Trust in science. Expert opinion, context, and open data all help, and in the era of AI, we should aspire to audit trails that confirm data provenance. Key to trust is accountability of researchers, however, and for that we need identity verification. New blog post. openrxiv.org/trust-signal...

What do you think is your most under-recognised paper and why should people be paying more attention? Go on, take the chance to toot your own horn 🎺

New book haul 📚

As long as a paper's methods are described *somewhere* I don't think it matters where they are located in a manuscript

Many people write grant applications assuming their reader is engaged, but the opposite is more likely. Your evaluators are time- and attention-poor, so you should design your application with this in mind. This slide is part of a talk I'm giving today on writing grant applications

Most autistic students who need accommodations in higher education never get them. In our latest preprint, we describe where the gaps are and what students said about them doi.org/q8js

In our new preprint, we describe the unique needs and experiences of autistic students in higher education osf.io/preprints/ps...

It’s paddle season

Timeline cleanse

Nothing says “emergency” quite like a 2x3x3 interaction effect (p = 0.043)

Social media platforms and how academics use them change, but one thing doesn't

did… did a journal write this?

We have a tradition in our lab: every PhD or Master’s publication gets its own custom mug ☕

Procrastination, 2026 style

A major release of the RoBMA package (4.0.0) is now on CRAN. Main highlights: - fully redesigned interface closely matching metafor - dedicated brma, bselmodel, bPET, bPEESE for single model fits - fit diagnostics features (residuals, influence, etc) (overview at fbartos.github.io/RoBMA/articl...)

Here’s a GIF I put together illustrating the advantage of an allostatic system vs. a homeostatic system for an upcoming talk 🧵

Effect size distributions (ESDs) can help determine small/medium/large effects, but publication bias may lead to imprecision. Iceberg plots visualize a raw ESD against a pub bias-adjusted ESD. Build this using data typically reported in published meta-analyses via our new ‘ESDist’ R package 📦

A meta-analysis is not a ‘easy’ publication. Search strategy development, screening, more screening, data extraction, requesting additional data, risk of bias assessments, the actual synthesis, write up—it all takes time. Many people heavily underestimate the resources required to complete one

Bring back printed course readers! 📗

New preprint: Preregistration is supposed to make research more transparent. But if registry records aren't kept up to date, the transparency benefit erodes. We checked 1,137 psych meta-analyses on PROSPERO and found the records are systematically out of sync with reality osf.io/preprints/ps...

Just want to check again: Has anyone with a PROSPERO systematic review registration received an email reminder to update their registration? The documentation says this is the case, but I've never received anything like this, and neither have a couple of others who responded to this original post

Really happy this paper is finally out. This began as an R script I wrote for a paper in 2016, and the Ben (first author) took this idea and improved it, iteration after iteration, into an R package.

Selecting an effect size for power analysis is hard. Many researchers fall back on Cohen's thresholds, but they have no empirical basis and vary wildly by field. Our new paper offers a better option: field-specific effect size distributions built from meta-analytic data doi.org/10.3758/s134...

My article "Three-Sided Testing to Establish Practical Significance" with @jackfitzgerald.bsky.social is now published in AMPPS! journals.sagepub.com/doi/10.1177/... Three-sided testing is an improved version of TOST that lets you test for equivalence, superiority and inferiority simultaneously.

My superpower is accepting peer review requests weeks apart whose deadlines all somehow fall within the same couple of days

For those with healthy enough boundaries to avoid bluesky over the weekend, here's a thread on my new preprint formalising the Allostatic Theory of Oxytocin (and what it taught me about theory-building in psychology)

Why are oxytocin study results so inconsistent? Methods get most of the blame, but the problem is also theoretical as existing accounts are too vague to falsify. In a new preprint, I formalise the Allostatic Theory of Oxytocin as a computational model osf.io/preprints/ps... 🧵

I always imagined that papers would become quicker to write as I got more experienced with research, but the opposite has happened instead because now I better understand how papers can be improved and spend much more time tweaking and rewriting

According to PROSPERO's documentation, automated emails are sent to reminder authors to update their registrations when their review is completed and to include citation details when it is published www.crd.york.ac.uk/PROSPERO/hel... Has anyone with a registration gotten this automated email?

