Tessera
Folgen

Stefano Coretta

@scoretta.bsky.social

Lecturer/Assistant Prof at UoE — Research Methods, Statistics, Open Research, Non-dual monism — #neurodiverse #lgbtq #chronicillness stefanocoretta.github.io

212 Posts352 Follower392 Folgt
Stefano Coretta@scoretta.bsky.social · 3mo ago

The more I teach Bayesian stats the more I am surprised that students/researchers naturally go to “evidence weighting” and “credibility based on CrIs including 0 or not” despite me warning against it. It is probably so engrained in our minds that you just got back to it.

Nicola Rennie@nrennie.bsky.social · 3mo ago

📊 Five #ggplot2 functions I wish I'd known about earlier 📊 I've written a short blog (with examples) of some of the lesser-known {ggplot2} functions and arguments that make it easier to create better charts! Link: nrennie.rbind.io/blog/five-gg... #RStats #DataViz

Five ggplot2 functions I wish I’d known about earlier – Nicola Rennie

There are a few small tweaks you can make to your ggplot2 code to improve your charts. However, they’re not often mentioned. So here’s a few functions and arguments I wish I’d known about earlier.

nrennie.rbind.io

Dan Quintana@dsquintana.bsky.social · 3mo ago

Selecting an effect size for power analysis is hard. Many researchers fall back on Cohen's thresholds, but they have no empirical basis and vary wildly by field. Our new paper offers a better option: field-specific effect size distributions built from meta-analytic data doi.org/10.3758/s134...

Abstract Effect sizes are useful for understanding the magnitude of study results and for planning new studies via power analysis. However, despite their wide usage, effect sizes are often misinterpreted. This is mostly due to an over-reliance on general effect size benchmarks that were not intended for broad application across diverse research fields. Inaccurate effect size interpretations can lead to incorrect conclusions about the magnitude of study results and incorrect sample size estimates, thereby increasing the likelihood of false-positive results. This article introduces the ESDist R package, which is designed to calculate empirically derived effect-size benchmarks or a range of reliably detectable empirical effect sizes for a specific research question or field of interest by computing effect size distributions (ESDs). This package can be used on data that can be easily extracted from pre-existing meta-analyses to help researchers more accurately plan new studies or to better understand how an individual study might relate to other studies in their field. ESDist includes a set of features that make it easy to use in a priori power analysis. Moreover, the package includes a feature for estimating effect size benchmarks that account for publication bias and are weighted by effect sizes' variances, which addresses existing limitations of using ESDs for study planning or interpretation.
Stefano Coretta@scoretta.bsky.social · 3mo ago

Mathematical formalisation of verbal process models is what we need in linguistics. (Note that this does not entail reductionism nor determinism nor other things associated with scientific realism, which I don’t agree with myself. I got your back, constructivist friends).

Dan Quintana@dsquintana.bsky.social · 3mo ago

Why are oxytocin study results so inconsistent? Methods get most of the blame, but the problem is also theoretical as existing accounts are too vague to falsify. In a new preprint, I formalise the Allostatic Theory of Oxytocin as a computational model osf.io/preprints/ps... 🧵

Oxytocin's effects on human behaviour are inconsistent across studies, persisting despite improvements in methodological rigour. Part of this problem is theoretical: existing verbal accounts do not specify predictions precisely enough to be falsified. Here I formalise the Allostatic Theory of Oxytocin as a computational model, deriving eight testable propositions and evaluating each against simulation evidence. Three structural predictions that follow directly from the model's assumptions were confirmed. Of five genuinely discriminating tests, three were supported. The two unsupported propositions revealed boundary conditions the verbal theory could not identify. Formalisation underscored the importance of validating auxiliary assumptions regarding optimal oxytocin dosing and the measurement of baseline endogenous oxytocin levels, without which tests of the theory cannot be conclusively interpreted. Altogether, this formalisation provides a more precise and falsifiable account of oxytocin's role in behaviour than currently exists, and a template for incrementally refining verbal theories in psychology through formal specification.
Riccardo Fusaroli@fusaroli.eurosky.social · 4mo ago

I ported the notes to Quarto and now all the old links are broken 😅. Here the new link to the chapter on Bayesian models of cognition: fusaroli.github.io/AdvancedCogn... Working now on 3 chapters on models of categorization (exemplar, prototype, rule-based)

10  Bayesian Models of Cognition – Advanced Cognitive Modeling Notes

fusaroli.github.io

Riccardo Fusaroli@fusaroli.eurosky.social · 4mo ago

First tentative draft of huge chapter on bayesian models of cognition for my Cognitive Modeling (in @mc-stan.org )course notes: fusaroli.github.io/AdvancedCogn... I'll probably need to split it, as I want to add models of sequential update and empirical data 😅

Stefano Coretta@scoretta.bsky.social · 4mo ago

Had a blast at #BAAP2026! [That also involved ripping my jeans while performing ballet at the bar (the drinks bar) for some of the conference dinner attendees who surely didn't ask for it] Jokes apart, amazing research being done and was happy to see Bayesian stuff!

Stefano Coretta@scoretta.bsky.social · 5mo ago

If research has no applied end-goal, it still matters. (Note that “no applied end-goal” doesn’t mean “no potential application”. I believe that all research has potential for application, that’s not the point I am making).

Stefano Coretta@scoretta.bsky.social · 5mo ago

A lot of (non-philosophical) discourse around science in linguistics and adjacent disciplines lacks a solid philosophical backdrop. What’s the point in trying to convince people that what you do is “science” when demarcation is a problem in itself?

Katrin Auspurg@kauspurg.bsky.social · 5mo ago

Do researchers share their code upon request? Does running their orginal code on the original data produce the original results? We provide evidence in a new Royal Society Open Science publication. Studying more than 1,000 articles which use data from the European Social Survey, we find that... 🧵

Bild
Stefano Coretta@scoretta.bsky.social · 5mo ago

🎉 - NEW MANUSCRIPT with @jess-hampton.bsky.social Beyond the horizon: a more-than-human and holistic approach to language lingbuzz.net/lingbuzz/009...

Beyond the horizon: a more-than-human and holistic approach to language - lingbuzz/009790

This position paper proposes a more-than-human and holistic approach to language that moves beyond entrenched divides between structuralism, constructivism, positivism, and methodological camps. Writi...

lingbuzz.net