Dorothy V M Bishop

In a study of patterns of language laterality in left-and right-handers, Woodhead et al.(2021) noted that several tasks showed no bias to the left-hemisphere in left-handed individuals. This might appear to suggest that these functions were mediated by the two hemispheres working together equally in left-handers. Here, I consider an alternative possibility: that individuals show lateral bias on these tasks, but the bias can occur to either left or right. Further analysis of the distributions of data from individuals in Woodhead et al is compared with simulated data. The pattern of results suggests that the impression of bilateral language processing may be an artefact of reliance on group data: even though the group mean does not differ from zero, a high proportion of individuals are biased to left or right.

Recent studies in psychology have documented how analytic flexibility can result in different results from the same dataset. Here we demonstrate a package in the R programming language, DeclareDesign, which uses simulated data to diagnose the properties of analytic designs. To illustrate features of the package, we contrast two analyses of a randomised controlled trial (RCT) of GraphoGame, an intervention to help children learn to read. The initial analysis (NFER) found that the intervention was ineffective, but a subsequent reanalysis (Cambridge) concluded that GraphoGame significantly improved children’s reading. With DeclareDesign we can simulate data where the truth is known, and thus can identify which analysis is optimal for estimating the intervention effect, using “diagnosands”, including bias, precision, and power. The simulations showed that the NFER analysis accurately estimated intervention effects, whereas selection of a subset of data in the Cambridge analysis introduced substantial bias, overestimating the effect sizes. This problem was exacerbated by inclusion of multiple outcome measures in the Cambridge analysis. Much has been written about the dangers of performing reanalyses of data from RCTs that violate the randomisation of participants to conditions; simulated data make this message clear and quantify the extent to which such practices introduce bias. The simulations confirm the original NFER conclusion that the intervention has no benefit over “business as usual”. In this tutorial we demonstrate several features of DeclareDesign; this package can simulate observational and well as experimental research …

In October 2017, Proceedings of the Royal Society B published an article by Le Floch & Ropars [1] claiming that dyslexia could be caused by a visual anomaly that led to confusion between images from the two eyes when reading letters. This article attracted considerable media attention, with an Altmetric score of 829, including coverage by 83 news outlets, and has subsequently been cited in promotional material for devices that are designed to ameliorate the problem. A number of international experts raised concerns about the study on the postpublication peer review site PubPeer [2], but this has not led to any moderation of the claims made for therapeutic implications of the study. Given that our view is that the study suffers from methodological, interpretive and ethical problems that should have precluded publication, we are grateful to the editors for providing this opportunity to document these issues.

In this exploratory study we compare and contrast two methods for deriving a laterality index (LI) from functional magnetic resonance imaging (fMRI) data: the weighted bootstrapped mean from the Laterality Toolbox (toolbox method), and a novel method that uses subtraction of activations from homologous regions in left and right hemispheres to give an array of difference scores (mirror method). Data came from 31 individuals who had been selected to include a high proportion of people with atypical laterality when tested with functional transcranial Doppler ultrasound (fTCD). On two tasks, word generation and semantic matching, the mirror method generally gave better agreement with fTCD laterality than the toolbox method, both for individual regions of interest, and for a large region corresponding to the middle cerebral artery. LI estimates from this method had much smaller confidence intervals (CIs) than …

Background: Organisations known as paper mills charge authors to place fraudulent papers in the academic literature. Publishers have been slow to tackle the problem, but are now starting to devise methods for identifying paper mill products. However, little attention has been paid to the topic of complicit editors, who can take over special issues of journals and then publish many fraudulent articles. To date, activities of such editors have been documented by a handful of sleuths on social media and on the website PubPeer. This paper reports a descriptive study that documents more systematically the presence of “red flags” indicative of paper mill activity in special issues from the Wiley-Hindawi Open Access publishing partnership.Methods: A spreadsheet was created from the Hindawi website with records for all published articles during 2022. Initial analysis focused on initial Editor Response Time (RT) in ten journals that had been identified by sleuths as having high rates of problematic papers. In a second step, the whole dataset was scrutinized to identify editors who had handled at least 10 articles. These were divided into those who had had at least one PubPeer comment flagging dubious content or citations, and those with no PubPeer comments.Results: A cutoff of 22 days was identified as corresponding to the 2nd percentile of Editor RT for regular articles not in special issues. Plots show that in the 10 selected journals, some special issues have 50% or more of Editor RTs below this cutoff, raising questions as to whether an appropriate peer review process had taken place. Editors with articles flagged on PubPeer processed significantly …

Fraudulent published papers used to be thought to be rare, but in recent years there has been growing awareness of coordinated activities by organizations that charge authors a fee to plant articles in reputable journals. These are known as paper mills. We reflect here on how Mind, Brain and Education suffered from such an attack in 2022 and 2023, discussing what we have learnt from this experience about how paper mills might operate in social sciences, how to spot signs that an article is fraudulent, and what steps might be taken to prevent such attacks in future.

Studies have highlighted an association between motor laterality and speech production laterality. It is thought that common demands for sequential processing may underlie this association. However, most studies in this area have relied on relatively small samples and have infrequently explored the reliability of the tools used to assess lateralization. We, therefore, established the validity and reliability of an online battery measuring sequence-based motor laterality and language laterality before exploring the associations between laterality indices on language and motor tasks. The online battery was completed by 621 participants, 52 of whom returned to complete the battery a second time. The three motor tasks included in the battery showed good between-session reliability (r ≥ .78) and were lateralized in concordance with hand preference. The novel measure of speech production laterality was left …

This study reports on the feasibility of using the Test of Complex Syntax- Electronic (TECS-E), as a self-directed app, to measure sentence comprehension in children aged 4 to 5 ½ years old; how testing apps might be adapted for effective independent use; and agreement levels between face-to-face supported computerized and independent computerized testing with this cohort. A pilot phase was completed with 4 to 4;06-year-old children, to determine the appropriate functional app features required to facilitate independent test completion. Following the integration of identified features, children completed the app independently or with adult support (4–4;05 (n = 22) 4;06–4;11 months (n = 55) and 5 to 5;05 (n = 113)) and test re-test reliability was examined. Independent test completion posed problems for children under 5 years but for those over 5, TECS-E is a reliable method to assess children’s understanding of …