your feedback was kind gentle and correct and still i resent you

New blog post: Evaluating Dr. Cuddy’s Claim that the Debunking of Power Posing is a Myth. daniellakens.blogspot.com/2026/05/eval... On an AI generated description of a non-existent study, incorrectly citing findings from studies, and the importance of scientific criticism.

Small study shows surprisingly encouraging results of a critical thing college class in reducing four biases psycnet.apa.org/record/2027-...

Given the prevalence of LLM-generated citations, it's time we generated Active Citations. These citations give the source excerpt for each citation. Paper: www.princeton.edu/~amoravcs/li... Example: amoravcs.scholar.princeton.edu/sites/g/file...

I’ll be in Oslo May 26-30 and Trondheim June 1-4. Anything I should do or know? Where are my Norwegian peeps?

I'll be in Boston Sunday night for a spell. Boston peep, where are you?

AI is supposedly sycophantic, but Claude Opus 4.7 routinely reminds me to use its advice when coding. I sometimes can't get it to budge and accept my view. Maybe consider the latest models before labeling chatbots are sycophants.

I am once again begging, praying, weeping for lists of written instructions rather than videos. I do not want to watch your two minute video in the hope of catching the two seconds where you whizz past what goddamn menu the thing is under. Write. It. Down.

I have so much grading to do that my home is spotless.

Given the prevalence of LLM-generated citations, it's time we generated Active Citations. These citations give the source excerpt for each citation. Paper: www.princeton.edu/~amoravcs/li... Example: amoravcs.scholar.princeton.edu/sites/g/file...

Pro tip: boost and engage (intelligently) with science posts. Some scientists are feeling lonely.

When reviewing manuscripts/papers, have you ever made a visual or diagram the sole focus of your feedback? I keep wondering that the core representation of reviewing can do for surfacing issues and persuading authors. What would it look like if peer review tech were primarily visual? #peerreview

"Reviewers... are going to become less tolerant of these “obvious” big mistakes now that they can be identified by a variety of cheap tools [GenAI review apps] prior to publication... The least we can do is ask them to review the best possible version..." via @quentinandre.bsky.social

#ATProtoScience tools can enable sensemaking and maybe some nerdy jobs in the near future: "Similar to how social media made influencers... maybe there'll be influencers [as scholarly] sensemakers... [attendee: high need for cognition people]..." Yep, that's my dream job, @ronentk.me

Logic mode: Guns don't kill people. People kill people.

This proposes a peer review credit system. I like it. But review credits must not only be generated by journals. To break the journal monopoly on peer review, other venues - such as journal-agnostic review, and post-publication review - should also draw from the pool. www.pnas.org/doi/10.1073/...

Hackers have compromised Canvas, the learninge management system used to teache college courses. The hackeres have made the system slow, inefficient, unresponsyve, and yn manye cases ympossible to use. Studentes and Professors so far have not noticed any difference.

Loving the use of automated/AI-assisted feedback *before* submission. We have a shortage of these tools.

I've built a new tool! You can upload your pre-analysis plan or registered report, pre-submission to a registry or journal, and it will screen it for completeness, clarity, and consistency. 1/ 🧵

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...

How much does it cost to buy your way onto a scientific paper? To answer this question (and better understand the global market for academic reputation manipulation), we compiled BuyTheBy, a massive dataset of thousands of advertisements from paper mills in seven different countries. Link below!

Unexpectedly, I found someone who loves alliteration more than me.

Never let a crisis go to waste.