PDF-Direct Firefox extension ('Bypass "fancy online reader" and just get the darn pdf') updated: addons.mozilla.org/en-US/firefo... & now supports ACM ACS Cambridge University Press Elsevier IEEE JSTOR Nature OUP PLOS PNAS Royal Society Publishing Sage Science Springer Nature Taylor & Francis Wiley

Computationally formalising a verbal theory is an eye-opening experience because it forces you to openly state all the assumptions that the theory rest on

New preprint! Led by @aisartorius.bsky.social, we ran a meta-analysis on the effect of intranasal oxytocin on appetite + caloric intake. We found oxytocin reduces appetite in clinical populations, but data are too underpowered to draw conclusions in healthy populations doi.org/10.64898/202... 🧵

New preprint! Led by @aisartorius.bsky.social, we ran a meta-analysis on the effect of intranasal oxytocin on appetite + caloric intake. We found oxytocin reduces appetite in clinical populations, but data are too underpowered to draw conclusions in healthy populations doi.org/10.64898/202... 🧵

My first thought was “why create fake authors if you wouldn’t get the credit?”. But then it clicked—this is probably a citation boosting play

It's been a few years since I've tried writing a manuscript in Rstudio using markdown (now quarto). It seems A LOT better now but I guess the collaboration issue still remains? We have this same problem with Overleaf, some collaborators simply prefer Word

29 years of #rstats community knowledge was sitting in hard-to-search pipermail archives. So I built a more modern home for it. Introducing the R Mailing List Archives: 631,000+ messages from 32 lists, fully searchable and available as open data. r-mailing-lists.thecoatlessprofessor.com

Nice to wake up to the news that our paper on the ESDist R package for calculating effect size distributions has been accepted for publication 🎉 Since our initial 2023 (!?) preprint we’ve made lots of improvements to the package based on feedback

Here are some options for hosting a living systematic review project A) A Quatro/Shiny website (we have code for this that can be adapted: github.com/iaiversen/AM...) B) OSF/Zenodo (easier but much less flexibility) C) A handful of journals support living systematic reviews

Conventional systematic reviews can become outdated before they're even published. In our new @nathumbehav.nature.com piece, we discuss how living systematic reviews can help keep evidence current by continuously integrating new evidence. www.nature.com/articles/s41...

Most oxytocin genetic studies investigate the link between oxytocin receptor gene variants + phenotypes of interest (usually in small samples). Here we took a different approach by investigating the genetics of the brain structures that *produce* oxytocin, in a sample of over 39k people

I'm peer reviewing a poorly written review right now, but it least I know it wasn't LLM generated. That is, unless the authors are playing 4D chess and included prompts to make the writing inconsistent. Who knows anymore? 🤷‍♂️

The actual audience when you deliver a talk 🙂🙂😗😊🧐🤓🙂 🙂🤔🥱🤔🙂🤨🙂 ☺️🙂😊🙂☺️😌🧐 What you see when you deliver a talk 🤨🥱🥱🤨🤨🥱😑 🥱🤨🤨🥱😑😐😐 🤨🥱🥱🤨🤨🥱😑

In our new Molecular Psychiatry paper, we demonstrate genetic overlap between oxytocin-vasotocin hypothalamus subunit architecture and several psychiatric + metabolic traits 🧠 Fantastic work from @aisartorius.bsky.social + team rdcu.be/e7THc

I’m reminded of this comic almost every time I check LinkedIn

Oxytocin is typically described as a "social" hormone. In our new article, we propose that it should instead be viewed a hormone that modulates behavioral flexibility doi.org/10.1016/j.ne...

How have I not heard if this Norah Jones cover of Black Hole Sun before? youtu.be/XbQ08Ixczvo?...

Ok pals. Looking to move all my teaching materials open and online, and am developing new modules that I want to start off this way. Content = reading, quizzes, videos, and code tutorials. Opinions on the best platform? Good examples of best practice I can steal from? Many thanks.

One really cool mentimeter feature is embedding within powerpoint, but due to institutional security policies this no longer works. A good solution to this problem I figured out that avoids switching between app windows is to run the slides in one browser tab and the mentimeter in another tab

New post! It may seem ambitious to ask for individual patient data from clinical trials to be shared, anonymized, for use by other researchers. But the history of medicine shows us that clinical trials have already undergone a series of transformations that once seemed equally bold:

Whenever @psyarxivbot.bsky.social posts a preprint title in all caps it reads like the title of an 'At the Drive In' song: SENSORY BASED MODULATIONS ON VISUAL MEMORY WORKING MEMORY CLASSROOM SCREENER PSYCHOMETRIC PROPERTIES IN AMBULATORY ASSESSMENT