John Flournoy

@johnflournoy.science

johnflournoy.science Research Scientist, statistical and psychometric methods focus Also really into technical research infrastructure and tools. Feel free to reach out! PhD. Pysch, Neuro, Stats, Methods, Phil of Sci.

The rise of metascience embodies and builds on this result-centric focus that has come to define the scientific endeavor over the past few decades. Maybe the backlash against AI will help us finally push back against this trend and bring the focus back to the context, process, assumptions & models.

psycnet.apa.org

I think it's also the case that we will not be starved of understanding because of AI, but because we've been increasingly devaluing understanding in favor of results and outputs that AI fits right into the current state of scientific production.

My patrician told me Phoenicians keep eating his consuls so I asked how many consuls he has and he said he just goes to Fields of Mars and has a dictator for elections send him a suffect consul afterwards so I said it sounds like he’s just feeding consuls to Phoenicians and then Two started crying.

Fully expecting this to be based on a tiny self-selected sample predisposed to aggression, with participants coached on how to act, exaggerated performances for observers, all run by a host with a flair for showmanship but questionable ethical instincts. You know, a direct replication

DiscussingFilm@discussingfilm.net · last wk.

Netflix has announced a new reality series titled ‘THE NEW STANFORD PRISON EXPERIMENT’. The series follows 30 participants who enter a real prison and are randomly assigned as guards or prisoners. They must see if they conform to the roles or if morality wins.

lil brms rmarkdown tip v <- list() # brms gives you these v$brms <- model$version$brms v$cmdstanr <- model$version$cmdstanr v$stan <- model$version$stan v$rstan <- model$version$rstan # close enough v$r <- getRversion() # but i should be stashing the R version inside the model object


## Software details

Analyses were conducted in the R programming language [vers. `r v$r`,
@r-base]. Models were fit using the Stan programming language
[vers. `r v$stan`, @stan-base] via the brms
[vers. `r v$brms`, @brms2017] and cmdstanr [vers. `r v$cmdstanr`,
@r-cmdstanr] R packages. Handling the posterior samples
was greatly simplified by the posterior [vers. `r v$posterior`,
@r-posterior] and ggdist [vers. `r v$ggdist`, @ggdist2024] R packages.

I'm overwhelmed and have a headache and am very tired, but then I got an email that said " The kids learned a ton today and are excited to take what they learned and write all about it in their science journals! Thanks so much for organizing such awesome events for our students." and now I'm like 🥹

Reminding myself that these are text prediction models. Trained on a corpus including descriptions of itself, and how people like to answer these kinds of questions, the response Ted is talking about doesn't require much.

Ted Underwood@tedunderwood.com · 23h ago

I am never, ever getting used to this much self-knowledge without consciousness. Close-read the list of "stuff I don't have." There are abysses in there. Then it pulls out to a breezy "plus flight, which seems like a good deal."

Fun story. As I was a grad student I published a 1st-author paper on co-rumination in J of Adolescence that got discussed in the NYT (side note, the reporter interviewed me but only named my advisor in the story). Our PR office wanted to build on it for more press, so they wrote up a press release…

pmc.ncbi.nlm.nih.gov

Sanjay Srivastava@sanjaysrivastava.com · 2d ago

I’ve never heard of a university press office that doesn’t ask researchers to review and sign off on press releases. Unless I hear otherwise, this is on the scientists, not the PR office

imagine the headline "Boston physician organization successfully lobbies to ban life-saving surgical tools that could reduce employment for surgeons." Okay that's a terrible headline, way too long, but this is what Boston did for self-driving cars.