Abstract
Realizing a sustainable future requires the active participation of today’s young generations. How can we foster adolescents’ proenvironmental engagement? In a preregistered, cross-national, longitudinal field experiment, we tested a novel approach to promoting adolescent behavior change: motive alignment. We hypothesized that presenting proenvironmental behavior as aligned with adolescents’ developmentally salient motives (in this case, autonomy and peer status) would enhance their engagement. In Study 1 (the Netherlands), a motive-alignment (versus control) intervention implemented in secondary schools increased adolescents’ proenvironmental food choices on the day of the intervention. Although this effect faded after 4 weeks, participants in the intervention condition continued to donate more to an environmental organization at follow-up. In Study 2 (China), we replicated these findings. Motive alignment thus offers a promising approach to strengthen interventions promoting adolescents’ proenvironmental behavior; it presents scalable and affordable opportunities to foster youth behavior change.
Keywords
Introduction
Most adolescents are concerned about climate change and motivated to contribute to its mitigation (Hickman et al., 2021; United Nations Development Programme, or UNDP, 2021). Still, efforts to encourage adolescents’ day-to-day proenvironmental behavior (i.e., behavior that benefits the environment, or harms the environment as little as possible; Steg & Vlek, 2009) exhibit only moderate success (Świa˛tkowski et al., 2024; van de Wetering et al., 2022). Theory and preliminary research suggest that motive-alignment (or values-alignment) interventions—that is, those that frame behavior as compatible with individuals’ current, core motives—can enhance efforts to promote behavior change (Bryan, 2020; Thomaes et al., 2023). However, this research has been conducted exclusively with Western samples and, for proenvironmental behavior, in controlled research settings (e.g., van de Wetering et al., 2025a). Thus, motive alignment’s real-world applicability to promote proenvironmental behavior and its replicability in other cultural contexts remain unknown.
In a preregistered longitudinal field experiment conducted in secondary schools in the Netherlands (Study 1) and China (Study 2), we evaluated the impact of a motive-alignment intervention on adolescents’ proenvironmental behavior. We hypothesized that the motive-alignment intervention (compared with a traditional educational intervention) would promote more proenvironmental behavior both in the short term (i.e., on the day of the intervention) and in the longer term (at a 4-week follow-up).
Sustainability motive alignment
Motive-alignment interventions rest on the premise that motives (i.e., values and goals) drive behavior (Berkman et al., 2017; Dweck, 2017) and that individuals will be more likely to adopt new behaviors if they see how those behaviors help them realize their preexisting motives (Bryan et al., 2019). Consequently, motive-alignment interventions help address the challenge of motivating behavior change in contexts—such as environmental contexts—in which the expected benefits of the targeted behaviors may not naturally align with people’s personally salient goals. Rather than aiming to change people’s priorities (e.g., persuading them to prioritize long-term interests and concerns), motive-alignment approaches reframe targeted behaviors as a means to achieve immediately relevant priorities (e.g., autonomy and peer status). These approaches thus imbue targeted behaviors with benefits that carry current motivational relevance, unlike traditional pragmatic appeals for behavior change, which tend to emphasize long-term payoffs (Bryan, 2020). Research has demonstrated the motivational power of motive alignment across different populations (e.g., adolescents and adults, by linking targeted behaviors to motives salient for each group; Hecht et al., 2023; Yeager et al., 2018) and across diverse behaviors (e.g., promoting healthy eating, reducing social-media use; Bryan et al., 2016, 2019; Galla et al., 2021).
