The sound of machine learning that is just logistic regression

Brad has been conducting outstanding meta research for several years. In this thread he points out how his and his coworkers’ work questions long-standing ‘knowledge’ about internal vs external foci of attention. A debate many of us were taught was settled when it is clearly not.

Super-interesting thread, and threads within the thread!

Good. Lots of people jumped on the “EXTERNAL FOCUS IS THE ONLY WAY” bandwagon despite plenty of anecdotal evidence to the contrary. Turns out the “real” evidence for external focus’ superiority wasn’t real strong.

From dancing to carrying a couch, we often coordinate our actions with others. Our paper addresses how the accuracy and time delays of visual and haptic feedback influence collaborative interactions between partners. journals.physiology.org/doi/abs/10.115…

It is astonishing that psychologists—trained to understand cognitive biases and errors—publish analyses that have been performed by only one person and make claims on that basis. (This is independent of the fraud discussion. People just make lots of mistakes.) /1

Love the honesty and openness in this thread. Some very relatable feelings!

I’m convinced researchers in my field treat sample size estimation as a tick box exercise. The thinking goes “let’s just do whatever it takes to placate reviewers and editors so that our paper gets published.” They fundamentally have no interest in high precision results.

Great job Laura St. Germain (she/her) at kicking off SCAPPS 2023 with your Young Scientist Award talk!

We are excited to share our new paper that dives into the role of reinforcement-based processes in task redundant motor exploration (doi.org/10.1098/rspb.2…)!

It is 2024 and it is still common practice in the medical literature to put a bunch of covariates in a regression model for some bad health outcome and declare the significant associations as "risk factors" that people should worry about. Pseudo science

Still time to sign up for this journal club event happening next Wednesday Feb 7th 👇

Always love reading about the progress in these updates (blog.system76.com/post/closing-i…) from System76. Super excited for the alpha release of COSMIC!

New paper out in EXBR! Me and @JCSkewesDK looked at how concepts relevant to decision-making and learning (e.g. confidence and prediction errors) are expressed in the movements associated with decision actions. I had great fun working on this! link.springer.com/article/10.100…

TBF, responder analysis provides a good means of distinguishing those who don’t understand clinical trials from those who do. #cuttingremark

Join us for a discussion with Eimear Dolan on May 15th Register: mcmaster.zoom.us/meeting/regist… All welcome!

Finally, while there are many individual articles on how to enhance motor behaviour research (e.g., journals.humankinetics.com/view/journals/… on power analyses from Brad McKay & colleagues), start with this fantastic overview of methodology provided by Dr Keith Lohse: journals.humankinetics.com/view/journals/…

Our paper “OPTIMAL theory’s claims about motivation lack evidence in the motor learning literature” is finally out! Grab a copy here: doi.org/10.1016/j.psyc…

Awesome work by Jan Calalo on this one!