The Dead Salmons of AI Interpretability arxiv.org/abs/2512.18792 The famous #fMRI cautionary tale re-born as an analogy for the AI age! @practicalfmri.bsky.social
Max Korbmacher
@maxkorbmacher.bsky.social
Imaging neuroscientist. Brain (bio)markers, big data, MRI, meta & open/better science. Views are my own.
If a transgender patient visits a psychiatrist in a hospital, is it fair to say that they have «severe psychiatric morbidity»? We believe research should be more rigorous. With 49 co-authors, including: @maxkorbmacher.bsky.social @erininthemorning.com @natacha.bsky.social @frende.bsky.social
David R. Banos@davidrbanos.bsky.social · 2mo agoOur letter of concern about the Finnish study is now published in Acta Paediatrica. The Finnish study has received a total of 5 letters/commentaries, so far. Most of the overlapping concerns are statistical/methodological. onlinelibrary.wiley.com/doi/10.1111/...
Nice example of how research data can be validated against real world clinical data. doi.org/10.1016/j.co...
Redirecting
doi.org
We just launched #GuesstheReplication (GutheR). It is a challenging game where you predict replicability based on the abstract and title of an original study. Play it now and beat my high score at lukasroeseler.github.io/GuessTheRepl... or use our free slides for teaching zenodo.org/records/1985...
Cognitive function at age 13 predicts dementia 60 years later, even controlling for cardiovascular and cardiometabolic risk. @KWalhovd www.medrxiv.org/content/10.6...
medrxiv.org
Fun facts about biological differences in sleep duration. But the title is misleading. Not about ageing but cross-sectional correlation. doi.org/10.1038/s415...
Sleep chart of biological ageing clocks in middle and late life - Nature
A cross-organ, multi-omics U-shaped relationship between sleep duration and biological ageing clocks highlights the potential of sleep optimization to promote healthy ageing, lower disease risk and ex...
doi.org
Interested in neuroimaging and cognition in aging and want to work with us? We have an open position @KWalhovd @LCBC_UiO 2411.webcruiter.no/Main2/Recrui...
Researcher in neuroimaging and cognitive change of aging
We have a 3-year researcher position for analysing longitudinal brain MRI data in relation to cognitive- and other biomarker data.The shift to a biomarker-based definition of Alzheimer’s disease (AD) ...
2411.webcruiter.no
pubmed.ncbi.nlm.nih.gov/42083734/htt...
Spatial and Volumetric Characteristics of Glioblastoma: Associations With Clinical Presentation and Survival - PubMed
Anatomical tumor distribution provided prognostic information beyond overall tumor burden in glioblastoma. Regional patterns expand on the anatomical basis of symptoms. The survival benefit associated...
pubmed.ncbi.nlm.nih.gov
Use of brain age models frequently lead to wrong conclusions - unfortunately it is not easy to fix. Fortunately, there are better alternatives. @edvardg.bsky.social @fmrib-steve.bsky.social @didacvp.bsky.social @maxwellelliott.bsky.social @k.b.walhovd www.medrxiv.org/content/10.6...
Reconsidering Brain Age: Why Age-Prediction Models Fail as Measures of Brain Aging
Brain age models - machine-learning predictions of chronological age from brain imaging - are widely interpreted as markers of accelerated brain aging. Here we show that this interpretation cannot be ...
medrxiv.org
🏓😊 Dürr, P., El Gheche, M., Maeda, G.J. et al. Outplaying elite table tennis players with an autonomous robot. Nature 652, 886–891 (2026). doi.org/10.1038/s415...
Outplaying elite table tennis players with an autonomous robot - Nature
An autonomous robot system, Ace, combines event-based vision and reinforcement learning to compete with elite human table tennis players, highlighting the potential of physical AI agents to perform co...
doi.org
What can we learn from 1,000 meta-analyses across 10 different disciplines? doi.org/10.1017/rsm....
What can we learn from 1,000 meta-analyses across 10 different disciplines? | Research Synthesis Methods | Cambridge Core
What can we learn from 1,000 meta-analyses across 10 different disciplines? - Volume 17 Issue 1
doi.org
Paradigm shifts are BS. "We find that three key measures of scientific progress—major discoveries, methods and fields—each demonstrate that science evolves cumulatively." doi.org/10.1098/rspa...
Debunking revolutionary paradigm shifts: evidence of cumulative scientific progress across science
Abstract. How can scientific progress be conceived best? Does science mainly undergo revolutionary paradigm shifts? Or is the evolution of science mainly c
doi.org
Nature recently published massive studies on the credibility of social science. We — the Norwegian co-authors on the robustness-paper, Gerit Pfuhl, @oysteinhernaes.bsky.social, @maxkorbmacher.bsky.social, and me — wrote an opinion piece about it (in Norwegian): www.forskersonen.no/forskning-kr...
Hva kjennetegner god vitenskap?
KRONIKK: Troverdig forskning er ikke forskning uten usikkerhet. Det er forskning som viser oss usikkerheten i stedet for å skjule den.
forskersonen.no
Brian Nosek@briannosek.bsky.social · 4mo agoSCORE, a collaboration of 865 researchers, is now released as three papers in Nature, six preprints, and a lot of data (cos.io/score/). SCORE examined repeatability of findings from the social-behavioral sciences and tested whether human and automated methods could predict replicability.
New paper alert 🤓🚨 Korbmacher, M., Boen, R., Andreassen, O.A. et al. White matter microstructure differences between 15q11.2 copy number variation carriers and non-carriers in mid-to-late life. Transl Psychiatry (2026). doi.org/10.1038/s413...