Applied to the environmental domain, the sustainability motive-alignment hypothesis (Thomaes et al., 2023) posits that adolescents are more likely to adopt proenvironmental behavior if they construe such behavior as an opportunity to fulfill their developmentally salient, short-term motives for autonomy (Fuligni, 2019; Rodríguez-Meirinhos et al., 2020) and peer status (Crone & Dahl, 2012; Tomova et al., 2021). Although these motives remain important throughout the life span (Ryan & Deci, 2000), physiological, psychological, and environmental changes during adolescence make them especially salient during this period (Blakemore & Mills, 2014; LaFontana & Cillessen, 2010). In their quest for self-definition, adolescents are driven to express themselves and find their place in society (Fuligni, 2019). They are also acutely sensitive to their peers’ opinions (Dahl et al., 2018; Foulkes & Blakemore, 2016) and readily adapt their behaviors or lifestyles to gain peer approval (Paluck et al., 2016; Veenstra & Laninga-Wijnen, 2022). Rather than emphasizing the long-term importance of lifestyle changes to prevent future climate catastrophes—as is typical in environmental education (e.g., Greenpeace UK, 2020; National Oceanic and Atmospheric Administration, n.d.)—motive-alignment approaches frame proenvironmental behavior as relevant to adolescents’ pertinent motives to express autonomy and fit in with peers.
Previous work, conducted in controlled research settings with Western participants, suggests that motive-alignment approaches can—at least for certain behaviors—promote adolescents’ proenvironmental engagement (van de Wetering, 2025a; van de Wetering 2025b). For example, one experiment found that presenting proenvironmental behavior as an opportunity to stand up to companies contributing to deforestation increased adolescents’ proenvironmental behavior intentions and their signings of an environmental petition (van de Wetering et al., 2025a).
Cross-cultural replicability
Theoretically, motive-alignment approaches should be robust across cultural contexts as long as the interventions speak to adolescent motives that are salient across cultures. We conducted this research in China and the Netherlands, two nations that differ culturally in multiple ways (e.g., power, distance, individualism, indulgence; The Culture Factor Group, n.d.; Hofstede, 2011). Still, adolescents worldwide share strong motives for autonomy and peer status (Anderson et al., 2015; Nalipay et al., 2020). We thus evaluate the replicability of motive-alignment approaches by conducting Study 1 in the Netherlands and Study 2 in China. Examining proenvironmental behavior interventions in China is also practically and societally important, as China is the world’s largest emitter of greenhouse gases and home to the second-largest youth population (Ritchie & Roser, 2020; United Nations International Children’s Emergency Fund, n.d.).
The present research
This preregistered, cross-national, longitudinal field experiment tested the effects of a 15-min motive-alignment intervention to promote adolescents’ proenvironmental behavior in the natural context of their school. We randomly assigned participants to a motive-alignment or traditional (active-control) educational intervention. We assessed adolescents’ proenvironmental behavior in two ways. On the day of the intervention and at a 4-week follow-up, we observed adolescents’ proenvironmental food choices (i.e., their selection of a snack with an “Eco-label A” or an “Eco-label D”). At follow-up, we also measured participants’ donations to a proenvironmental organization. Using Bayesian informative hypothesis testing, we evaluated the hypothesis that the motive-alignment intervention would foster more proenvironmental behavior than the control intervention.
Research Transparency Statement
General disclosures
Studies 1 and 2
Study 1
Does a motive-aligned intervention, compared with a traditional environmental-education intervention, increase proenvironmental behavior among adolescents in the Netherlands?
Method and participants
Study 1 included 473 adolescents (Mage = 15.9 years, SDage = 0.9 years; 52.6% female, 33.6% male, 1.9% other, and 11.8% unreported; 63.6% Dutch, 16.9% Dutch and other ethnicity, 7.6% other ethnicity, 11.8% unreported). We included in this sample only youth who were present on the day of the experimental intervention (i.e., participated in either the control or motive-alignment intervention), as preregistered. We recruited participants from four secondary schools located in suburban regions throughout the Netherlands, which serve mostly middle-class communities. Following our preregistered power analyses and sampling plan, we stopped recruiting additional schools after reaching a minimum of 300 participants. At the end of the study, participants received €8 as compensation for their time and effort, along with a debriefing letter explaining our research aims, use of observations, and study materials. The Institutional Review Board at Utrecht University in the Netherlands approved of this study (Protocol No. 23-0103).
