iiinova — where your digital self keeps evolving

Reposted by John Sakaluk

Jennifer Tackett @jnfrltackett.bsky.social · 29/09/2026

Very excited for this review with Michael Hallquist on adolescent personality pathology, a critically understudied topic that has only become moreso in recent years as seen by booming publication rates for depression and anxiety - PD lines at the bottom www.annualreviews.org/content/jour...

1125

Reposted by John Sakaluk

Bill Chopik @chops310.bsky.social · 11/09/2026

I'm giving an overview presentation of what my lab does and I think this meme accurately captures it.

2546

Reposted by John Sakaluk

Vincent Arel-Bundock @vincentab.bsky.social · 04/09/2026

Big news! 🎉 𝚖𝚊𝚛𝚐𝚒𝚗𝚊𝚕𝚎𝚏𝚏𝚎𝚌𝚝𝚜 1.0.0 for #Rstats is out. It’s a big number and it feels like a big step. I wrote a blog on the challenges of interpreting statistical models, SPEED, cool new features, the future, and a 5 year package development and writing odyssey. arelbundock.com/posts/margin...

1642199

Reposted by John Sakaluk

Jenny Bryan @jennybryan.bsky.social · 02/09/2026

👉 "But in this new era, we must delete code even harder, because if we don’t constantly remove complexity, we are going to drown in generated kudzu slop that overtakes codebases." 🌿

04710

Reposted by John Sakaluk

Vicki @vickiboykis.com · 02/09/2026

New post: we need to be deleting more code. vickiboykis.com/2026/09/01/b...

John Sakaluk @johnsakaluk.bsky.social · 04/09/2026

The "proportional responsibility commitment" system you're proposing is, I think, the correct/sensible/least complicated to implement *if the cumulative workload in the system stays sane*. But it's not--we're well past that now, and i don't think we can go back. Could be wrong, but that's my $0.02

020

John Sakaluk @johnsakaluk.bsky.social · 04/09/2026

FWIW: I think the problem is that there are too many (and growing) submissions to journals, so we need a higher layer of dissemination (that enjoys a higher level of acclaim, after experiencing a higher level of scrutiny), and then we accept less/lighter-touch reviewing at the submission level.

120

Reposted by John Sakaluk

Michèle Nuijten @michelenuijten.bsky.social · 04/09/2026

Very proud of my PhD student @cas-goos.bsky.social, whose very hard work on measurement (reporting) quality in psychology has now been published in PLoS One! A finding that stuck with me is that reliability is much more often reported if the recalculated reliability was nice and high 👀

0102

John Sakaluk @johnsakaluk.bsky.social · 01/09/2026

New dependent-effect-size-friendly meta-analysis R package just dropped! Check it out! 👇

150

John Sakaluk @johnsakaluk.bsky.social · 01/09/2026

Congrats! Our group is very excited to dive in with this pkg (along with y'all's recent adjacent papers) !

030

Reposted by John Sakaluk

Josh Grubbs @joshuagrubbsphd.com · 28/08/2026

1/43 Ok folks, after much teasing and anticipation, I'm THRILLED to share our new paper in Annual Review of Clinical Psychology. Not gonna lie: I'm proud of this one, and I hope it makes an impact on the field (so please share!). A thread 🧵👇

To Be a Mental Disorder: Revisiting Behavioral Addictions
Joshua B. Grubbs, * Nicholas C. Borgogna, Beáta Bőthe, Hyoun S. Kim, Rory A. Pfund, Shane W. Kraus

Manuscript to appear in the 2027 edition of Annual Review of Clinical Psychology, this preprint may not
exactly represent the final printed copy.

