Jack Dempsey

"Why are people leaving academia?" Literally this person.

The fucking gall of putting into a review that "you're surprised we didn't care about your points" is absurd, and sorry but you're theory is not the end all be all. Earlier work already covered that honey

Can't decide if tidyverse is the spicy version of SQL or if SQL is the spicy version of tidyverse?

Thankfully they're not making me replace all instances of my name with "Author" because I've done that before and...there is no clearer way of signaling who I am than that haha

Ahhh! Thanks - that may potentially prevent getting this manuscript sent back again! :)

Holy shit, the "anonymous" URL of an OSF repository still has the non-anonymous URL inside of it? This is such a poetic example of how double blind doesn't actually exist.

Journals literally cannot understand the concept that a numerical decrease that is insignificant should not be considered a decrease. Why am I even running models then? If you just want the means then stop asking for stats.

British American speakers - is there a stereotype for American English speakers speaking fast or slow?

I'm going to be presenting in every poster session at #HSP2025 so I'll most likely see you there!

And are these learning styles in the room with us right now?

Excited to party with other psycholinguists at Planet Word for #HSP2025 - that sounds so cool!

I'm also presenting two posters (eye-tracking/Cascade Reading, an NC experiment), as well as a demo session with Julie Van Dyke about engaging communities with syntax! Hope to see everyone soon!

Hot take, but you don't have to wait until the day something's due to turn it in.

To anyone who hasn't received a review request from an MDPI journal yet...just don't do it. Unless you want dozens more requests from random journals thrown at you monthly for the next several years.

I think it depends on how your model is fit to the data. If the ID variable is entered as a predictor like the others in an ANCOVA then it would be a covariate analysis. If the ID is entered as a random, group-level effect in a mixed model I would just say a multilevel analysis.

Cascade Reading Introduction Webinar Are you a K-12 teacher with students who struggle with reading comprehension? Are you an English, Spanish, French, German, or Italian teacher with students who struggle to understand the grammar of th...

Interested in learning more about Cascade Reading, how it can improve reading and language comprehension outcomes, and how it can be leveraged for research or implementation into e-learning platforms? Come check out a free webinar I'm
hosting Tuesday, March 4th at 4:00 PM US Central Time!

Image of book cover for THE AI CON with title, subtitle (How To Fight Big Tech's Hype and Create the Future We Want) and author names (Emily M. Bender and Alex Hanna).

How about a little good news in this *gestures around*... you can now get ebooks via bookshop.org, including THE AI CON! (Available May 13, but you can pre-order now.)

bookshop.org/p/books/the-...

w/ @alexhanna.bsky.social

I feel like psycholinguistics journals may have coddled me about APA requirements - learning the hard and irritating way that other journals don't just leave those details to copy editors.

OpenAI Furious DeepSeek Might Have Stolen All the Data OpenAI Stole From Us OpenAI shocked that an AI company would train on someone else's data without permission or compensation.

OpenAI is furious that DeepSeek might have stolen all of the data OpenAI stole from all of us

www.404media.co/openai-furio...

Ahhhh, okay that might be why, thanks - felt like I was going crazy

Multiple posters accepted at #HSP2025 this year - excited to talk about linguistically-driven text formatting and noisy-channel stuff!

Also, I appreciate wanting to know more about different modeling approaches, but in 2025 can we please as a field move past questioning why someone used Bayesian inference? I like maximal random effects structures, and so should you!

This is not the first time I've received a review saying "I don't know Bayesian hierarchical models, why didn't they use mixed-models?". Am I the one getting terminology wrong here? Aren't these the same thing? I think frequentists see Bayesian and think the model structure is different but it's not

@othershorestudio.bsky.social Amazing time watching Bi An's movie premiere of Permanent Residence tonight. Fantastic and unnerving, way to go!!!

This is a cool resource, I'll definitely be sharing!

I can dm you!

Going to start posting on here more frequently - might mention work-related things as it relates to psycholinguistics, but this is mostly for my academic persona. Give me all your garden paths.

This works provides a starting point for future research into psycholinguistic processes during source code reading, both for debuggin and for comprehension.

We also found that gaze durations were longer on bugged regions that produced an undesired result, not when the bug was syntactic in nature. This suggests that reading for debugging employs top-down strategies in early stages of processing, highlighting context's central role in source code reading.

We found that reading for debugging leads to more rereading and a much higher rate of skipping than reading for comprehension or for proofreading text. We also found that code complexity positively predicts the number of regressive saccades, likley reflective of a greater need for context.

We had experienced programmers' eye-movements and accuracy in a bug detection and identification task. One goal was to provide descriptive information about global eye-movements when reading for debugging in Python, and the other was to examine eye-movement behaviors associated with particular bugs.