Experimental procedure and measures
Intervention development
We based the control intervention on real-world, typical environmental-education materials (Greenpeace UK, 2020; World Wildlife Fund, n.d.). We initially developed two motive-alignment intervention approaches, on the basis of prior work (Bryan et al., 2019; Thomaes et al., 2023). One approach highlighted youth as increasingly taking action to lead societal progress on climate change; another approach highlighted youth as rebelling against companies that contribute heavily to climate change. To receive feedback and optimize the intervention approaches, we conducted interviews with Dutch (n = 4) and Chinese (n = 6) adolescents and secondary school teachers (N = 2). Next, we conducted a pilot in separate samples of Dutch (N = 148) and Chinese (N = 318) adolescents to examine how they perceived both motive-alignment intervention approaches. Although both approaches promoted more self-reported motive alignment (e.g., I find it admirable when people do good things for the environment) relative to the control intervention, the approach emphasizing youth as leading societal progress showed more consistent effectiveness across cultures (see the Supplemental Material available online for detailed methods and results). Thus, we adopted this approach in the present research.
Preintervention
Approximately 2 weeks before the intervention, participants completed an online survey in class. Among other variables not analyzed in the present study (see the Supplemental Material for all variables), the survey included measures of climate-change skepticism and brand sensitivity. These measures allowed us to evaluate randomization between conditions and explore possible country differences and moderation effects. We used three items to measure climate-change skepticism (e.g., I think that global warming is really happening vs. I doubt that global warming is really happening; Grapsas et al., 2023) and five items to measure brand sensitivity (e.g., When I buy something, I prefer a well-known brand; Perrin-Martinenq, 2004; see the Supplemental Material for descriptives).
Intervention
Participants were randomly assigned at the individual level (i.e., not at the school or classroom levels) to either a motive-alignment intervention or a traditional (active control) intervention. The intervention took approximately 15 min to complete and was administered in class by our research team. Students received a printed handout with a uniform cover page, including reading materials and writing activities about climate change. Students completed the intervention independently, with teachers and research assistants (both blind to condition) present to answer questions.
The reading materials differed between conditions, but the writing activities were the same (see the Supplemental Material for the full interventions in both conditions in English, Dutch, and Chinese). To harness adolescents’ motives for peer status and autonomy, we ensured that the motive-alignment reading materials conveyed that climate change is viewed as highly important among peers. For example, we quoted a real statistic from the People’s Climate Vote, the largest public-opinion survey on climate change (UNDP, 2021), saying, “Almost 70% of youth see climate change as a worldwide emergency.” These materials also portrayed youth as “taking matters into their own hands” rather than waiting for governments and companies to act, and “increasingly making pro-environmental choices in their everyday lives.” To further highlight that peers care about proenvironmental behavior and see it as a way to make a positive societal impact (i.e., alignment with status and autonomy), we included in the materials a quote “from a 17-year-old-student” that we developed with youth in the pilot interviews: “Even by doing small things I can already do my best for the climate, by living in an environmentally friendly way. My generation is going to make a difference, and I am proud of that.” The motive-alignment reading materials also featured photos of youth engaging in environmental activism, along with a (fictitious) infographic depicting increasing rates of youth proenvironmental engagement over the past decades. This infographic was designed to be indistinguishable from the real infographic used in the control condition depicting increasing rates of global CO2 emissions over the past centuries (Ritchie & Roser, 2020). Although the intervention infographic is fictitious, it reflects a real trend of growing youth engagement in proenvironmental behavior (The Commission to the European Parliament, 2025; Penker, 2024). These materials thus framed proenvironmental behavior as trending among peers (appealing to the peer-status motive), and as allowing independently minded youth to lead societal progress (appealing to the autonomy motive).
The control condition also aimed to motivate adolescents to engage in proenvironmental behavior, but it used a different (non-motive-aligned) approach to do so. The control condition emphasized the personally and societally significant (albeit temporally distant) consequences of climate change (e.g., “putting cities such as [Amsterdam/Shanghai] in danger of disappearing under water”). In addition, it included explicit (albeit prescriptive) appeals from an authority figure to engage in proenvironmental behavior; for example, there was a quote from a “science teacher” stating that “Young people can and must make a difference. They should work together to demand and participate in one of the largest and most urgent transitions in human history.” Although the control condition contained motivational content, we did not expect it to be very effective, given that such traditional appeals often fail to align with adolescents’ short-term orientation and heightened need for autonomy.