Grubbs, J.B., Borgogna, N.C., Bőthe, Beáta, Kim, H.S., Pfund, R.A., & Kraus, S.W. (2027) To Be a Mental Disorder: Revisiting Behavioral Addictions. Annual Review of Clinical Psychology. 10.1146/annurev-clinpsy-061225-045608

Behavioral addictions have been recognized in various forms for centuries, yet only three—Gambling Disorder, Gaming Disorder, and Compulsive Sexual Behavior Disorder—are currently included in formal diagnostic systems. In the past decade, research on these conditions has expanded dramatically in both volume and methodological rigor, drawing on longitudinal, neurobiological, and cross-cultural designs and evolving into a truly global science. Even so, the field remains beset by unresolved controversies. Chief among these is a persistent tendency to import substance use disorder concepts—tolerance, withdrawal, and dependence—that lack conceptual and empirical support when applied to behavior. Related challenges concern the over-pathologizing of normal engagement and a reliance on confirmatory, self-report measurement of uncertain validity. We argue that advancing the field requires sustained funding, theory grounded in the disorders themselves, objective and intensive-longitudinal methods, greater diversity, and public health and regulatory attention to increasingly technological environments.

68236

Reposted by John Sakaluk

Kristin Jankowsky @krijanko.bsky.social · 28/08/2026

When you answer "I often don't know who I really am" – which time frame are you thinking of? PD diagnoses require symptoms lasting 2+ years, but common questionnaires specify no time frame at all. We tested whether that matters: 🧵 osf.io/zdbmc with @jzpsych.bsky.social and @uschroeders.bsky.social

osf.io

OSF

1245

Reposted by John Sakaluk

Mark Dingemanse @markdingemanse.net · 28/08/2026

I give my students Dennett's "Higher-order truths about chmess" to habituate them to the idea that not all questions are equal, and there is chaff to be separated from wheat link.springer.com/article/10.1...

Many projects in contemporary philosophy are artifactual puzzles of no abiding significance, but it is treacherously easy for graduate students to be lured into devoting their careers to them, so advice is proffered on how to avoid this trap.

161

Reposted by John Sakaluk

Tom Hardwicke @tomhardwicke.bsky.social · 27/08/2026

Send your best work to Psychological Science! Four special issue calls: 1. Descriptive Research in Psychological Science www.psychologicalscience.org/publications... 2. Advancing Psychological Science Using Large Language Models www.psychologicalscience.org/publications...

12619

John Sakaluk @johnsakaluk.bsky.social · 20/08/2026

Dumb question: how are you using typist to benefit your package? Are these outputs from your packages? Or is typist helping with documentation...? Asking for a friend who thinks typist looks cool but isn't exactly sure why they would use it (yet)

100

Reposted by John Sakaluk

Eytan Adar @eytan.adar.prof · 11/08/2026

I wrote a post about the emergence of AI-native students who do their research using GenAI tools. I think that there are some real risks to job prospects and the likely changes in their fields.

1378

Reposted by John Sakaluk

Steve Kambouris @steve-kambouris.bsky.social · 11/08/2026

Further to this: the news about OSF is the perfect time to recommend @re3data.bsky.social, the Registry of Research Data Repositories - you can search for potential repositories for your data, and filter the results by a vast number of repo features (e.g., does the repo offer versioning or not)

re3data.org

Home | re3data.org

02316

John Sakaluk @johnsakaluk.bsky.social · 11/08/2026

(also thanks: I didn't know Borealis was a thing, and your post motivated me to search and Western's a participating institution)

030

John Sakaluk @johnsakaluk.bsky.social · 11/08/2026

Seems like there's a GH action (github.com/marketplace/...) that could play nice with Borealis for the same kind of workflow that @steve-kambouris.bsky.social describes here: bsky.app/profile/stev...

github.com

Dataverse Uploader Action - GitHub Marketplace

Uploads a repository to Dataverse

130

John Sakaluk @johnsakaluk.bsky.social · 11/08/2026

Indeed. A surprise, in the sense of "whoa, big piece of scientific infrastructure going down!", but not a surprise considering what the price of memory has been doing, post AI-boom, to a variety of markets (e.g., even Apple has had to begrudgingly raise prices)

030

John Sakaluk @johnsakaluk.bsky.social · 11/08/2026

Very cool. Yeah, it would be an unpopular opinion with the masses, but while there would be some initial onboarding/usability challenges to confront, I feel like a lot of scientific systems could be substantially improved with more Git + GitHub usage. But the appetite for that reform would small..

