Replication of the natural selection of bad science

Impact of redefining statistical significance on P-hacking and false positive rates: An agent-based model

In recent years, concern has grown about the inappropriate application and interpretation of P values, especially the use of P<0.05 to denote "statistical significance" and the practice of P-hacking to produce results below this threshold and selectively reporting these in publications. Such behavior is said to be a major contributor to the large number of false and non-reproducible discoveries found in academic journals. In response, it has been proposed that the threshold for statistical significance be changed from 0.05 to 0.005. The aim of the current study was to use an evolutionary agent-based model comprised of researchers who test hypotheses and strive to increase their publication rates in order to explore the impact of a 0.005 P value threshold on P-hacking and published false positive rates. Three scenarios were examined, one in which researchers tested a single hypothesis, one in which they tested multiple hypotheses using a P<0.05 threshold, and one in which they tested multiple hypotheses using a P<0.005 threshold. Effects sizes were varied across models and output assessed in terms of researcher effort, number of hypotheses tested and number of publications, and the published false positive rate. The results supported the view that a more stringent P value threshold can serve to reduce the rate of published false positive results. Researchers still engaged in P-hacking with the new threshold, but the effort they expended increased substantially and their overall productivity was reduced, resulting in a decline in the published false positive rate. Compared to other proposed interventions to improve the academic publishing system, changing the P value threshold has the advantage of being relatively easy to implement and could be monitored and enforced with minimal effort by journal editors and peer reviewers.

Increasing Value and Reducing Waste of Research on Neurofeedback Effects in Post-traumatic Stress Disorder: A State-of-the-Art-Review

Post-Traumatic Stress Disorder (PTSD) is often considered challenging to treat due to factors that contribute to its complexity. In the last decade, more attention has been paid to non-pharmacological or non-psychological therapies for PTSD, including neurofeedback (NFB). NFB is a promising non-invasive technique targeting specific brainwave patterns associated with psychiatric symptomatology. By learning to regulate brain activity in a closed-loop paradigm, individuals can improve their functionality while reducing symptom severity. However, owing to its lax regulation and heterogeneous legal status across different countries, the degree to which it has scientific support as a psychiatric treatment remains controversial. In this state-of-the-art review, we searched PubMed, Cochrane Central, Web of Science, Scopus, and MEDLINE and identified meta-analyses and systematic reviews exploring the efficacy of NFB for PTSD. We included seven systematic reviews, out of which three included meta-analyses (32 studies and 669 participants) that targeted NFB as an intervention while addressing a single condition-PTSD. We used the MeaSurement Tool to Assess systematic Reviews (AMSTAR) 2 and the criteria described by Cristea and Naudet (Behav Res Therapy 123:103479, 2019, https://doi.org/10.1016/j.brat.2019.103479 ) to identify sources of research waste and increasing value in biomedical research. The seven assessed reviews had an overall extremely poor quality score (5 critically low, one low, one moderate, and none high) and multiple sources of waste while opening opportunities for increasing value in the NFB literature. Our research shows that it remains unclear whether NFB training is significantly beneficial in treating PTSD. The quality of the investigated literature is low and maintains a persistent uncertainty over numerous points, which are highly important for deciding whether an intervention has clinical efficacy. Just as importantly, none of the reviews we appraised explored the statistical power, referred to open data of the included studies, or adjusted their pooled effect sizes for publication bias and risk of bias. Based on the obtained results, we identified some recurrent sources of waste (such as a lack of research decisions based on sound questions or using an appropriate methodology in a fully transparent, unbiased, and useable manner) and proposed some directions for increasing value (homogeneity and consensus) in designing and reporting research on NFB interventions in PTSD.

Replicability of simulation studies for the investigation of statistical methods: the RepliSims project

Results of simulation studies evaluating the performance of statistical methods can have a major impact on the way empirical research is implemented. However, so far there is limited evidence of the replicability of simulation studies. Eight highly cited statistical simulation studies were selected, and their replicability was assessed by teams of replicators with formal training in quantitative methodology. The teams used information in the original publications to write simulation code with the aim of replicating the results. The primary outcome was to determine the feasibility of replicability based on reported information in the original publications and supplementary materials. Replicasility varied greatly: some original studies provided detailed information leading to almost perfect replication of results, whereas other studies did not provide enough information to implement any of the reported simulations. Factors facilitating replication included availability of code, detailed reporting or visualization of data-generating procedures and methods, and replicator expertise. Replicability of statistical simulation studies was mainly impeded by lack of information and sustainability of information sources. We encourage researchers publishing simulation studies to transparently report all relevant implementation details either in the research paper itself or in easily accessible supplementary material and to make their simulation code publicly available using permanent links.

Correction to: 'The natural selection of bad science' (2016) by Paul E. Smaldino and Richard McElreath

[This corrects the article DOI: 10.1098/rsos.160384.][This corrects the article DOI: 10.1098/rsos.160384.].