Both conditions contained a uniform section providing information on what youth can do to protect the environment. The section presented examples of proenvironmental behaviors that young people can engage in, such as using a refillable water bottle and choosing products with a green Eco-label. This section was followed by writing activities that were also the same in both conditions. Participants wrote a very brief action plan describing ways they could help protect the environment. They then briefly responded to questions about the materials (e.g., “What information did you find the most interesting and why?”), which were intended to strengthen their engagement with the intervention contents and reinforce the cover story of this research (i.e., that students were providing feedback on environmental-educational materials).
Proenvironmental food choice
After completing the intervention, participants received a coupon to redeem a free snack (i.e., potato chips) from a food stand that was set up at their school (a behavioral task adapted from Bryan et al., 2016). To minimize suspicion and demand effects, we informed students at the start that they would have the opportunity to collect free snacks twice, first on the day of the intervention and again at follow-up, as part of their compensation for providing feedback on the educational materials. Moreover, the food stands were located in common areas, away from classrooms, to separate them from the study activities. Participants wrote their names both on the coupons and the returned intervention handouts, which allowed us to test whether participants’ snack selection differed between conditions. At the food stand, participants could choose among four types of chips: (a) classic flavor, unfamiliar brand (Quinn), Eco-label A; (b) classic flavor, familiar brand (Lays), Eco-label D; (c) paprika flavor, unfamiliar brand (Quinn), Eco-label A; and (d) paprika flavor, familiar brand (Lays), Eco-label D (see Fig. 1; classic and paprika are locally popular chip flavors, as confirmed during the pilot interviews). We coded proenvironmental behavior (i.e., choosing Eco-label A chips) dichotomously (Eco-label A = 1, Eco-label D = 0). We used Eco-labels to indicate the environmental impact of equivalent food options, as they present a systematic approach to informing consumers and encouraging proenvironmental behavior (Taufique et al., 2022; Prieto-Sandoval et al., 2016; Yokessa & Marette, 2019).
Snack options presented at the food stands in the Netherlands. Paprika (paprika) and naturel (plain) are locally popular chip flavors.
Follow-up
Approximately 4 weeks following the intervention, participants completed an online survey in class at school. Along with payment information (needed to process the financial compensation), the survey included a proenvironmental donation measure—an unobtrusive element in the administrative part of the study. On the same day, we again measured participants’ proenvironmental food choices at the food stands.
Proenvironmental donation
At the start of the study, we offered participants €5 for taking part in our research. In the follow-up survey, we offered participants an additional €3 and asked them whether they would like to keep the additional compensation (i.e., donate €0) or donate €1, €2, or €3 to a youth-led environmental organization. Donation tasks like these are often used to assess proenvironmental behavior, because donating to an environmental organization involves a consequential trade-off between environmental and personal costs and benefits, similar to those people encounter when making real-world proenvironmental decisions (Lange, 2022). We coded proenvironmental donations continuously, from 0 to 3.
Proenvironmental food choice
Upon completing the follow-up survey, participants received a coupon to use that day at a food stand that we set up at their school, allowing us to observe their proenvironmental food choice using the same procedure as on the day of the intervention.
Results
As preregistered, we used Bayesian informative hypothesis testing to evaluate evidence for our hypothesis that participants in the motive-alignment (vs. control) condition would be more likely to engage in proenvironmental behavior. We conducted separate analyses for each measure of proenvironmental behavior, including all participants with available data for that measure. We compared the model reflecting our hypothesis (H+) to the model reflecting its complement (Hc), allowing us to determine whether the data supported our hypothesis or not (van Lissa et al., 2021). We identified which model received the most support using Bayes factors (BF; indicating the relative support for the hypothesized models) and posterior model probabilities (PMPs; indicating the probability of each model given the data; Bergh et al., 2021; Hoijtink et al., 2019; Raftery et al., 1997). Because we assigned H+ and Hc equal prior chance, the posterior probabilities (the percentages of the posterior distribution falling in each region) were equal to the PMPs. Posterior probabilities can equivalently be presented as corresponding odds (i.e., their ratio), which, in our setting of equal prior model probabilities, is equivalent to a Bayes factor. A BF > 1 indicates support for H+, whereas a BF < 1 indicates support for Hc. The more the BF is above 1, the stronger the evidence for H+, whereas the more the BF is below 1, the stronger the evidence for Hc. A PMP ranges from 0 to 1, with higher values indicating a higher probability that the model represents the data. To aid interpretation across analytic practices (i.e., Bayesian and frequentist), we also included descriptive effect-size estimates as odds ratios (ORs) or Cohen’s ds. All data and code to reproduce our analyses are available on the Open Science Framework (OSF) at https://osf.io/rc8bv.