White matter microstructure differences between 15q11.2 copy number variation carriers and non-carriers in mid-to-late life - Translational Psychiatry
Translational Psychiatry - White matter microstructure differences between 15q11.2 copy number variation carriers and non-carriers in mid-to-late life
doi.org
Ren, J., Zhang, W., Dahmani, L. et al. Parkinson’s disease as a somato-cognitive action network disorder. Nature (2026). doi.org/10.1038/s415...
Parkinson’s disease as a somato-cognitive action network disorder - Nature
The substantia nigra and all Parkinson’s disease deep-brain stimulation targets are selectively connected to the somato-cognitive action network rather than to effector-specific motor regions.
doi.org
New paper in Imaging Neuroscience by James D. Kent, Alejandro de la Vega, et al: Neurosynth Compose: A web-based platform for flexible and reproducible neuroimaging meta-analysis doi.org/10.1162/IMAG...
Would be great to make migraine meds obsolete for rTMS responders. Q is whether the classification into responders can be done with cheaper means than fMRI... link.springer.com/article/10.1...
Prospective open-label trial of personalised connectivity-guided transcranial magnetic stimulation therapy for migraine - The Journal of Headache and Pain
The Journal of Headache and Pain - Migraine is a leading cause of disability worldwide, affecting hundreds of millions of individuals. Repetitive transcranial magnetic stimulation (rTMS) offers a...
link.springer.com
Multimodal meta-analysis of brain integrity in disorders of consciousness A Sala, M Meys, N Alnagger, N Beliy, S Abagnale… - 2026 scholar.google.com/scholar_url?...
scholar.google.com
New post! There was a lot of innovation in medicine and biomedical research this year, and I've tried to summarize the biggest ones in this blogpost. Medical breakthroughs in 2025. Plus a serious note at the end. www.scientificdiscovery.dev/p/medical-br...
Medical breakthroughs in 2025
... and a happy new year.
scientificdiscovery.dev
Sex‐Modified Effects of APOE‐ε4 on Spatial Patterns of Brain Atrophy: A Multi‐cohort Study Zhang et al doi.org/10.1002/alz7...
doi.org
Seems like gold for the Computational neuroscientists and data farmers among us: BrainScape: An open-source framework for integrating and preprocessing anatomical MRI datasets Muhammad Nabi Yasinzai, Remika Mito, Mangor Pedersen doi.org/10.1162/IMAG...
BrainScape: An open-source framework for integrating and preprocessing anatomical MRI datasets
Abstract. MRI has revolutionized our ability to investigate and understand brain structure and function in health and disease. A large amount of MRI data is widely available to researchers, both from ...
doi.org
Are tele-neuropsychology and in-person assessment scores meaningfully different? A systematic review and meta-analysis Jessica I. Alva et al doi.org/10.1080/1385...
Are tele-neuropsychology and in-person assessment scores meaningfully different? A systematic review and meta-analysis
Objectives: Despite growing evidence supporting tele-neuropsychology (teleNP), clinicians have voiced concerns about comparability to traditional in-person testing and the limited availability of t...
doi.org
"Grassroots networks can help implement and harmonize open research efforts", by @agataboch.bsky.social, @tkalan.bsky.social, @maxkorbmacher.bsky.social, Lewend Mayiwar, @julienmayor.bsky.social & @dsquintana.bsky.social - doi.org/10.7557/11.8....
We're delighted to welcome TWO new diamond-OA PCI RR-friendly journals Registered Reports in Linguistics, ed. by @scoretta.bsky.social ky.social & @jess-hampton.bsky.social + Replication Research @r2journal.bsky.social ed. by @aufdroeseler.bsky.social & team rr.peercommunityin.org/about/pci_rr...
Hello world! Follow this account to receive announcements, invitations for open peer review, and publications from Replication Research.
Brain-age models with lower age prediction accuracy have higher sensitivity for disease detection Marc-Andre Schulz, Nys Tjade Siegel, Kerstin Ritter journals.plos.org/plosbiology/...
Brain-age models with lower age prediction accuracy have higher sensitivity for disease detection
There is increasing interest in developing imaging-based models to characterize 'brain-age', but how accurate are these for predicting neurological disease? This study shows that simpler models, which...
journals.plos.org
Is the “standard workflow” holding back fMRI analysis? Mass-univariate analysis is still the bread-and-butter: intuitive, fast… and chronically overfitted. Add harsh multiple-comparison penalties, and we patch the workflow with statistical band-aids. No wonder the stringency debates never die.
Nice one, from @drg.bsky.social and @jamesheathers.bsky.social link.springer.com/article/10.1...
“Cake causes herpes?” - promiscuous dichotomisation induces false positives - BMC Medical Research Methodology
Background Continuous biomedical data is often dichotomized into two or more groups for analysis, despite long-standing warnings from statisticians that this constitutes bad practice. This dichotomisa...
link.springer.com
Smeland, O.B., Kutrolli, G., Bahrami, S. et al. A genome-wide analysis of the shared genetic risk architecture of complex neurological and psychiatric disorders. Nat Neurosci (2025). doi.org/10.1038/s415...
A genome-wide analysis of the shared genetic risk architecture of complex neurological and psychiatric disorders - Nature Neuroscience
Smeland et al. demonstrate greater genetic overlap between neurological and psychiatric disorders than previously recognized, along with diverse neurobiological associations. The findings support a mo...
doi.org