140

Reposted by John Sakaluk

PsyArXivBot @psyarxivbot.bsky.social · 10/08/2026

Machine Psychometrics for Artificial Agents: osf.io/up3ye

032

John Sakaluk @johnsakaluk.bsky.social · 10/08/2026

Which is a shame, because relationship + sexuality science folks seems to have woken up in the last year or so, and realized they can mixture-model the same data they've factor analyzed, and it's given them a whole new lease on life for construct proliferation.

010

John Sakaluk @johnsakaluk.bsky.social · 10/08/2026

My recollection is that @kevinmking.bsky.social has a few. I often (as Reviewer 2) ask authors to defend choice of mixtures vs. Haslam et al.'s taxometric meta-analysis: www.cambridge.org/core/service... I think there are arguments to be made, in select cases, but most don't/can't make them.

cambridge.org

270

John Sakaluk @johnsakaluk.bsky.social · 10/08/2026

If you do #dyadic data analysis of any kind, consider filling out our survey; we'd love to learn what y'all are doing with your dyadic data!

076

Reposted by John Sakaluk

McKell Jorgensen-Wells @mjorgwells.bsky.social · 31/07/2026

I'm starting my dissertation research on the use of / barriers to dyadic data analysis in relationships research! If you've used dyadic data in your research, I'd love for you to take my and @johnsakaluk.bsky.social‘s survey: uwo.eu.qualtrics.com/jfe/form/SV_.... Thank you!

045

Reposted by John Sakaluk

Erin Reed @erininthemorning.com · 04/08/2026

Days after a WNBA team co-ownwer was suspended for confronting anti-trans activists courtside, Cheryl Reeve of the Minnesota Lynx wore a "Trans Kids Belong" shirt while facing down Sophie Cunningham's team, the Fever. The Lynx won 108-100. Subscribe to support our journalism.

erininthemorning.com

WNBA Coach Cheryl Reeve Wore “Trans Kids Belong” While Beating Sophie Cunningham’s Team

The crowd was equally indignant, reportedly booing whenever Cunningham touched the ball.

91309197

Reposted by John Sakaluk

Eric Hehman @erichehman.bsky.social · 30/07/2026

McGill social psych is now officially hiring at the assistant prof level, hooray Happy to chat with anyone about the position mcgill.wd3.myworkdayjobs.com/McGill_Caree...

05347

Reposted by John Sakaluk

Aki Vehtari @avehtari.bsky.social · 29/07/2026

priorsense R package has now JOSS paper "priorsense: Efficient prior and likelihood sensitivity checks for Bayesian models in R" you can cite, too doi.org/10.21105/jos... with Noa Kallioinen, Topi Paananen, and @paulbuerkner.com priorsense achieved also a gold badge from @ropensci.org review!

doi.org

priorsense: Efficient prior and likelihood sensitivity checks for Bayesian models in R

Kallioinen et al., (2026). priorsense: Efficient prior and likelihood sensitivity checks for Bayesian models in R. Journal of Open Source Software, 11(123), 11036, https://doi.org/10.21105/joss.11036

16816

John Sakaluk @johnsakaluk.bsky.social · 17/06/2026

Apropos of nothing, anyone interested in joining a "software* in social/personality psychology"-themed SPSP submission? *could be R/python packages, dashboards, web-based or mobile apps, etc.,

186

John Sakaluk @johnsakaluk.bsky.social · 12/06/2026

Social/Personality Psych people: what are some of your favourite research software products/papers that have come out in the last ~5 years?