Preliminary analyses
Descriptive statistics for the Study 1 outcome variables are presented in Table 1. Participants’ gender and age were not associated with their proenvironmental behavior (see the Supplemental Material). Random assignment to the experimental conditions was successful; participants’ age (BF = 11.466, PMP = .920), gender (BF = 3.637, PMP = .784), climate-change skepticism (BF = 14.870, PMP = .937), and brand sensitivity (BF = 19.088, PMP = .950) were equivalent between groups. Thus, any differences in proenvironmental behavior between conditions cannot be attributed to differences in participants’ age, gender, openness to climate-change information, or preferences for familiar versus unfamiliar brands. As anticipated, we obtained fewer observations of participants’ food choices than of their donations, because participants had to actively visit the food stands (which were located outside of the classroom) for their snack, whereas the donation measure was part of the survey they completed in class (see Fig. 2 for an overview of participant flow).
Study 1 Descriptives for Primary Outcome Variables in the Full Sample and by Condition
Note: Food Choice 1 refers to observation on the day of the intervention; Food Choice 2 refers to observation at follow-up after 4 weeks. Food-choice frequencies and percentages indicate, for each subsample, the number and proportion of participants selecting Eco-label A chips. Donation means and standard deviations indicate, for each subsample, average donations and their variation (participants could donate either 0, 1, 2, or 3 euros).
Study 1 and 2 sample sizes and number of participants, with observations for the proenvironmental behavior measures for both conditions. MA = motive-alignment condition; control = control condition.
Confirmatory analyses
Our confirmatory analyses supported the hypothesis that the motive-alignment (vs. control) intervention would promote more proenvironmental behavior. On the day of the intervention, participants in the motive-alignment (vs. control) intervention condition more often chose Eco-label A chips (BF = 7.027; PMP = .875; OR = 1.27). At follow-up, participants’ Eco-label A choices were comparable between conditions (BF = 1.439; PMP = .590; OR = 1.06). However, at follow-up, participants in the motive-alignment (vs. control) intervention condition did donate more money to the environmental organization (BF = 8.966; PMP = .900; Cohen’s d = 0.13). Based on the descriptive effect sizes, the motive-alignment intervention, compared with the control intervention, led to 27% higher odds of selecting the Eco-label A chips on the day of the intervention and to 19.75% larger donations at follow-up.
Robustness analyses
We conducted robustness analyses evaluating condition differences in proenvironmental food choice while controlling for participants’ brand sensitivity at baseline. The results remained unchanged. Again, on the day of the intervention, participants in the motive-alignment (vs. control) intervention condition more often chose Eco-label A chips (BF = 9.584; PMP = .906). At follow-up, participants’ Eco-label A choices were comparable between conditions (BF = 2.249; PMP = .692).
In addition, we used inverse probability weights (IPW) based on observed covariates to examine the robustness of our results to the effects of missing data (following Gerber & Green, 2012; Gomila & Clark, 2022; see the Supplemental Material for details). Our IPW-adjusted analyses yielded results that were consistent with the unweighted results for all three outcomes, leading to the same substantive conclusions (see the Supplemental Material).
Study 2
In Study 2, we replicated Study 1 in secondary schools in China.
