Gerit Pfuhl

Governments and organizations increasingly endorse the importance of mental health in all forms of policy, though emphasis in measurement still remains on indicators such as GDP, depression, and other common disorders. In this study of 92 countries ( n = 53,799), we assess the mental health of individuals around the world alongside measures of financial and social well-being. Findings include a trend of languishing mental health across many groups, particularly among younger adults and within higher-income countries. While positive trends are observed for education and income, a number of concerning trends also emerge, particularly the poor results for young adults and generally moderate results for employed individuals. In sum, we find an alarming but perhaps unsurprisingly low level of positive mental health in a substantial number of countries, but also identify specific economic, social, and other indicators that policymakers could target in improving the well-being of the populations they serve.

The descriptors of the European avalanche danger scale, in use since 1993, are being revised. In June 2025 EAWS forecasters agreed on a new provisional wording and, in doing so, provided its content validation. Because the descriptors are first of all a public communication tool, we ran a comprehension validation: a survey testing whether end users read the wording as forecasters intend. It ran online from 3 February to 8 April 2026 in six languages and retained 17,335 responses after cleaning, of which 15,749 were completed. All wording comparisons were randomised between subjects. We evaluated each proposal against three conditions: correctness (do users recover the order forecasters intend?), simplicity (how easy is the task?), and increase (does the wording convey how much the danger increases from level to level?). For avalanche occurrence the main finding is reassuring: all four wording variants were ranked accurately (83–92% correct), and most differences that reach significance in the pooled data do not survive within individual languages. Only the variant that drops the distribution wording is clearly weaker. On the increase condition the variants do separate. Respondents placed the four statements on a continuous danger scale; wording level 2 as “in some places” rather than “in a few places” moves it closer to level 3 and shrinks the perceived difference between moderate to considerable from 36% to 31% of the distance between levels 1 and 4 (Cohen’s d = 0.39, p < 0.001). For size and impact the wording effect is large: adding size class words (small, medium, large, very large) to the impact wording raised accuracy from 61% to 95% and self-reported ease from 4.7 to 6.1 (Cohen’s d = 1.04). Most of that failure comes from using “bury and kill” at both levels 1 and 2, which makes them hard to tell apart. We also report how users read seven phrases that forecasters already use. For avalanche occurrence, the largest differences are between languages rather than between wordings.

Governments and organizations now increasingly endorse the importance of mental health in all forms of policy, though emphasis in measurement still remains on indicators such as GDP, depression, and other common disorders. In this study of 92 countries (n = 53,774), we assess the mental health of individuals around the world alongside measures of financial and social well-being. Findings include a trend of languishing mental health across many groups, particularly among younger adults and within higher-income countries. While positive trends are observed for education and income, a number of concerning trends also emerge, particularly the loss of the U-shape for age and generally moderate results for employed individuals. In sum, we find an alarming but perhaps unsurprisingly low level of positive mental health in a substantial number of countries, but also identify specific economic, social, and other indicators that policymakers could target in improving the well-being of the populations they serve.

Many current issues (e.g., climate change) plausibly also affect future generations, but research suggests that people are less concerned about the well-being of future humans. Future humans are temporally, but also socially distant. We tested whether intergenerational judgment tracks more closely with temporal discounting or social discounting. In a preregistered study, N=5,801 undergraduate participants from 35 universities across 12 countries completed measures of social discounting, temporal discounting, intergenerational discounting, and intergenerational concern. Lower social and temporal discounting each predicted less intergenerational discounting and greater intergenerational concern, although, social discounting emerged as the stronger predictor. Exploratory analyses showed a social-by-temporal discounting interaction for intergenerational discounting. Participants who extend strong, temporally-invariant concern to future generations also showed less discounting across intergenerational but also social and temporal domains. The overall pattern generalized across preregistered secondary discounting metrics. These findings suggest that concern for future generations reflects both intertemporal and interpersonalvaluation.