133

John Sakaluk @johnsakaluk.bsky.social · 01/06/2026

People can copy/refactor effortlessly with LLMs, but "taste", and vision for match between user base and functionality are things that still matter, and can (but not necessarily will) win competition for long-term user allegiance

010

John Sakaluk @johnsakaluk.bsky.social · 01/06/2026

Totally. I had a kind of "come to Jesus" realization about this risk awhile ago. I think the problem you're asking after requires every developer to ask: are you prepared to out-develop a competitor on the merits (and let the market decide)?" "Yes" is not the right answer for every project.

110

John Sakaluk @johnsakaluk.bsky.social · 01/06/2026

I totally see the sensibility in that kind of approach, with the caveat, that I think "I deserve to write/get a paper out of this code, and people should cite it!" should be reserved more for "formal package" level commitment and less for "I slapped some CC-0 code up online" one-offs.

120

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

Bottom line: we need a higher bar (and stronger incentives and supports) for quality development and maintenance of open-source research software.

170

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

Meanwhile, we need to figure out a better merger of os software development and publishing. This issue is not a one-off; most of the new functionality I've tried (each with fancy papers) in the last year is DOA or breaks soon upon arrival.

161

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

Because at the end of the day, users don't have time/energy to fuss with broken software. So to aspiring devs out there: if you see someone developing in an area of functionality you're passionate about, don't cede the turf. Compete on good design and usability, and may the best software win!

120

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

With open-source development, in contrast, a first-mover might be able to stymie someone publishing for awhile, but they can't stop them from developing and distributing an alternative. And if their alternative is better, and the functionality is needed, eventually the better product will win out.

130

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

If you're first to publish on a popular idea, you often assume the role of gatekeeper, regardless of the quality of your contribution, and regardless of your technical chops for evaluating the robustness of others' work. It allows first-movers to exert a wild amount of control over new work.

131

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

There's a "permissionless" nature to open-source development that (with appropriate license use/compliance, of course), makes competition of between scientific products feel different and healthier than how different scientists compete over ideas in the published literature.

120

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

Eventually, though, through this realization, my frustration gave way to (precarious) optimism: the fact of the matter is that if this package is fragile and its design sucks (and I think it is, and it does), someone can, and perhaps will, come along and make/promote a better version of it.

140

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

The journal often wants to publish these sorts of papers because methods/tooling papers can become citation gold mines. But once published in the current system, what's the structural incentive for the original developer to maintain and/or improve the quality of their software? There is none.

160

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

To name a few reasons: -reviewers are already too hard to come by -reviewers with software development chops are rare -most non-developer reviewers wouldn't know what makes software design good/bad -many general reviewers can feel statistics imposter syndrome (on top the burden of reviewing code)

1100

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

How could a respectable methods journal put its stamp of approval on such a flimsy piece of software? But thinking about it more, and talking it through with some colleagues, in the context of the current peer-review system, this makes a lot of sense. Why?

130

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

At first, my reaction was anger. Having had a paper recently rejected (from a different journal) detailing a package that (immodestly) we had worked hard to make robust and well-documented--and it seemed like no reviewer had actually looked at our codebase--made this situation seem outrageous.

140

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

...argument design that was restrictive/unintuitive, bad documentation, and for all intents and purposes only one substantive .R file where each of the ~10-12 functions--monolithic monsters that try to do a bajillion things each--were crammed into an 8000+ line script file.

130

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

devtools::check() produced errors, warnings, and notes galore. Not that CRAN standards are be-all/end-all--packages that live elsewhere can totally be useful and robust--but all signs pointed to this being very breakable. No unit tests, well-known error-prone NAMESPACE practices...

170

John Sakaluk @johnsakaluk.bsky.social · 28/05/2026

A small 🧵 about frustrations and opportunities w/ open-source research software for the social sciences: Came across a new paper in a top methods journal, explaining/demo'ing a new R package. Sounded v. cool, so was excited to dig in. What I found was largely a mess...

43210