intentions to act sustainably [44]. Accordingly, a large portion of variability in behavior remains unexplained by intentions and likely stem from contextual, structural, or additional internal factors. This underlines the importance of investigating potential systemic interventions that address barriers beyond individual intentions. Indeed, current behavioral models, such as the Theory of Planned Behavior or the Value-Belief-Norm theory, have faced criticism for being overly individualistic and linear, overlooking structural factors, limiting their effectiveness at promoting meaningful climate action [45]. Future work would therefore benefit from integrating an interplay between individual behaviors, collective phenomena, and structural factors 46•, 47, 48, informing more effective and integrated solutions 49, 50•.

At the system level, interventions primarily target behavioral barriers to climate mitigation, aiming to facilitate and accelerate climate awareness and action through structural changes.

Notable pathways for systemic changes that stand to catalyze climate action include policy innovations, infrastructure development, as well as algorithmic, entertainment, and educational deployment of relevant information, among others. Intervening through these channels has the potential to facilitate and accelerate climate solutions directly, as well as indirectly (Figure 1). The indirect pathway, we posit, relies on increasing the effectiveness of interventions at the individual and collective levels, which can then further make the system even more conducive to action, in a positive feedback loop [51]. This mechanism could operate through social tipping points (i.e. moments at which incremental shifts in norms or behaviors reach a critical threshold that triggers rapid, self-reinforcing change) and cascading network effects, collectively driving transformational climate solutions at scale [52].

First, policy changes and incentives are essential pathways to removing financial barriers to engaging in climate mitigation behaviors. For example, the upfront costs associated with taking high-impact climate actions, such as purchasing an electric vehicle or installing solar panels, can be prohibitively high for many individuals, discouraging such actions despite their pro-environmental beliefs and attitudes. Thus, policies ensuring subsidies for these choices could make high-impact behaviors economically accessible to a broader population. For instance, in a meta-analysis of household-scale interventions, monetary incentives were found more effective at reducing energy consumption than nonmonetary interventions [53], although the impact of financial incentives compared to nudges is context dependent [54]. Similarly, economic incentives to install solar panels have been shown to narrow the adoption gap between Democrats and Republicans [55]. Additional strategic policies, such as subsidizing railroad infrastructure, implementing congestion taxes, or funding alternative meat research, can also trigger the adoption of low-carbon alternative transportation and food, with large carbon mitigation potential. Similarly, decision aids, design architecture, and smart defaults [56] could be integrated with individual- and collective-level interventions to optimize impact [57]. Unlike individual or collective interventions that rely on personal choices or social mobilization, policy interventions operate at a broader structural scale, altering economic or legal constraints that shape entire communities and markets. This distinction underscores the potential for policy measures to create large-scale shifts in climate-relevant behaviors.

Another promising structural approach to facilitate climate action involves digital technologies and prevalent algorithms. Digitization is a key factor in the transition to sustainable industrial processes [58]. For instance, using digital technology in agriculture is essential for mitigating the effects of climate change and addressing food insecurity [59]. A case study in Indonesia highlights the importance of integrating digital solutions into local waste management, a system-level intervention that involves not only individual waste-disposal habits but also infrastructure upgrades, governance structures, and economic incentives. By optimizing waste recycling, reducing greenhouse gas emissions, and driving economy-wide decarbonization [60], this example demonstrates how large-scale, multi-stakeholder efforts can facilitate significant climate benefits. Similarly, calls for a ‘Digital Green Deal’ [61] represent a high-level policy framework that coordinates digital innovation, regulatory measures, and economic restructuring across entire regions or nations. Although grounded in top-down approaches, such large-scale strategies also rely on bottom-up participation from local communities and businesses, illustrating the interdependence of systemic and collective efforts.

Moreover, recent work has identified problematic biases embedded in the algorithms behind prevalent digital technologies with potentially negative effects on climate awareness and action efforts 62, 63. In a recent global study, internet search algorithms were found to be influenced by pre-existing subjective climate sentiments but not by objective climate impacts in a given geographic region, distorting users’ climate cognitive concepts and corresponding behaviors [62]. Given that billions of people use these technologies [64], and most of them trust the information retrieved from digital search to be objective depictions of the world [65], the fact that search outputs may reinforce pre-existing climate sentiments rather than convey the objective threat posed by climate change can be a structural barrier to climate awareness and action around the world. Thus, interventions aimed at correcting such algorithmic biases and communicating objective climate risks in digital search outputs could have the potential to scale and streamline climate mitigation efforts.

In the domains of education and entertainment, increasing climate literacy through information conveyed in textbooks [66], movies [67], or games [68] also have the potential to catalyze widespread climate awareness and action [69]. For instance, gamification, which involves incorporating game-like features in nonentertainment contexts, has been found effective at addressing societal issues, such as misinformation thanks to its effectiveness and scalability 70, 71. A burgeoning body of research has started exploring the effects of gamifying climate education, incorporating social and affective engagement to trigger behavior change 72, 73. For instance, the ‘Cranky Uncle’ game inoculates players against climate misinformation [70], and the ‘Climate Duel’ game promotes actual behavioral changes such as sustainable consumption [74]. By reaching many users simultaneously and shaping public discourse around climate issues, such interventions function at a systemic level, affecting collective awareness and norms through accessible, scalable mediums.