#SciComment The state of sleuthing, in a nutshell 🥜

C'mon, now! Let's get a Community Notes type feature for Bluesky out already. 🪧

1. Steve Jobs realized that "People hate their phones". Apple invented a better one, made bank, & changed the world. Academics hate their scientific publishers. Bluesky invented a better publishing platform, & could bank some of the ~$20B/year sci pub revenue, changing science 🧪 @toni.bsky.team 🧵

Vibecoding this after two years: A Zotero-to-PubPeer integration (alpha, shipping in a few days). Suppose you want to share your comments about specific parts of a report. Just follow five simple steps: 1. Highlight and annotate in Zotero 1/n 🧵

Not sure how or whether to suspect AI, but the @zotero.org team is cooking with version 9 just a few months after version 8. www.zotero.org/blog/zotero-9/

Who are some of your favorite, lesser-known sleuths? 🔦 I'm looking for blogs and social media accounts to find patterns in critiques (strategies, communication frames, collaboration practices). Any rising stars? Anyone who isn't in the spotlight? 📸 #metascience #openscience

Ever wanted to share an important *connection* between two pieces? Like a thread explainer for a complex paper, or a blog debunking some hyped up claim on social media?

If only there were, like, peers who could review those journal articles for payment. Chawla, D. S. (2026). Scientific Journals Need Dedicated Fact-Checkers. Undark Magazine. Retrieved from undark.org/2026/04/09/o...

just a lil existentialist chart!

Eventually benchmarks will need to be standardized to next-gen models like making IQ tests harder over time to account for the Flynn Effect.

Webinar tomorrow: Retracted research doesn’t simply disappear once it is withdrawn from the scientific record. In many cases, it continues to circulate across news outlets, social media platforms and public conversations 1/3

You'd think astronauts would use Linux...

Welp! Education (my old field) is not very reproducible: I saw it coming from a mile away (reason I left), but still... www.nature.com/articles/d41...

The only real AI benchmark I need is one focused on correctly generating citations and references with just a link or document provided.

I'm thinking that the true pre-registration should be on the reviewer side. Preregister your claims about some result being "obvious" before reading the Results section or the end of the Abstract.

Sigh...😩 I'm considering not reviewing manuscripts that don't use a reporting guideline for work that could have benefited from it. Don't make reviewers make such basic comments as: - report standard deviations - link to materials and data - visualize the study design - justify the sample size

Scientist friends: When you're writing a paper, have you ever chosen not to cite someone's work because that person is a jerk? Obviously you can't avoid citing vital work, but there are lots of times when you could choose any of several papers to cite. What factors into your decision then? 🧪

Regular reminder that RELX -- owner of academic publisher Elsevier -- is one of those brokers providing data to ICE.

If you like keeping up the latest research tools and want to plug into a community developing scholar-centered software, stream ATProto Science 2026 talks this Friday.

*IT* finally happened. I requested that one of my papers be cited in a peer review I authored. I had just cause, OK?

The "Theory Paper Format" is a template for writing psychological theory papers, developed by the Taskforce Theoretical Psychology (Lorentz Workshop, 2025). Use it to structure your theoretical paper, or (for editors) consider creating journal space for theory papers doi.org/10.17605/OSF...

Sleuthing can be confrontational and grim at times, but once in a while a positive interaction and outcome surfaces. In this PubPeer thread, Dorothy Bishop finds stronger results for an effect in a viral paper and helps authors make a correction (pubpeer.com/publications...).

psycnet.apa.org/record/2026-... replicationindex.com/2025/04/27/r... So many supposed "pillars" of social psychology have bitten the dust over the last decade. Why is it that there seems to be so little anger and rage at the underlying incompetence and dishonesty?

Research sleuths are starting to organize, but the field, and the people, remain idiosyncratic — my latest for @cenmag.bsky.social: cen.acs.org/research-int... @mcintold.bsky.social, @reeserichardson.bsky.social, @elisabethbik.bsky.social, @sholtodavid.bsky.social, @eugenie-reich.bsky.social

Methodologists' theoretical papers: "you MUST do X or your estimates are meaningless." Methodologists' empirical papers: [does not do X]

I think I might have sprung forward a little too much.

Last week, I mentioned this in passing in a workshop: In 1938 Enrico Fermi won a Nobel Prize for discovering two new elements of the periodic table. Lise Meitner shortly showed that Fermi was mistaken and instead had produced known lighter elements by fission. She did not win a Nobel prize.