Method and participants
Study 2 included 554 adolescents (Mage = 15.8 years, SDage = 0.7 years; 49.1% female, 45.3% male, 5.6% unreported; 87.4% Han Chinese, 4.2% other ethnicity, 8.5% unreported). We included in this sample only youth who were present on the day of the experimental intervention (i.e., who participated in either the control or motive-alignment intervention), consistent with our preregistration. We recruited participants from two secondary schools located in urban regions of Nanjing (eastern China) and Wuhan (central China), serving mostly middle-class communities. Following our preregistered power analyses and sampling plan, we again stopped recruiting additional schools after reaching a minimum of 300 participants. At the end of the study, participants received ¥50 (equivalent to the compensation in Study 1) for their time and effort. The Institutional Review Board at Nanjing Normal University in China approved of this study (Protocol No. NNU202310010).
Procedure and measures
Study 2 followed the same procedure and used the same measures as Study 1, with two exceptions. First, in Study 2, we administered paper-and pencil surveys rather than an online survey, because online surveys were not permitted in the classrooms. Second, in Study 2, participants chose between classic-flavored chips and tomato-flavored chips (rather than classic and paprika), in keeping with the local popularity of chip flavors (confirmed during pilot interviews; see Fig. 3). For the donation measure, we offered participants an additional ¥15 (equivalent to the additional compensation in Study 1) and asked whether they would like to keep the additional compensation (i.e., donate ¥0) or donate ¥5, ¥10, or ¥15 to a youth-led environmental organization. We recoded these values as 0, 1, 2, and 3, respectively, to ease interpretation across studies.
Snack options presented at the food stands in China. “番茄味” (tomato) and “原味” (plain) are locally popular chip flavors.
Results
We followed the same preregistered analysis plan as in Study 1.
Preliminary analyses
Descriptive statistics for the Study 2 outcome variables are presented in Table 2. Participants’ gender and age were not associated with their proenvironmental behavior (see the Supplemental Material). Random assignment to the experimental conditions was successful; participants’ age (BF = 17.547, PMP = .946), gender (BF = 23.359, PMP = .959), climate-change skepticism (BF = 3.575, PMP = .781), and brand sensitivity (BF = 18.481, PMP = .949) were equivalent between groups. Thus, any differences in proenvironmental behavior between conditions cannot be attributed to differences in participants’ age, gender, openness to climate-change information, or preferences for familiar versus unfamiliar brands. As anticipated, and as in Study 1, we obtained fewer observations of participants’ food choices than of their donations (see Fig. 2).
Study 2 Descriptives for Primary Outcome Variables in the Full Sample and by Condition
Note: Food Choice 1 represents observations on the day of the intervention; Food Choice 2 represents observations at follow-up after 4 weeks. Food-choice frequencies and percentages indicate, for each subsample, the number and proportion of participants selecting Eco-label A chips. Donation means and standard deviations indicate, for each subsample, average donations in yuan and their variation (participants could donate either 0, 5, 10, or 15 yuan, recoded as 0, 1, 2, 3, respectively, for ease of interpretation across studies).
Confirmatory analyses
Our confirmatory analyses again supported the hypothesis that the motive-alignment (vs. control) intervention would promote more proenvironmental behavior. On the day of the intervention, participants in the motive-alignment (vs. control) intervention condition more often chose Eco-label A chips (BF = 5.477; PMP = .846; OR = 1.21). At follow-up, participants’ Eco-label A choices were comparable between conditions (BF = 1.698; PMP = .629; OR = 1.06). However, at follow-up, participants in the motive-alignment condition again donated more money to the environmental organization (BF = 12.435; PMP = .926; Cohen’s d = 0.13). Examination of the descriptive effect size shows that the motive-alignment intervention, compared with the control intervention, led to 21% higher odds of selecting Eco-label A chips on the day of the intervention and to 10% larger donations at follow-up.
Robustness analyses
As in Study 1, we conducted robustness analyses evaluating condition differences in proenvironmental food choice while controlling for participants’ brand sensitivity at baseline. Again, the results remained unchanged. On the day of the intervention, participants in the motive-alignment (vs. control) intervention condition more often chose Eco-label A chips (BF = 6.612; PMP = .869). Again, at follow-up, participants’ Eco-label A choices were comparable between conditions (BF = 2.412; PMP = .707).
