Frederik Aust

Ah yes, the eternal question. Always a good excuse to bump this *epic* CV answer by @noahgreifer.bsky.social

stats.stackexchange.com/questions/54...

Hippocampal encoding of memories in human infants Humans lack memories for specific events from the first few years of life. We investigated the mechanistic basis of this infantile amnesia by scanning the brains of awake infants with functional magne...

Why do we not remember being a baby? One idea is that the hippocampus, which is essential for episodic memory in adults, is too immature to form individual memories in infancy. We tested this using awake infant fMRI, new in @science.org #ScienceResearch www.science.org/doi/10.1126/...

Bayesian Inference and Testing Any Hypothesis You Can Specify - Alexander Etz, Julia M. Haaf, Jeffrey N. Rouder, Joachim Vandekerckhove, 2018 Hypothesis testing is a special form of model selection. Once a pair of competing models is fully defined, their definition immediately leads to a measure of ho...

(I’m really unfair in service of a joke here. These guys totally know their MCMC and Bayes. They continue:

„However, each test carries with it potentially unique implementation (i.e., computational) challenges.“

Nice paper!)

„With Bayesian methods, researchers can test any hypothesis they can specify sufficiently.“

Hybrid Workshop "Discovering Statistics Using JASP", June 30 - JASP - Free and User-Friendly Statistical Software We are happy to announce that the JASP team will offer a week of workshops in Amsterdam this summer.  In this blogpost we’ll highlight the first of four workshops: “Discovering Statistics Using JASP“....

🛠️☀️ 1 of 4 Hybrid Summer Workshops:

Improve your general statistics knowledge.
No experience w/ stat-techniques or JASP necessary.

Based on fresh-out-the-press textbook.
Taught by @ejwagenmakers.bsky.social
ECs for students available.

jasp-stats.org/2025/03/03/h...

Clearly this is an Ichthyophis! Get on top of your limbless amphibians, man.

I think you won‘t get around the nonlinear syntax. If I understand correctly, p. 34 shows how they do it. There is also some discussion of priors in there.

This is essentially a 3PL model (plus the lapse rate), right? This and other types of IRT models (and their implementation) are discussed in this paper: arxiv.org/pdf/1905.09501

Within the context of the study, there is a finite population of relevant stimuli, and the study had used nearly all of them (80%~).

That's when I came across this really cool paper about finite population correction (FPC) for (two-level) mixed linear models! Definitely worth a read!

2/

Variance, bias, and computational cost of estimating the Bayes factor using bridge sampling and the Savage-Dickey density ratio: http://osf.io/37xnb_v1/

Amazing collective problem solving in ants:

Sage Journals: Discover world-class research Subscription and open access journals from Sage, the world's leading independent academic publisher.

We show this for the Bayesian equivalent here:

journals.sagepub.com/doi/pdf/10.1...

Anyone Can Learn Echolocation in Just 10 Weeks—And It Remodels Your Brain Human echolocation repurposes parts of the brain’s visual cortex for sound, even in sighted people

'Anyone can learn echolocation in just 10 weeks' [via Scientific American] 🧪🦇🔈📍🧠

And it remodels your brain [the visual cortex, for the better, even if you are a 'sighted' individual]

www.scientificamerican.com/article/anyo...

#echio #location #human #visual #cortex #neuroscience

"When all meta-analytic evidence is taken into account, self-report measures are superior at predicting behaviour compared to implicit measures"

Very important question, though I'm not read up on it. But I do think some are too quick to dismiss self-report measures.

www.nature.com/articles/s44...

Our Introduction to Bayesian Data Analysis for Cognitive Science (with Nicenboim and Schad) is now in production with CRC Press. It will remain freely available here:

bruno.nicenboim.me/bayescogsci/

Method creep: When a grad student’s interest in methodology and statistics outcompetes and eventually extinguishes their curiosity about object level questions.
Many such cases.
#stats

$statcheck$ is flawed by design and no valid spell checker for statistical results The R package $statcheck$ is designed to extract statistical test results from text and check the consistency of the reported test statistics and corresponding p-values. Recently, it has also been fea...

Just found out about the JATSdecoder alternative to statcheck. Seems to have had a lot of work put into it, to be more robust to e.g. APA violations arxiv.org/abs/2408.07948 Any experiences?

Was spricht dafür, vom Lehrstuhlsystem in ein Departmentsystem zu wechseln und wie macht man das, wenn man es sinnvoll findet? - Dazu wurde ich von "Forschung und Lehre" befragt.

#DieAndereUniversität #IchBinHanna

Das Netzwerk Nachhaltige Wissenschaft hat einen Katalog von kritischen Fragen zur Exzellenzstrategie an GWK, DFG und den Wissenschaftsrat geschickt. Nun haben uns die ersten Antworten dazu erreicht. Fragen und Antworten findet Ihr hier:

netzwerk-nachhaltige-wissenschaft.de?p=287

Minimal exposure durations reveal visual processing priorities for different stimulus attributes - Nature Communications Minimal exposure durations required for visual processing of stimulus attributes can reveal the system’s priorities. Here, the authors show a distinct sequence of minimal exposures needed for face pro...

In our new @NatureComms paper, we investigate the visual system's priorities for extracting meaningful information from faces and bringing it to conscious awareness. We use a new tachistoscope that enables visual presentations as short as 0.002 ms: nature.com/articles/s41... (🧵👇)

@danielheck.bsky.social recently preprinted a paper on this response format.

osf.io/preprints/ps...

We're still doing this for the next 2 weeks, and looking for more volunteers. The interviews so far have been enormously helpful for us! Please consider sharing with someone who might be a good match for us to talk to, or send me their contact info via a DM and we'll reach out.

Our neural network model of differentiation and integration of competing memories, developed by Victoria Ritvo and Alex Nguyen, has now been published in @elife.bsky.social:
Paper: https://tinyurl.com/4b3k95rd
Code: https://tinyurl.com/mr25rk9e
#neuroskyence #psychscisky #neuroai (1/5)

All the reporting functions are fully compatible. The document template qapaja is more of an experiment at this point and not really functional. I do want to continue working on it next year.

I agree, but would note that the „Quarto“ syntax for chunk options is available in knitr/rmarkdown as well.

But I‘m also obviously quite invested in R Markdown (currently writing another package with a trimmed down papaja template to write student papers).

If I was to start from scratch, I would probably teach Quarto. I think it‘s still fine to teach R Markdown, but I would use the #|-syntax for chunk options to make a later transition to Quarto easier.

Random question - has anyone ever done a formal investigation into publication bias in rodent studies?

I always wonder how many lab-bench mouse studies are unpublished because they didn't find that some chemical caused more tumors. Is there a realistic estimate of this number?

Cognitive science journal Open Mind to be cofunded by Harvard Library and MIT Libraries through 2027 in an effort to advance open access publishing The MIT Press, Harvard Library, and MIT Libraries champion scholar-led open access publishing with Open Mind, a venue for quality research in cognitive science.

I'm so grateful to the Harvard and MIT libraries, in partnership with @mitpress.bsky.social, for making this happen:
mitpress.mit.edu/cognitive-sc...

I hope this will be the first step towards broader changes in how open access publishing is funded. Tell your libraries to join forces!

A case study on the challenges of theory testing in psychology “But when does lack of 'simplicity' in the protective belt of theoretical adjustments reach the point at which the theory must be abandoned?” - Lakatos, 1976 What does it take to falsify a…

What does it take to falsify a theory? New post by Maria Robinson & Jamal Williams on #psynomPBR paper by Robinson, Williams, John Wixted & Brady.