The study, published in Frontiers in Behavioral Economics, was conducted by economist Lea Cassar, economist Tim Michael Neu, and psychologist Petra Jansen. With financial support from a large German health insurer, the team offered a free eight-week online mindfulness course to students at the University of Regensburg in the 2024 summer term. Because not all applicants could be accommodated, the 304 eligible respondents were randomly assigned in equal numbers to a treatment group that received the course and a control group that received nothing, using a re-randomization procedure to ensure the groups were balanced on baseline characteristics. The study was pre-registered on the Open Science Framework before the intervention began, and outcomes were measured twice—once at application and once a week after the course ended—allowing powerful within-subject comparisons.

The course itself was no throwaway wellness webinar. It consisted of eight live 90-minute Zoom sessions led by two certified teachers, with groups of roughly thirty participants each. The first four sessions drew on the well-established Mindfulness-Based Stress Reduction program developed by Jon Kabat-Zinn, combining formal sitting meditation, body awareness, and mindful movement. Sessions five and six were based on the Mindful Self-Compassion curriculum, the seventh on Rick Hanson’s neuroplasticity training, and the eighth integrated elements from MBSR, self-compassion practice, and the tradition of Thich Nhat Hanh, with a deliberate emphasis on connectedness. Participants completed weekly homework, often recording and uploading videos of fifteen to twenty minutes of meditation. Attendance averaged 63.6 percent across the weeks, and 71.2 percent of treated students attended at least four sessions.

The researchers’ central hypothesis was that mindfulness training would spill over into socially relevant economic behavior. Their theoretical framework held that increases in mindfulness improve mental health and strengthen connectedness—to oneself, to others, and to nature—which in turn should make people behave in more environmentally friendly ways and distribute income more equally. Crucially, rather than relying on attitude questionnaires, which dominate this literature and are vulnerable to social desirability bias, the team measured actual behavior with incentivized tasks in which real money and real carbon dioxide reductions were at stake.

The pro-environmental measures were inventive. In a certificate game, students bid from a 100-euro endowment for a genuine CO2 certificate from the EU Emissions Trading System, worth 100 euros and permitting 1.2 tons of emissions; winning bidders saw their certificates retired, while their unspent money was paid out—creating a real trade-off between planetary and personal gain. A travel lottery forced choices between flight and train vouchers of varying values, capturing willingness to forgo money for lower-emission transport. A product choice task measured whether students would exert costly effort, decoding product labels, to identify and select the environmentally friendly option. Distributive preferences were measured with a fairness allocation game in which participants acted as impartial spectators, dividing a 30-euro wage between two real Prolific crowdworkers who differed in productivity, so that fairness views could be separated from self-interest. Survey measures of climate and redistribution policy preferences rounded out the battery.

On its own terms, the intervention worked. Mindfulness, measured with the Freiburg Mindfulness Inventory, rose by roughly 0.24 to 0.38 standard deviations relative to the control group, depending on the statistical specification, a highly significant effect. The secondary psychological outcomes told a partly encouraging story: perceived stress fell by about 0.28 standard deviations, self-compassion increased by about 0.37 standard deviations, and connectedness to nature improved by around 0.23 standard deviations. But flourishing and connectedness to others were untouched, meaning the hypothesized mediating channels were only selectively activated.

When the researchers turned to behavior, the expected cascade of change never materialized. Across the three environmental tasks, the point estimates were positive—between 0.03 and 0.26 standard deviations—but none reached statistical significance in the endline or value-added models, and the pattern did not hold consistently across specifications. Policy preferences on climate protection versus economic growth, government climate action, reducing inequality versus growth, and progressive taxation were likewise unaffected, with consistently positive but insignificant coefficients. The fairness allocation game showed no treatment effect in aggregate or in any of its individual stages, in which one worker’s performance ranged from one percent to more than one hundred percent better than the other’s. In short, students who became measurably more mindful, more self-compassionate, and more connected to nature did not become measurably greener or more egalitarian in anything they actually did or endorsed.

Why did meaningful psychological change fail to translate into behavioral change? One candidate explanation is selection. Participation was voluntary, and the course attracted a self-selected, predominantly female sample—about 81 percent women, against roughly 61 percent in the university’s student body overall—over-representing humanities students. Prior research shows that women tend, on average, to hold stronger pro-environmental and redistributive preferences, and the baseline data confirmed it: 80.9 percent of participants bid at least 50 euros in the CO2 game, about 70 percent already prioritized climate action and inequality reduction over economic growth, and 86.3 percent said higher earners should pay a larger share of their income in taxes. The course, in other words, may have recruited precisely the people whose behavior it was hoped to change, leaving almost no headroom for improvement. Yet post-stratification weighting to match the university’s gender composition produced nearly identical estimates, suggesting selection alone cannot explain the null results.

Alternative explanations point to dosage and mechanism. Moderation analyses found that students whose psychological gains were largest did not show stronger behavioral effects, which argues against the idea that the channels were simply too weakly activated for some individuals. However, exploratory analyses splitting participants by attendance hint at a dose-response pattern: high attenders showed marginally significant positive effects on several environmental outcomes and larger psychological improvements, while low attenders showed none. This pattern is observational and open to endogeneity concerns, but it suggests that a more intense or in-person intervention might yet move behavior. The authors also note that the randomized incentive structure produced modest expected payouts, though social desirability bias, had it operated, would presumably have inflated rather than suppressed treatment effects.

The broader implications are uncomfortable for a popular narrative. Much of the existing literature linking mindfulness to pro-environmental attitudes and prosociality rests on cross-sectional correlations or self-reported intentions; the small number of randomized studies with incentivized outcomes, including a recent trial finding that mindfulness training actually reduced preferences for pro-environmental outcomes, paints a far more cautious picture. This study—one of the largest and best-controlled to date, combining live instruction, a real-world health insurance partnership, and hard behavioral measures—supports that caution. Mindfulness programs remain relatively cheap, scalable, and genuinely effective at reducing stress and enhancing well-being, and those benefits are not in question. But the hope that meditation can quietly serve as climate policy or redistribution policy in disguise is, for now, unsupported. If societies want citizens to fly less, pay for emissions reductions, or back progressive taxation, the evidence suggests they will need to appeal to something other than eight weeks of breathing exercises.

Subject of Research: Effects of mindfulness training on pro-environmental and distributive behavior in a randomized controlled trial

Article Title: From mental well-being to pro-environmental and distributive behavior? Evidence from a randomized mindfulness intervention

Article References: From mental well-being to pro-environmental and distributive behavior? Evidence from a randomized mindfulness intervention. (n.d.). https://doi.org/10.3389/frbhe.2026.1875793

Image Credits: AI Generated

DOI: 10.3389/frbhe.2026.1875793

Keywords: mindfulness, randomized controlled trial, pro-environmental behavior, distributive preferences, mental health, behavioral economics, climate policy, self-compassion, connectedness to nature, university students, stress reduction, spillover effects