This is the authors' abstract. We don't add key points for this paper.
Many research fields rely on assessing the behaviour of wild or captive animals, using observational and experimental protocols. This work is highly susceptible to sampling biases which, when unrecognized or undisclosed, can cause replication failure. Here, we illustrate how the STRANGE framework can be used to improve reporting quality and, ultimately, replicability in animal behaviour research. Using a combination of real-world and hypothetical case studies, we demonstrate how structured reporting of information about subjects and testing procedures can help contextualize mixed findings, enabling both improved assessments of research replicability and deeper biological insight. As a flexible tool for identifying, mitigating and reporting sampling biases, the STRANGE framework is already being leveraged productively by a wide range of user groups, including researchers, consortia, educators and journals. Our case study analyses have revealed several specific opportunities to support further community engagement and uptake. We recommend using the STRANGE framework to develop teaching materials on sampling biases, to support the work of many-labs consortia, to prepare robust preregistrations and registered reports and to conduct systematic retrospective analyses of seemingly unreliable findings. With its light-touch, researcher-led approach, the STRANGE framework has initiated an exciting grassroots push towards improved study designs, reporting and replicability in animal behaviour research.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Webster et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: