iiinova — where your digital self keeps evolving

Nadja Rupprechter @naru96.bsky.social · 12/10/2026

Yay! I try to have fun with it! Not every post will be a banger, but every post one might sit on is a missed chance to make someone laugh or feel seen 🙂‍↕️

000

Reposted by Nadja Rupprechter

Julia M. Rohrer @dingdingpeng.the100.ci · 12/10/2026

Causal peeps: Somebody asked me whether I had readings to recommend on the role of causality in psychological theories -- for example, are all theories inherently causal or otherwise. Any ideas/pointers? They already found work by @mieronen.bsky.social and @bringmannlaura.bsky.social 👀

1169

Reposted by Nadja Rupprechter

Julia M. Rohrer @dingdingpeng.the100.ci · 18/09/2026

Cluster analysis can be applied to these, and clusters will be found. But that doesn’t mean those clusters are meaningful.>

man to computer: Find clusters
computer: >CLUSTERS FOUND
man: oh my god.

4769

Reposted by Nadja Rupprechter

Julia M. Rohrer @dingdingpeng.the100.ci · 04/10/2026

Researcher insisting that attributing Y to X is not meant to be a causal statement:

Elizabeth Holmes: why would I deceive you? There’s no reason for me to do that

0758

Nadja Rupprechter @naru96.bsky.social · 08/10/2026

Inducing 2012 nostalgia by having this song on repeat for 4 weeks already. Depression is cured!

000

Reposted by Nadja Rupprechter

Peter Friese @peterfriese.dev · 08/10/2026

Coding agents often waste reasoning tokens on routine routing decisions. Sai Rahul explains how to pair Codex with Jev, a fast decision model, to handle judgement calls in 50-300ms at minimal cost. go.peterfriese.dev/ai-agentic-...

x.com

Rahul (@sairahul1) on X

How To Code Almost Forever for $20/Month With Codex + GPT-6 Luna

261

Reposted by Nadja Rupprechter

Wyatt Walls @wwalls.bsky.social · 07/10/2026

I'm pivoting to something LLMs can't do: day-drinking

33911

Nadja Rupprechter @naru96.bsky.social · 08/10/2026

Very relevant point you bring up. I am curious to see how many conversations in the GPT pipeline are companion-oriented (and vice versa on c.AI, as there are more fun/casual role-play oriented Chatbots on there too). Also already curious if an a-priori mix of data for coding/training makes sense. 🧐

010

Nadja Rupprechter @naru96.bsky.social · 01/10/2026

Talk is done! Thanks to everyone attending. It was such a fruitful discussion in the panel – really enjoy the vibes here at the @dgpuk-meth.bsky.social conference! ⭐

071

Nadja Rupprechter @naru96.bsky.social · 30/09/2026

Not me finding a spelling error in Deutsche Bahns Safety Note lol Any usual events 🤷

000

Reposted by Nadja Rupprechter

Nicole Bizzotto @nicolebizzotto.bsky.social · 29/09/2026

📣 Important #ICA27 update for colleagues planning to submit: you can now be designated as presenter on up to THREE submissions 🎊

074

Nadja Rupprechter @naru96.bsky.social · 29/09/2026

Convinced there's a correlation between doing research and loving lore-heavy stuff (RPGs, games, fictional universes, historical facts...). Maybe doing research is just a way to get more lore about the real world.

static.klipy.com

Me Trying to Find the Lore in a Giant Book

ALT: Me Trying to Find the Lore in a Giant Book

040

Nadja Rupprechter @naru96.bsky.social · 13/09/2026

Ever wondered what (test!) chatlog data donations from ChatGPT and Character.AI look like – and how to set up extraction in UZH's Data Donation Module? datadonation.uzh.ch Little blog post (part 1) here: nadjarupprechter.com/nadja-ddm-bl... #commsky #datadonation #ai

1105

Reposted by Nadja Rupprechter

Chris Carpenter @chriscarpenter.bsky.social · 08/09/2026

Deeply disturbing. I generally support posting datasets but researchers need to do better about anonymity. Some of these cases would warrant an IRB investigation.

041

Nadja Rupprechter @naru96.bsky.social · 03/09/2026

With seasonal depression just around the corner sad scrub daddy just dropped

000

Nadja Rupprechter @naru96.bsky.social · 03/09/2026

This afternoon is all about studying at UZHs @ikmz.bsky.social! Really happy talking to interested young people about what communication science is all about 🙌

070

Nadja Rupprechter @naru96.bsky.social · 03/09/2026

I second!!! especially CFA for reflective constructs — excellent addition

020

Reposted by Nadja Rupprechter

Media Psychology & Methods @mepsyme.bsky.social · 14/08/2026

And that’s a wrap on our second retreat! The city of Bern treated us very well. We explored the city, enjoyed a movie night (Bicentennial man 🔥), visited the museum of communication, and exchanged ideas. We even spotted the Berner bear! (1/2)

1183

Reposted by Nadja Rupprechter

Klaus Pforr @klauspforr.eurosky.social · 31/08/2026

language precision is like shower water temperature in old buildings

2172

Reposted by Nadja Rupprechter

Juha Karvanen @juhakarvanen.bsky.social · 01/09/2026

When does“variance explained” actually mean causation? In our new preprint, Olli Saarela and I develop graph-based causal variance decompositions. The framework clarifies when components of an ordered variance decomposition have causal interpretations. arxiv.org/abs/2608.27140

arxiv.org

Graph-based causal variance decompositions: When "variance explained" means causation

Recursive application of the law of total variance decomposes the marginal variance of an outcome into components attributed to explanatory variables and a residual component. The resulting decomposit...