Also, as in Study 1, we used IPW based on observed covariates to examine the robustness of our results to the effects of missing data (Gerber & Green, 2012; Gomila & Clark, 2022). Our IPW-adjusted analyses again yielded results that were consistent with the unweighted results for all three outcomes, leading to the same substantive conclusions (see the Supplemental Material).
Studies 1 and 2: Exploratory Cross-National Comparisons and Moderation Analyses
Across conditions, Chinese participants (Study 2) engaged in more proenvironmental behavior than Dutch participants (Study 1). Specifically, Chinese participants made more proenvironmental food choices, both on the day of the intervention (BF = 123.883; PMP = .992; OR = 1.40) and at follow-up (BF = 1,009.342; PMP = .999; OR = 1.58). Chinese participants also donated more to the environmental organization (BF = 2.29e13; PMP = 1.000; Cohen’s d = 0.73).
For both studies, to gain insight into individual-level factors that might enhance or limit the intervention’s effectiveness (Bryan et al., 2021), we explored moderation by age, gender, and climate-change skepticism. For each moderator, we evaluated evidence for (and against) a no-moderation model (H: Condition × Moderator = 0) because we had no prior expectations regarding moderation. We found moderate to strong evidence for the no-moderation model in both studies, for all moderators, and for all proenvironmental-behavior dependent measures (BFs > 3.287, PMPs > .767; see the Supplemental Material for detailed results). Thus, the motive-alignment intervention was equally effective at promoting proenvironmental behavior regardless of participants’ gender, age, and climate-change skepticism.
General Discussion
In a preregistered longitudinal field experiment conducted in two countries, we found that a motive-alignment intervention more effectively encouraged adolescents’ real-world proenvironmental behavior than a traditional environmental-education intervention. The motive-alignment intervention, compared with the traditional intervention, resulted in greater engagement across diverse proenvironmental behaviors. Averaging across studies, adolescents who received the motive-alignment (vs. traditional) intervention were 24% more likely to choose foods with an Eco-label A on the day of the intervention, and they donated approximately 15% more to an environmental organization at 4-week follow-up. Notably, the intervention effects were robust across samples (i.e., in the Netherlands and China), across tested individual-level factors (i.e., participants’ gender, age, and climate skepticism), and when controlling for baseline brand sensitivity (for the food-choice measure).
Implications for theory and practice
This research advances the literature by showing that motive alignment, a relatively novel approach to motivating behavior change (Bryan et al., 2019), can effectively promote adolescents’ proenvironmental behavior in the real-world context of their schools. Consequently, this research adds evidence that motive alignment offers a promising approach to encouraging behavior change in contexts where pragmatic appeals have failed (Bryan, 2020; Hecht et al., 2023). By imbuing targeted behaviors with benefits that individuals care about in the here and now, this approach reconfigures the perceived costs and benefits of those behaviors, helping individuals overcome motivational barriers to engagement (Thomaes et al., 2023; Yeager et al., 2018). In our experiments, participants knew that engaging in proenvironmental behavior could come at a cost (i.e., receiving potentially less tasty chips from an unknown brand, or earning less financial compensation). Still, presenting proenvironmental behavior as a respected, increasingly popular way to make an impact made more youth willing to incur such costs.
Our research also illustrates that motive-alignment interventions can be effective across cultures. We suggest that the particular approach to motive alignment used here may have contributed to this cross-cultural replicability. Different from motive-alignment interventions used in prior research (e.g., Bryan et al., 2019; van de Wetering et al., 2025a), our materials emphasized a positive, empowering message about youth (that youth are increasingly taking climate action into their own hands) rather than a negative, critical message about authorities (how companies are misbehaving and taking advantage of youth). In our pilot research, this positive framing more strongly facilitated motive alignment for youth in both cultures (see the Supplemental Material). Thus, motive-alignment approaches may hold promise across cultures, provided the intervention materials speak effectively to the core motives relevant to the target group.
