Lightnews — Scholar-powered news

Shravan Vasishth

@shravanvasishth.bsky.social

250 followers 63 following 45 posts

Professor of Psycholinguistics, University of Potsdam, Germany. Outside of work: Classical guitar, beginning student of piano. I like our cats, on occasion. Web page: vasishth.github.io

Posts Media Videos Starter Packs

Pinned

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

The physical book copies have arrived: www.routledge.com/Introduction...

3 17

Shravan Vasishth @shravanvasishth.bsky.social · 25d

New preprint investigating whether lossy-context surprisal can account for the locality and expectation effects found in Russian, Hindi, and Persian reading data: osf.io/preprints/ps...

OSF

osf.io

Shravan Vasishth @shravanvasishth.bsky.social · Aug 31

We are done with the ninth Statistical Methods for Linguistics and Psychology (SMLP) summer school, Potsdam, Germany. The tenth edition is planned for 24-28 August 2026.

3 16

Shravan Vasishth @shravanvasishth.bsky.social · Aug 17

A common disease in academia is that for any position X there is an academic with an opposing position not-X. It's just how academics are. One can't just blindly follow one or another person's recommendation, but develop a good understanding of it oneself, and draw one's own conclusions.

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 17

I guess then we should also recommend against using p-values because they are sensitive to the likelihood function assumed? :) Who was it that gave this recommendation? I'm guessing SIngmann but maybe I'm wrong.

Shravan Vasishth @shravanvasishth.bsky.social · Aug 17

Because this is a frequently asked question in summer schools I teach, I am thinking about adding a video lecture on this in my online materials for the book. Do also read the online chapter from our book on this topic:

bruno.nicenboim.me/bayescogsci/...

D Model comparison - Extended | Introduction to Bayesian Data Analysis for Cognitive Science

Introduction to Bayesian data analysis for Cognitive Science.

bruno.nicenboim.me

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 17

Also:

Cumming, G. (2014). The new statistics: Why and how. Psychological science, 25(1), 7-29.

Kruschke, J. K. (2011). Bayesian assessment of null values via parameter estimation and model comparison. Perspectives on Psychological Science, 6(3), 299-312.

epub.ub.uni-muenchen.de/74222/

Bayesian Decisions using Regions of Practical Equivalence (ROPE): Foundations

epub.ub.uni-muenchen.de

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 17

Some other books to read:

Royall, R. (2017). Statistical evidence: a likelihood paradigm. Routledge.

Spiegelhalter, D. (2024). The art of uncertainty: how to navigate chance, ignorance, risk and luck. Random House.

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

Spiegelhalter, D. J., Abrams, K. R., & Myles, J. P. (2004). Bayesian approaches to clinical trials and health-care evaluation. John Wiley & Sons.

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

You can use the credible interval to think about whether the pattern is consistent with the predicted effect. Even better if you can compare that interval to a model's a priori predicted interval. Read Spiegelhalter's 2004 book:

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

That's what the likelihood ratio test (aka anova) does, you compare likelihoods under the null and non-null to decide whether to reject the null. Bayes factors compare marginal likelihoods but the idea is the same. See our BF chapter.

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

E.g., if my data point is 0, the likelihood of mu being 0 (assuming sd=1) is dnorm(0,0,1). The lik. of mu being 10 is dnorm(0,10,1), which is much smaller than that of dnorm(0,0,1). So I'd favor mu=0 over mu=10. If the data point were 10, the lik. of mu=10 is much higher, and I would reject mu=0.

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

So with Bayes factors you'd compute the marginal likelihoods under the two models and compare them.

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

To make a discovery claim, I would have to set up two models, m0 where the effect is 0, and mfull, where it is not. Then I can talk about the relative evidence assuming a null or a non-null effect. 2/2

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

I would say that both credible intervals in my example are *consistent* with the effect being positive, but I would not go so far as to make a discovery claim. 1/2

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 14

The physical book copies have arrived: www.routledge.com/Introduction...

3 17

Shravan Vasishth @shravanvasishth.bsky.social · Aug 13

Also see: pure.uva.nl/ws/files/636...

pure.uva.nl

1 1

Shravan Vasishth @shravanvasishth.bsky.social · Aug 13

Just a quick question first: If the Bayesian 95% Credible interval is [0.0000001,100] you would reject the null, and if it is [-0.0000001,99], would you fail to reject or even accept the null?

2 1

Reposted by Shravan Vasishth

Dario Paape @dariopaape.bsky.social · Aug 5

New preprint with @shravanvasishth.bsky.social: We use a computational model to look at how linguistic context affects first-pass attachment and reanalysis. Assuming a tripartite mixture of reanalyzed, non-reanalyzed and inattentive trials outperforms LLM surprisal. lingbuzz.net/lingbuzz/009...

Context ameliorates but does not eliminate garden-pathing: Novel insights from latent-process modeling - lingbuzz/009217

A theoretically important claim in psycholinguistics is that in English, linguistic context largely eliminates garden-path effects in temporary complement clause/relative clause ambiguities (e.g., Alt...

lingbuzz.net

1 9

Shravan Vasishth @shravanvasishth.bsky.social · Jul 10

Next week onwards, I'm teaching a five-day introductory course on Bayesian Data Analysis in Gent. Newly recorded video lectures to accompany the course are now online: vasishth.github.io/LecturesIntr...

Shravan Vasishth's Intro Bayes course home page

vasishth.github.io

5 13

Shravan Vasishth @shravanvasishth.bsky.social · May 22

In person (no streaming/zoom) sentence processing workshop at Potsdam with Tal Linzen, Brian Dillon, Titus von der Malsburg, Oezge Bakay, William Timkey, Pia Schoknecht, Michael Vrazitulis, and Johan Hennert:

vasishth.github.io/sentproc-wor...

Sentence processing workshop, May 27, 2025

vasishth.github.io

2 6

Shravan Vasishth @shravanvasishth.bsky.social · May 16

Three year postdoc position in Cognitive Science, Potsdam, Germany:

www.uni-potsdam.de/en/verwaltun...

Temporary positions for academic staff

Temporary positions for academic staff at the University of Potsdam

www.uni-potsdam.de

5 6

Shravan Vasishth @shravanvasishth.bsky.social · May 11

New study on local coherence, led by Pia Schoknecht :

The time course of local coherence effects: Evidence from self-paced reading times and event-related potentials

osf.io/preprints/os...

OSF

osf.io

Shravan Vasishth @shravanvasishth.bsky.social · May 8

Scientists wanting to move to Europe from the US take note of this funding opportunity:

erc.europa.eu/news-events/...

Choose Europe for Science: ERC welcomes new budget for 'super grants'

The Scientific Council of the European Research Council welcomes the offer of substantial additional budget from the European Commission for the development of a new ERC funding instrument offering la...

erc.europa.eu

2 2

Shravan Vasishth @shravanvasishth.bsky.social · Mar 24

@cos.io Are you aware that in restructuring your website you have destroyed many repositories? The new Files directory is empty, and add-ons to github repos no longer work. WTF?

Shravan Vasishth @shravanvasishth.bsky.social · Mar 20

Computational Psycholinguistics Meeting 2025

cpl2025.sites.uu.nl

When: December 18–19, 2025

Where: Utrecht, the Netherlands

Abstract submission deadline: June 15, 2025

Organizers: Jakub Dotlačil, Lena Jäger, Bruno Nicenboim, Ece Takmaz

Computational Psycholinguistics Meeting 2025 | Universiteit Utrecht

Universiteit Utrecht

cpl2025.sites.uu.nl

3 1