Metascience Matters

Randy Ellis

The research integrity channel for people who want science to work. Metascience Matters covers scientific fraud, replication failures, peer review, research incentives, AI/data reliability, and the people building better science. Hosted by Randy Ellis, a computational biologist and neuroscientist. New interviews, clips, and explainers on how science breaks and how to fix it. Guests, tips, and collaborations: metasciencematters@gmail.com.

  1. 6d ago

    Brain stimulation, fraud, and the RIOT Science Club | Sam Westwood

    Samuel Westwood is Senior Lecturer in Psychology (Education) at the Institute of Psychiatry, Psychology & Neuroscience at King's College London. Samuel joined King’s College London in 2023. He specialises in neuromodulation, neurodevelopmental disorders, and open and reproducible research practices. His work focuses on providing best practice guidance in translational cognitive neuroscience and psychiatric research.  Samuel earned a degree in BSc Psychology at the University of Birmingham, then went on to achieve a PhD in Cognitive Neuroscience at Aston University, after which he was a post-doc at King’s in the Department of Child & Adolescent Psychiatry. Before joining King’s, Samuel was a lecturer at the University of Westminster and the University of Wolverhampton. OUTLINE: 0:00 - Highlights 2:06 - Intro 3:01 - How Sam got interested in brain stimulation 5:25 - Publishing null results from transcranial direct current stimulation 10:49 - Conducting a systematic review of the field 13:31 - Methodological issues across types of brain stimulation 16:11 - Which brain stimulation/clinical indication combination are most promising? 24:05 - Clinical trial of transcranial direct current stimulation and cognitive training for ADHD in adolescent boys 31:51 - Personalized approaches to brain stimulation and cognitive training 37:23 - Pursuing metascience more broadly 42:34 - Founding the RIOT Science Club 48:07 - Most desired structural changes to science over the next 5-10 years 51:15 - Dealing with scientific fraud, incentives 57:11 - Advice for young scientists 59:25 - Forthcoming developments and where Sam's work can be found online 1:00:58 - Outro

  2. Sep 23

    Registered reports | Chris Chambers

    Chris Chambers is a Professor of Cognitive Neuroscience at Cardiff University who focuses on the promotion and effectiveness of open research practices. He co-founded Registered Reports, Exploratory Reports, Verification Reports, the Transparency and Openness Promotion guidelines, the Royal Society Replications initiative, the UK network of open research working groups, the Peer Reviewers' Openness Initiative, the UK Reproducibility Network (UKRN), the GW4 Undergraduate Psychology Consortium programme for promoting reproducible science, the Royal Society Rapid Review Network for COVID-19 Registered Reports, and the Peer Community in Registered Reports. He currently chairs the Registered Reports committee supported by the Center for Open Science, serve on the UKRN steering committee and sits on the Advisory Board of Nature Human Behaviour. He is the author of the Seven Deadly Sins of Psychology: A Manifesto for Reforming the Culture of Scientific Practice, which won the 2018 British Psychological Society Book Award (Best Academic Monograph) and the 2018 PROSE Award from the Association of American Publishers. OUTLINE: 0:00 - Highlights 2:19 - Intro 3:06 - Starting registered reports 5:44 - Blogosphere conversations that spawned registered reports 7:26 - Pre-registration vs. Registered reports 10:39 - Precursors to registered reports 17:10 - Reviewing the first registered reports at Cortex 21:25 - Initial reactions from neuroscience and psychology to RRs 25:26 - The growth of RRs across journals 27:54 - Peer Community In Registered Reports (Registered Reports 2.0) 33:30 - Stage 1 RR triage/snapshot solution to the wait time problem 36:57 - New challenges with Registered Reports 2.0 39:53 - Programmatic research at Peer Community In Registered Reports 43:26 - Clinical trials and registered reports 47:47 - Current conversations around registered reports 50:39 - Researcher evaluation in a hypothetical world where null data are embraced 56:01 - Forthcoming developments related to registered reports 59:49 - Advice for young scientists 1:03:32 - Outro

  3. Sep 2

    Direct replications, P-curves, and Acknowledging Limitations | Beth Clarke

    Beth Clarke is an incoming Junior Professor of Psychological Metascience at the Leibniz Institute for Psychology (ZPID). Her work examines how research methods and practices have been changing in psychology (especially social and personality psychology) over the replication crisis. Google Scholar: https://scholar.google.com/citations?user=KTF_Du0AAAAJ&hl=en Looking our limitations in the eye: A call for more thorough and honest reporting of study limitations: https://compass.onlinelibrary.wiley.com/doi/10.1111/spc3.12979 The prevalence of direct replication articles in top-ranking psychology journals: https://pubmed.ncbi.nlm.nih.gov/39418470/ P-curving social and personality psychology experiments: Changes from 2006 to 2019: https://osf.io/preprints/psyarxiv/txzeu_v1 RegCheck: https://regcheck.app/ OUTLINE: 0:00 - Highlights2:34 - Intro3:29 - What Beth's professorship in metascience says about the maturity of the field7:10 - Initial interest in metascience10:54 - Hopefulness for the future of psychology, metascience, and science in general13:16 - Study of how researchers ignore the limitations of their work21:20 - Thoughts on registered reports 22:56 - Are certain subfields of psychology better at acknowledging limitations? 24:52 - Should limitations be acknowledged in abstracts, titles, and elsewhere in papers?25:55 - Study of the prevalence of direct replication articles in top psychology journals34:42 - How do you choose what studies to replicate?36:06 - P-curving social and personality psychology experiments45:34 - Developing RegCheck, an AI tool to compare preregistrations and their associated papers50:30 - What systemic interventions would you most like to see in the next 5-10 years that would most benefit the scientific community?54:37 - Directions for Beth's future lab55:51 - Advice for undergraduates59:16 - Outro

  4. Jun 27

    Replication research, diamond open access publishing | Lukas Röseler

    Lukas Röseler is the Managing Director at the Center for Open Science at the University of Münster. He is the Editor-in-Chief at the journal Replication Research, part of the Replication Journal Federation, is an active member of the Framework for Open and Reproducible Research Training (FORRT) community, and developed a video game called Guess the Replication, where players guess whether studies successfully replicated.  Replication Research: http://replicationresearch.org/ Replication Journal Federation: https://forrt.org/rjf/ Framework for Open and Reproducible Research Training (FORRT) community: https://forrt.org/ Guess The Replication: https://lukasroeseler.github.io/GuessTheReplication/ Dominance and Prestige: Meta-Analytic Review of Experimentally Induced Body Position Effects on Behavioral, Self-Report, and Physiological Dependent Variables: https://www.researchgate.net/profile/Robert-Koerner/publication/360577810_Dominance_and_prestige_Meta-analytic_review_of_experimentally_induced_body_position_effects_on_behavioral_self-report_and_physiological_dependent_variables/links/62823b133a23744a728598c2/Dominance-and-Prestige-Meta-Analytic-Review-of-Experimentally-Induced-Body-Position-Effects-on-Behavioral-Self-Report-and-Physiological-Dependent-Variables.pdf OUTLINE: 0:00 - Intro 2:10 - Leading the Münster Center for Open Science 6:17 - How Lukas became interested in metascience 10:53 - Cultural change in science 13:43 - Being Editor-in-Chief of Replication Research 20:00 - Funding the journal, diamond open access 24:55 - Prestigious journals' money extraction scheme 28:17 - Distributed institutional support for diamond OA journals 31:50 - The Replication Journal Federation 39:24 - Journals for replication, or replications for journals? 44:34 - Guess The Replication 50:57 - Prediction markets for replication 54:32 - Meta-analysis of body positioning effects/power posing 1:02:46 - If science were aligned to truth, how would researchers be evaluated? 1:14:36 - Advice for young people

  5. May 27

    Data Provenance in Medical Machine Learning | Alexander Gibson

    Alexander Gibson is a PhD student at the Queensland University of Technology and the Australian Centre for Health Services Innovation studying the intersection of metascience and clinical machine learning. One of his focus areas is data provenance, the Who, What, Where, When, Why, and How of datasets, and how neglecting this can lead to bad outcomes in medical machine learning not only in research, but also for clinical practice and medical device approval.  CONTACT RANDY: Feedback: metasciencematters@gmail.com EPISODE LINKS: Alex's preprint on unreliable diabetes and stroke datasets: https://www.medrxiv.org/content/10.64898/2026.02.24.26347028v2 OUTLINE: 0:00 - Introduction 3:14 - The beginning of Alex's interest in clinical predictive modeling 5:05 - Alex's interest in metascience 6:42 - Choosing a dissertation topic/metrics hacking in machine learning 9:49 - Preprint on data provenance in medical datasets 12:33 - The diabetes and stroke datasets Alex investigated 16:46 - Major irregularities in the data 23:29 - TRIPOD+AI guidelines for auditing machine learning studies 25:26 - How unreliable studies can impact clinical practice and medical device patents 26:42 - Citation networks 27:37 - AI-generated formulaic medical machine learning studies 31:50 - Strategies for high-quality data provenance 33:53 - Patents citing unreliable studies, and how to integrate data provenance into peer review 35:23 - The biggest problems for clinical predictive modeling studies 37:02 - Resources and tools for improving rigor in machine learning 38:45 - Metrics reporting 40:45 - Choosing decision thresholds in predictive models 42:59 - The importance of clinical context in metrics reporting 45:21 - The unreasonable effectiveness of age and sex as predictors 47:53 - The roles of academia and industry in improving clinical machine learning studies 50:07 - Explanation versus prediction 52:51 - Advice and resources for students 54:27 - Outro

  6. Apr 22

    Fraud in Alzheimer's research | Charles Piller

    Charles Piller, an Investigative Correspondent for Science Magazine, is the author of the 2025 book Doctored: Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer's. He has received more than 40 honors or awards from American Association for the Advancement of Science/Kavli Foundation, National Institute of Health Care Management, Society of Professional Journalists, National Academies/Eric and Wendy Schmidt, Los Angeles Press Club, American Society of Tropical Medicine and Hygiene, Gerald Loeb Foundation and UCLA, Western Publications Association, First Amendment Coalition, California State University, Computer Press Association, American Society of Business Press Editors, California Newspaper Publishers Association, National Council on Crime and Delinquency, and other organizations. His work has been nominated for the Pulitzer Prize nine times by STAT, the Los Angeles Times and The Sacramento Bee. CONTACT RANDY: Feedback: metasciencematters@gmail.com EPISODE LINK: Blots on a field?  https://www.science.org/content/article/potential-fabrication-research-images-threatens-key-theory-alzheimers-disease Picture imperfect https://www.science.org/content/article/research-misconduct-finding-neuroscientist-eliezer-masliah-papers-under-suspicion Doctored: Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer's https://www.amazon.com/Doctored-Fraud-Arrogance-Tragedy-Alzheimers/dp/1668031248 OUTLINE: 0:00 - Introduction 6:05 - How Charles became a science journalist 7:58 - Charles' 2022 piece, "Blots on a field?" 14:39 - Responses from scientists to this piece 22:22 - Charles' 2024 piece, "Picture imperfect" 30:29 - How Charles began working on his book "Doctored: Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer's" 33:15 - Charles' experience working with scientific sleuths 36:45 - The incentive for positive data in science 46:09 - Charles' NYT op-ed about how to combat fraud in Alzheimer's research; attacks from the Trump administration on science 53:22 - Charles' thoughts on registered reports and other ways to improve scientific incentives 58:34 - Upcoming work 59:22 - Advice for listeners 1:02:20 - Outro

  7. Apr 1

    Why So Many Cancer Biology Findings Fail to Replicate | Tim Errington

    Tim Errington is the Senior Director of Research at the Center for Open Science. He led the Reproducibility Project: Cancer Biology, Systematizing Confidence in Open Research and Evidence (SCORE), as well as the implementation and evaluation of initiatives such as Registered Reports, Registered Revisions, responsible conduct of research trainings, and open science badges. CONTACT RANDY: Feedback: metasciencematters@gmail.com EPISODE LINKS: Investigating the replicability of preclinical cancer biology (Reproducibility Project: Cancer Biology led by Tim Errington): https://elifesciences.org/articles/71601 Bayer replication study: https://www.nature.com/articles/nrd3439-c1 Amgen replication study: https://www.nature.com/articles/483531a Reproducibility in Cancer Biology: Challenges for assessing replicability in preclinical cancer biology (Companion paper to Reproducibility Project: Cancer Biology): https://elifesciences.org/articles/67995 What is replication?: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3000691 Study comparing standard reports and registered reports in psychology: https://journals.sagepub.com/doi/10.1177/25152459211007467 Blog post on the seemingly magical success of revision experiments: https://rajlaboratory.blogspot.com/2013/06/the-magical-results-of-reviewer.html Google's AI co-scientist paper: https://arxiv.org/abs/2502.18864 Machine-readable documents: https://journals.sagepub.com/doi/10.1177/2515245920970949 How open science helps researchers succeed: https://elifesciences.org/articles/16800 ZBW's Expedition to Open Science Land: https://expedition-open-science.org/ OUTLINE: 0:00 - Introduction 4:47 - Tim's origin story as a cancer biologist 6:38 - Initial interest in metascience 9:24 - Starting the Reproducibility Project: Cancer Biology 12:07 - How were the studies that were replicated chosen? 14:41 - Publishing the Reproducibility Project: Cancer Biology as a registered report 17:26 - Results from the Reproducibility Project: Cancer Biology 20:28 - Tim's experience throughout the years running the Reproducibility Project: Cancer Biology 25:21 - The difficulty of running cancer biology studies 27:54 - Judging whether a replication is successful 31:23 - What has the response to the Reproducibility Project: Cancer Biology been? 37:52 - Why aren't replication rates higher? 40:26 - Challenges of running cancer biology replication studies 45:43 - Caveats of preclinical disease models 49:13 - The incentive for positive data in science 57:05 - Systemic intervention vs. Individual policing 1:01:04 - The value of registered reports 1:07:38 - Registered revisions 1:10:48 - Falsifying theories early at the preclinical stage 1:15:21 - Different institutions (e.g., academic, industry) conducting different studies (e.g., preclinical, clinical) 1:17:34 - New initiatives at the Center for Open Science (Replication project of social/behavioral sciences, automated tools for predicting replication success, LifeCycle journal) 1:23:02 - AI scientists are trained on biased literature; distrust of academic literature in drug discovery 1:28:46 - Peer review 1:32:51 - Narrative in science 1:35:06 - 100-200 years into the future 1:40:29 - Advice for high school/undergraduate listeners 1:42:51 - Metascience manifests in every field 1:44:46 - Philosophy of science 1:47:49 - Outro

  8. Apr 1

    The Replication Crisis Changed Psychology. Did It Fix It? | Brian Nosek

    Brian Nosek is the cofounder and Executive Director of the Center for Open Science. He co-developed the Implicit Association Test, a method that advanced the study of implicit bias. He then co-founded three non-profit organizations: Project Implicit to advance research about implicit bias, the Society for the Improvement of Psychological Science to improve the research culture in his home discipline, and the Center for Open Science (COS) to improve rigor, transparency, integrity, and reproducibility across research disciplines. He led the Reproducibility Project in Psychology, a replication of 100 studies from psychology, as well as multiple other replication projects, along with policy reforms such as open science badges, and the investigation of prediction markets for study replication. CONTACT RANDY: Feedback: metasciencematters@gmail.com EPISODE LINKS: Estimating the reproducibility of psychological science (Reproducibility Project in Psychology led by Brian Nosek): https://www.science.org/doi/10.1126/science.aac4716 False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant: https://journals.sagepub.com/doi/10.1177/0956797611417632 What is replication?: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3000691 Metascience conference: https://metascience.info/ International Conference on the Science of Science and Innovation: https://icssi.org/ International Congress on Peer Review and Scientific Publication: https://peerreviewcongress.org/ MetaROR: https://metaror.org/ OUTLINE: 0:00 - Introduction 2:56 - Brian's origin story as a psychologist 6:35 - Initial interest in metascience  10:57 - "Scientific Utopia," starting the Reproducibility Project in Psychology, and major fraud cases 17:57 - How did the Reproducibility Project in Psychology come together? 23:40 - Results of the Reproducibility Project in Psychology 27:03 - Response from the field, institutional changes 30:39 - Advice for high school/undergrad listeners 34:03 - Outro

See All (13)

About

The research integrity channel for people who want science to work. Metascience Matters covers scientific fraud, replication failures, peer review, research incentives, AI/data reliability, and the people building better science. Hosted by Randy Ellis, a computational biologist and neuroscientist. New interviews, clips, and explainers on how science breaks and how to fix it. Guests, tips, and collaborations: metasciencematters@gmail.com.

Information

  • Creator
    Randy Ellis
  • Years Active
    2025 - 2026
  • Episodes
    13
  • Rating
    Explicit
  • Copyright
    © Randy Ellis
  • Show Website
    Metascience Matters

You Might Also Like