Our research has practical implications for those seeking to encourage adolescents’ proenvironmental behavior (e.g., educators, parents, or policymakers). Our findings suggest that motive alignment offers a time- and resource-efficient approach to do so. Efforts to help adolescents internalize behaviors that contribute to environmental sustainability may benefit from emphasizing the immediately rewarding and developmentally salient aspects of such behavior, enabling adolescents to experience sufficient in-the-moment motivation to act. More broadly, the implications of this work extend to the promotion of “should” behaviors in general—that is, behaviors that individuals believe they should engage in, even if they struggle to do so in practice (Bryan, 2020). As our research suggests, motive alignment offers a promising means of bridging persistent gaps between intentions or attitudes and actual behavior.
Strengths, limitations, and future research
The present research provides a stringent field-experimental test of the sustainability motive-alignment hypothesis (Thomaes et al., 2023), demonstrating the possibility of encouraging youth’s proenvironmental behavior in the natural context of their schools. It also contributes to the field of adolescent behavior change by demonstrating that the benefits of motive alignment extend to an understudied class of behaviors (proenvironmental behaviors), and replicate across cultural contexts (the Netherlands and China). Though the observed intervention effects were not large, they were obtained relative to an active control intervention (i.e., a traditional educational intervention), suggesting that motive-alignment approaches can complement and strengthen current practices.
Our research also has limitations, which can inform future work. First, we compared a motive-alignment intervention with a control condition that included pragmatic and prescriptive appeals to behavior change. Our pilot study tested and established that the motive-alignment condition was more motive aligned than the control, but it did not test whether the conditions differed with respect to how motivating they were overall. More research is thus needed to strengthen confidence that motive alignment, rather than the inclusion of more motivational content, accounted for the observed effects. Second, although we took several steps to reduce the risk of demand effects (e.g., placing the food stands away from classrooms, emphasizing the optional nature of visiting the food stands), we cannot rule out that some participants were aware of our interest in promoting proenvironmental behavior, which could have influenced their behavior. We observed no signs of suspicion when collecting the data, and it is unlikely that demand effects alone could account for the observed condition differences. Nevertheless, future research using similar designs could include suspicion probes to increase confidence in the observed effects (Barrett et al., 2025). Third, our design choices to reduce suspicion contributed to a fairly high proportion of missing data at the food stands. Future researchers could use more embedded assessment strategies to maximize behavioral observations.
Fourth, we did not measure participants’ motivations for engaging in proenvironmental behavior (Spitzer et al., 2025). Future researchers could examine whether motive-alignment approaches foster internalized motivations, which are crucial for establishing long-term intervention effects (e.g., Vansteenkiste et al., 2004). Longer-term follow-ups are also needed to evaluate whether intervention effects are self-sustaining (e.g., though recursive processes or positive feedback loops; Yeager & Walton, 2011). Finally, future work is needed to develop scalable materials that can be integrated into existing environmental-education curricula, as these materials were developed only as a proof of concept.
Conclusion
This present research demonstrates that insights from developmental science can inform efforts to promote behavior change, particularly in contexts where pragmatic appeals have had limited impact. In particular, it provides guidance for educational and communication strategies that aim to engage youth in proenvironmental behavior in their day-to-day lives. Our findings thus illustrate how psychology can contribute to addressing climate change—an existential threat to human and planetary well-being that is rooted in human behavior (Intergovernmental Panel on Climate Change, 2023). We hope this work inspires further research into the psychology of young people regarding environmental sustainability, to help empower today’s youth to contribute to a more sustainable future for themselves and generations to come.
Supplemental Material
sj-docx-1-pss-10.1177_09567976261440239 – Supplemental material for Motive Alignment Promotes Adolescents’ Proenvironmental Behavior: A Field Experiment in Two Cultures
Supplemental material, sj-docx-1-pss-10.1177_09567976261440239 for Motive Alignment Promotes Adolescents’ Proenvironmental Behavior: A Field Experiment in Two Cultures by Jenna Spitzer, Stathis Grapsas, Astrid M. G. Poorthuis, Fan Li, Yue Song and Sander Thomaes in Psychological Science
Footnotes
Transparency
Action Editor: Roni Porat
Editor: Simine Vazire
Author Contributions
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
