Information-based interventions are still the most commonly used intervention approach to motivate people to engage in pro-environmental behaviors. However, recently, researchers assume that the effectiveness of such information depends on the extent to which recipients endorse biospheric values (the so-called ‘values-as-moderator’ hypothesis). If this hypothesis is correct, it would have important implications for the use of information appeals: they could not motivate the entire population, but only those population groups with high biospheric values. Two experimental studies using case vignettes were conducted to investigate the effects of biospheric values prioritization, an informational and a socio-normative appeal, as well as the interaction between biospheric values and those two interventions on two different pro-environmental behavioral intentions: everyday ‘low-cost’ behavior of purchasing (organic) coffee (Study 1) and the more difficult ‘high-cost’ behavior of adopting a heating style that prevents rebound effects (Study 2). In Study 1 we additionally examined the influence of the degree of privacy in which a behavior takes place on the purchase intention. Both studies were conducted as online survey studies in Germany. A total of 800 participants took part in each study. We used moderated logistic regression models to examine the main and interaction effects of biospheric values and the two different appeals on participants’ pro-environmental behavioral intentions. Both studies showed a significant main effect of the informational appeal on pro-environmental behavioral intentions. In Study 1, but not in Study 2, the main effect of the socio-normative appeal was also significant, especially if the purchase decision took place in a public space. However, no consistent evidence of a meaningful effect of biospheric values was found: the main effect of biospheric values was statistically significant only in Study 1, and the postulated interactions between biospheric values and interventions were not statistically significant in either study. Further theoretical and practical implications of the reported results are discussed.
12 April 2017 in AStA Advances in Statistical Analysis
International air traffic has contributed to the global spread of SARS-CoV-2 and its variants. In early 2023, wastewater-based epidemiology (WBE) has been implemented at airports as a surveillance tool to detect emerging variants at short notice. This study investigates the feasibility and challenges of applying WBE at Berlin Brandenburg (BER) Airport, including a rapid implementation of wastewater sampling and analysis under unprecedented circumstances. For this purpose, aircraft and airport wastewater was sampled over 13 weeks. Established sampling and analysis protocols for municipal wastewater treatment plants (WWTPs) had to be adapted to the specific conditions of the airport environment. SARS-CoV-2 RNA was quantified and sequenced, revealing SARS-CoV-2 mutations not previously observed in clinical surveillance data in Germany. Despite the logistical and methodological challenges, the study demonstrates that WBE can serve as an early warning system for pathogen introduction. However, our study also underscores the need for realistic timelines for the establishment and validation of WBE monitoring strategies in new contexts. Investments in the establishment of WBE systems, e.g., infrastructure, protocols, trained personnel, and a network of stakeholders at strategic nodes including airports, can act as an effective tool for pandemic preparedness and global health security.
10 February 2026 in Science of The Total Environment
Wastewater-based surveillance complements case-based surveillance systems by capturing pathogen signals shed in stool and other bodily excretions, enabling population-level monitoring independent of clinical testing. Its utility during the COVID-19 pandemic has been widely explored, but its responsiveness and interpretability relative to case-based systems remain insufficiently understood. We analyzed German nationwide data on COVID-19 or SARS-CoV-2 from July 2022 to December 2024, using wastewater surveillance and four case-based surveillance systems. These comprise syndromic surveillance systems at the population as well as the primary care level, and mainly laboratory-confirmed notification data, all aimed at monitoring COVID-19 incidence in Germany. We assessed agreement between wastewater viral load and disease incidence using visual inspection, cross-correlation analysis, and an estimated prevalence dynamic informed by a fecal shedding model. We derived retrospective translation factors and compared week-to-week trend directions between systems. Finally, we tested the predictive power of wastewater data using classification models to anticipate current week incidence trends. Wastewater SARS-CoV-2 viral load closely correlates with COVID-19 incidence trends from case-based systems, showing similar timing of peaks and troughs without notable time lags. Cross-correlation coefficients are highest with syndromic surveillance systems (up to 0.87) and lowest with notification data (0.43). Retrospective translation into incidence estimates works well on average, but week-to-week translation varies considerably. Wastewater-based models correctly predict the current week's trend, as indicated by at least three of the four case-based systems, with about 68 % probability. Wastewater surveillance correlates well with COVID-19 incidence, but real-time translation to incidence lacks precision. Trend prediction for the current week may demonstrate improved accuracy and may be valuable when case reporting is limited or delayed.
28 May 2026 in Proceedings of the National Academy of Sciences
Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. We experimentally test how collaboration between researchers and LLM assistants influences the reproduction of quantitative social science findings across different levels of AI autonomy. We randomly assigned 288 researchers to 103 teams working under three conditions: human-only, AI-assisted (using ChatGPT as a collaborative tool), or AI-led (ChatGPT operating with minimal human oversight). Teams reproduced published results from leading social science journals, detected coding errors, and proposed robustness checks. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%) and performed similarly on most outcomes, except human-only teams identified significantly more major coding errors. Both substantially outperformed AI-led teams, which achieved only a 37% reproduction rate, detected fewer errors across all categories, proposed weaker robustness checks, and required more time. This autonomous approach, however, likely represents only a lower bound of AI capabilities. Despite rapid model advances, expert human judgment currently remains indispensable for reliable empirical verification. While AI assistance did not degrade most outcomes, it provided no measurable advantages and was associated with reduced detection of major errors. However, the 37% autonomous reproduction rate indicates that AI could provide value in settings where scale or cost constraints preclude human review of papers, even though general-purpose LLMs offer no immediate advantages for human-supervised verification.
Background: Wastewater-based surveillance complements case-based surveillance systems by capturing pathogen signals shed in stool, enabling population-level monitoring independent of clinical testing. Its utility during the COVID-19 pandemic has been widely explored, but its responsiveness and interpretability relative to case-based systems remain insufficiently understood. Methods: We analyzed data on COVID-19 or SARS-CoV-2 from July 2022 to December 2024 across 131 weeks, using wastewater surveillance and four case-based surveillance systems. These comprise syndromic surveillance systems at the population as well as the primary care level, and notification data, all aimed at monitoring COVID-19 incidence in Germany. We assessed agreement between wastewater viral load and disease incidence using visual inspection, cross-correlation analysis, and an estimated prevalence dynamic informed by a literature-based fecal shedding model. We derived retrospective translation factors and compared week-to-week trend directions between systems. Finally, we tested the predictive power of wastewater data using classification models to anticipate current week incidence trends. Results: Wastewater SARS-CoV-2 viral load closely correlates with COVID-19 incidence trends from case-based systems, showing similar timing of peaks and troughs without notable time lags. Cross-correlation coefficients are highest with syndromic surveillance systems (up to 0.87) and lowest with notification data (0.43). Retrospective translation into incidence estimates works well on average, but week-to-week translation varies considerably. Wastewater-based models correctly predict the current week trend, as indicated by at least three of the four case-based systems, with about 68% probability. Conclusion: Wastewater surveillance correlates well with COVID-19 incidence, but real-time translation to incidence lacks precision. Trend prediction for the current week may demonstrate improved accuracy and may be valuable when case reporting is limited or delayed.