06617

Reposted by Nadja Rupprechter

simine vazire @simine.com · 28/08/2026

Psych Science (the journal) is broader than you might think! Do you do descriptive research? Or research w/ underrepresented populations or cultural frameworks? Or research on AI/LLMs for psych? Or research testing lab findings in the real world? We have a special issue for that!

24931

Nadja Rupprechter @naru96.bsky.social · 30/08/2026

Ohh no way — bummer! Glad you enjoyed it though 😊😊

010

Nadja Rupprechter @naru96.bsky.social · 30/08/2026

We had such a fun time at UZH/ETHs Scientifica! People really seemed to enjoy our booth about Deepfakes 🥳 @evavogel.bsky.social @larakobilke.bsky.social

272

Nadja Rupprechter @naru96.bsky.social · 26/08/2026

Fair! Let’s see how this one plays out

010

Nadja Rupprechter @naru96.bsky.social · 26/08/2026

😃 nice - zenodo is back on the menu!

231

Nadja Rupprechter @naru96.bsky.social · 23/08/2026

Why is this tempting me to put SOM in the ms just to bug you 😂

000

Nadja Rupprechter @naru96.bsky.social · 23/08/2026

I am so happy that bluesky added drafts!!

110

Nadja Rupprechter @naru96.bsky.social · 14/08/2026

Downloaded an AI voice changer to pitch my voice to male when playing Counterstrike 2 and guess what happened? I have never experienced such relaxed and easygoing games before 😭

210

Reposted by Nadja Rupprechter

Matti Vuorre @matti.vuorre.com · 13/08/2026

New on the PsyArXiv blog: "Preprints: Where They Fit in Today’s Public Access Policy Landscape" blog.psyarxiv.com/2026/08/13/p...

Quote from linker article:

Benefits of preprint publishing for researchers

Speed: Preprint publishing is fast, especially when compared with traditional journals.
Citable outputs: Preprint platforms, such as PsyArXiv, assign persistent identifiers like DOIs, making preprints fully citable.
Early credit: Preprints help researchers, especially those early in their career, demonstrate productivity and establish precedence.
Improving access: Posting a preprint increases accessibility by making an open access version of the work available.
Foundation for peer review: Preprints provide an opportunity for early feedback and can be peer-reviewed through platforms such as Review Commons and PREreview.
Documentation and verification: Versioning, timestamps, and persistent identifiers provide a transparent record of the publication.

082

Nadja Rupprechter @naru96.bsky.social · 13/08/2026

We just received the warmest welcome as new fellows of the Center for Reproducible Science (CRS) here at UZH 🥳(@philippesloksnath.bsky.social) Its amazing to be part of an exciting group of researchers that push for reproducible and trustworthy research practices: www.crs.uzh.ch

031

Reposted by Nadja Rupprechter

Aki Vehtari @avehtari.bsky.social · 13/08/2026

All four books I've co-authored are freely available online for non-commercial use: - Bayesian Workflow at avehtari.github.io/Bayesian-Wor... Links to other three books are in the quoted post 👇 (too many books to fit in one post!)

avehtari.github.io

Bayesian Workflow book: Website – Bayesian Workflow book

Website for the Bayesian Workflow book by Gelman, Vehtari, McElreath, et al. — case studies, code, and exercises in R and Stan.

8287134

Nadja Rupprechter @naru96.bsky.social · 11/08/2026

Team retreat with the best in Bern 🥳

0114

Nadja Rupprechter @naru96.bsky.social · 09/08/2026

Grieving someone you’ve never met 🥺—parasocial relationships make it possible. Didn’t expect to find this today in the International Society for Presence Research (ISPR) newsletter, edited by Matthew Lombard (which I highly recommend subscribing to!) ispr.info/2026/08/07/p...

ispr.info

Parasocial presence: Sydney Towle

[Apologies for the personal and lengthy introduction to this post. When a person who is close to us, or as many of us know, a pet, dies, even when the death is expected, we feel not only sadness and…

000

Nadja Rupprechter @naru96.bsky.social · 06/08/2026

I have to say it, the new mousemat goes HARD

010

Nadja Rupprechter @naru96.bsky.social · 30/07/2026

WD-40 mentioned!!!!!!! source: Putnam's Bowling Alone (2000)

050

Nadja Rupprechter @naru96.bsky.social · 22/07/2026

Anyone going to the @dgpuk-meth.bsky.social conference in Cologne in September/October, beware of weird spam mails about hotel bookings 🚨 #commsky

021

Nadja Rupprechter @naru96.bsky.social · 17/07/2026

Nice, I can require my BA students to use Quarto Markdown for their thesis analyses. Purely for their scientific development, of course, and definitely not because I'd rather not squint at 400 lines of unannotated raw R script with no structure 😇 Would you look at that!!

020

Nadja Rupprechter @naru96.bsky.social · 15/07/2026

New mug alert! A piece of Vienna at the Zurich department 💓

040

Nadja Rupprechter @naru96.bsky.social · 15/07/2026

Does any one of you smart folks have experience with, or pointers to literature on, bot exclusion strategies in surveys? We've built a few detection strategies into ours (CAPTCHA, AI/human task, attention checks, math tasks) – but it's not entirely clear how to use that information best. #commsky

010