Outdoor recreation offers a natural testbed of informal leader emergence (ILE): autonomy is prized, yet coordination can be safety-critical. We study ILE in backcountry (BC) skiing across two linked studies. Study 1 (215 skiers) compared group-function appraisals between ILE-perceivers and non-perceivers. ILE-perceivers rated Skill lower (overall group dynamics slightly lower), while Trust and Confidence were comparably good. Factor analyses revealed distinct evaluative lenses: with-leader accounts emphasized Openness, Relevant competence, and Sense of involvement; without-leader accounts emphasized Preparedness and Harmony—interpreted as group skill configuration and decision-making climate. Study 2 (250 skiers in 100 groups) used path analysis to test how skill configuration and decision processes relate to ILE and safety outcomes. Greater within-group dispersion of avalanche-assessment skill (Group Skill Dispersion; GSD) predicted more ILE and related negatively to both group mitigation practices enacted (nGMP) and Individual Confidence; Decision Involvement (DI) robustly increased confidence. By contrast, the group’s Maximum Assessment Skill, Democratic Decision Climate, and ILE itself were not reliable correlates of nGMP or Confidence once GSD and DI were modeled. Exploratory multi-group tests indicated the GSD–nGMP association was more negative when ILE was low; confidence pathways were largely invariant. Together, the studies suggest ILE in BC skiing is chiefly compensatory for uneven intra-group assessment skills, while skill configuration and participation are the proximate levers of group functioning. Practically, reducing GSD through partner matching and on-trip calibration, and facilitating DI, are key levers—and when GSD is inevitably high, recognizing a leader can restore group outcomes to those achieved under low GSD.

Discrepancies between what we think we do and what we actually do can have fatal consequences in avalanche terrain. This study investigates whether double-loop learning, reflecting on underlying assumptions rather than merely adjusting behaviors, can be stimulated in avalanche courses to improve safe decision-making. Through a season-long course with ten participants in Northern Norway, we analyzed instructor field notes and participant Reflectometer entries, identifying 201 key learning moments across seven themes. We found that double-loop learning emerged through both human interactions and environmental encounters, occurring bidirectionally between instructors and participants. Uncomfortable dislocatory moments more frequently triggered double-loop learning but were not necessary; positive moments also prompted assumption revision. Based on these findings, we propose a ‘mind-the-gap’ approach to avalanche instruction: helping learners experience, identify, and close discrepancies between intended and actual behavior. This practical method offers avalanche educators, and practitioners in other high-stakes domains, a tool for transforming invisible assumptions and decision-making traps into visible sources for learning.

The same dataset can be analysed in different justifiable ways to answer the same research question, potentially challenging the robustness of empirical science^1-3. In this crowd initiative, we investigated the degree to which research findings in the social and behavioural sciences are contingent on analysts' choices. We examined a stratified random sample of 100 studies published between 2009 and 2018, in which, for one claim per study, at least five reanalysts independently reanalysed the original data. The statistical appropriateness of the reanalyses was assessed in peer evaluations, and the robustness indicators were inspected along a range of research characteristics and study designs. We found that 34% of the independent reanalyses yielded the same result (within a tolerance region of ±0.05 Cohen's d) as the original report; with a four times broader tolerance region, this indicator increased to 57%. Of the reanalyses conducted, 74% reached the same conclusion as the original investigation, 24% yielded no effects or inconclusive results and 2% reported the opposite effect. This exploratory study indicates that the common single-path analyses in social and behavioural research should not be simply assumed to be robust to alternative analyses⁴. Therefore, we recommend the development and use of practices to explore and communicate this neglected source of uncertainty.

The same dataset can be analysed in different justifiable ways to answer the same research question, potentially challenging the robustness of empirical science1–3. In this crowd initiative, we investigated the degree to which research findings in the social and behavioural sciences are contingent on analysts’ choices. We examined a stratified random sample of 100 studies published between 2009 and 2018, where for one claim per study, at least five re-analysts independently re-analysed the original data. The statistical appropriateness of the re-analyses was assessed in peer evaluations, and the robustness indicators were inspected along a range of research characteristics and study designs. We found that 34% of the independent re-analyses yielded the same result (within a tolerance region of +/- 0.05 Cohen’s d) as the original report; with a four times broader tolerance region, this indicator rose to 57%. Regarding the conclusions drawn, 74% of analyses were reported to arrive at the same conclusion as in the original investigation; 24% to no effects/inconclusive result, and 2% to the opposite effect as in the original investigation. This exploratory study suggests that the common single-path analyses in social and behavioural research should not simply be assumed to be robust to alternative analyses4. Therefore, we recommend the development and use of practices to explore and communicate this neglected source of uncertainty.