Search

Methods sections are often missing essential details. Methodological shortcut citations, in which authors cite previous papers instead of describing the method in detail, may contribute to this problem. This meta-research study used 3 approaches to examine shortcut citation use in neuroscience, biology, and psychiatry. First, we assessed current practices in more than 750 papers. More than 90% of papers used shortcut citations. Other common reasons for using citations in the methods included giving credit or specifying what was used (who or what citation) and providing context or a justification (why citation). Next, we reviewed 15 papers to determine what can happen when readers follow shortcut citations to find methodological details. While shortcut citations can be used effectively, they can also deprive readers of essential methodological details. Problems encountered included difficulty identifying or accessing the cited materials, missing or insufficient descriptions of the cited method, and shortcut citation chains. Third, we examined journal policies. Fewer than one quarter of journals had policies describing how authors should report previously described methods. We propose that methodological shortcut citations should meet 3 criteria; cited resources should provide (1) a detailed description of (2) the method used by the citing authors', and (3) be open access. Resources that do not meet these criteria should be cited to give credit, but not as shortcut citations. We outline actions that authors and journals can take to use shortcut citations responsibly, while fostering a culture of open and reproducible methods reporting. Methods sections often lack critical details needed to reproduce an experiment; the practice of citing previous papers instead of describing the methods in detail may contribute to this problem. An analysis of >800 papers across three fields shows that >90% of papers use at least one shortcut citation, that these significantly impair reconstruction of the original method, and that <25% of journals have policies relating to previously described methods. [ABSTRACT FROM AUTHOR]

Scientific advances due to conceptual or technological innovations can be revealed by examining how research topics have evolved. But such topical evolution is difficult to uncover and quantify because of the large body of literature and the need for expert knowledge in a wide range of areas in a field. Using plant biology as an example, we used machine learning and language models to classify plant science citations into topics representing interconnected, evolving subfields. The changes in prevalence of topical records over the last 50 years reflect shifts in major research trends and recent radiation of new topics, as well as turnover of model species and vastly different plant science research trajectories among countries. Our approaches readily summarize the topical diversity and evolution of a scientific field with hundreds of thousands of relevant papers, and they can be applied broadly to other fields. Our ability to understand the progress of science through the evolution of research topics is limited by the need for specialist knowledge and the exponential growth of the literature. This study uses artificial intelligence and machine learning approaches to demonstrate how a biological field (plant science) has evolved, how the model systems have changed, and how countries differ in terms of research focus and impact. [ABSTRACT FROM AUTHOR]

Curated scientific databases catalogue and amplify research findings to maximize their reach. Authors should write their papers with this in mind, ensuring that data are accurate, easy to extract, and presented in standardized formats. Scientific data is becoming increasingly complex, making data curation increasingly difficult. In this Perspective, a group of biocurators present some simple guidelines for data accessibility that will help to increase the reach of published studies. [ABSTRACT FROM AUTHOR]

The letter responds to a previous paper and discusses the inclusion of both sexes in scientific studies. The authors argue that, unless there is evidence to the contrary, it is reasonable to assume equal variance between sexes and proceed with a standard sample size. They refute the claim that heteroscedasticity (unequal variance) between sexes is the norm, citing studies that support the assumption of homoscedasticity (equal variance). The authors emphasize the importance of adapting statistical analysis and power calculations if evidence of sex-related heteroscedasticity emerges. [Extracted from the article]