Dario Paape
@dariopaape.eurosky.social
psycholinguistics @ Potsdam, Germany https://d-paape.github.io
Ah, always a pleasure to hear that universal icebreaker of the International Fraternity of Ostensibly Heterosexual Men: "Germany? Beautiful women! But football?" *makes so-so gesture with hand*
Come to Rio for #CogSci2026, stay for the hot dog with peas, corn, quail eggs, raisins and tiny fries on it.
Rewatched Ghibli's Tales from Earthsea with the kids yesterday. I hated it when I saw it in the cinema 20 years ago. But turns out it's actually... not that bad?I mean, it's still confusing and the tone is... bleak, but artistically it's still quite enchanting. Kinda Ghibli's The Black Cauldron.
Who's attending #CogSci2026 in Rio next week? 🇧🇷 I will be presenting joint work with @shravanvasishth.bsky.social and @tallinzen.bsky.social on LLM surprisal in garden-path sentences on Friday! arxiv.org/abs/2602.04489
If I were a coach I would occasionally say to the referees "You're right we did commit a foul there" instead of arguing all the time. This builds my credibility as an objective interlocutor in their minds
Hmm, the specificity of "eye ointment for use on the eye" makes me wonder what other kinds of eye ointments there are... ? Maybe ointments made *from* eyes for use on one's... ears?
Is dit jetze Dialekt oder ham die einfach dit n fajessen
Being deeply interested in What The Hell Is Wrong With Germany, the Miscikowski et al. study where we have the highest level of "narcissistic rivalry" in their sample (doi.org/10.1080/1529...) made me wonder how this shaped me, in the sense that I've learned to abhor competition and zero-sum games.
Wondering if we can learn something by looking at different cultures' equivalents of the idiom "It's not rocket science". German has "It's not witchcraft" - different because it evokes something arcane/irrational rather than just "very complicated". What does your language use? #linguistics
Schafft es jemand, die Anzahl der möglichen syntaktischen Analysen zu zählen?
Small experiment: Which of the following statements makes more sense to you? A) "One dog year is/equals seven human years" or B) "One human year is/equals seven dog years", or C) both are fine. #linguistics
One thing I love about the modern world is that it makes garden paths so much more interesting when common sense isn't useful anymore even as a rough guide to wtf happened. (Source: BBC) #linguistics
Currently preparing slides for my science slam that will ask the REAL questions
Chat, what do we know about agreement attraction from inside NN compounds? Or maybe something else is happening here. AA on possessives is always interesting to me because it's the kind of error that I subjectively feel I would never conceivably make, because it just feels so... wrong.
Excited to share that @shravanvasishth.bsky.social will be visiting @princeton.edu virtually tomorrow (Tuesday, April 7) 🗣️ Talk: "Can Context Eliminate Garden-Pathing? A Bayesian Approach" 12:00 PM ET 🛠️Workshop: "Theory-Informed Model Specification" 11:00 AM ET The zoom is open to all!
New preprint: NPI and depth charge illusions in Pythia LLMs of different sizes and at different training checkpoints. NPI illusion goes away but depth charge illusion stays or even gets stronger. Amazingly for me, I come out in favor of... construction grammar? arxiv.org/abs/2603.27855
What can LLMs tell us about the mechanisms behind polarity illusions in humans? Experiments across model scales and training steps
I use the Pythia scaling suite (Biderman et al. 2023) to investigate if and how two well-known polarity illusions, the NPI illusion and the depth charge illusion, arise in LLMs. The NPI illusion becom...
arxiv.org
New open-access paper with @shravanvasishth.bsky.social: "Context ameliorates but does not eliminate garden-pathing: Novel insights from latent-process modeling" doi.org/10.1016/j.jm...
Redirecting
doi.org
New preprint with @tallinzen.bsky.social and @shravanvasishth.bsky.social: We jointly model reading data from four tasks (SPR, BSPR, eye tracking, Maze) with a latent-process mixture, and find that it outperforms LLM surprisal in terms of predictive fit: arxiv.org/abs/2602.04489
What do people think of this construction? I'm not a native speaker of English, but for me "it" can't refer back to "WB-57". #linguistics
Shoutout to the person who added this passive-aggressive-ish comment to the CogSci LaTeX template
PSA: We're moving the deadline for submissions to the JML special issue on nonveridical language comprehension (www.sciencedirect.com/special-issu...), co-guest-edited by @golfwriterkiel.bsky.social and me, back to end of February! We're looking forward to many interesting contributions!
Is it just me or does the chorus of Last Christmas not make sense? X gives their heart to Y, and on the very next day, Y gives "it" away, "it" presumably being X's heart. So... is X now in love with some unspecified person Z or does "it" have a different antecedent here somehow? #linguistics
I think "(in)tolerance towards coordination of unlike things" is probably a good indicator of general linguistic pedantry in a person. Here we have "snow and slipperiness removal", which I personally don't feel good about. #linguistics
