Maximilian Maier

Funnily enough, these are the supplements from a large-scale collaboration on Reproducibility in Management Science, which also means getting the approval of all primary authors for using the data would likely be virtually impossible! Any previous experiences with this policy would be very helpful!

I was just trying to download some replication files from Management Science and came across this highly unusual policy! This seems very problematic as it would allow the journal to reject any work scrutinising the paper completely at their discretion, without possibility of submitting elsewhere.

Today's HotFresh recommended paper is:

Maier, M., Harris, A. J. L., Kellen, D., & Singmann, H. (2025). Decision making under extinction risk. Cognitive Psychology, 159. doi.org/10.1016/j.co...

I am recruiting (Canadian🇨🇦) graduate students at York University for Fall 2026 in Social and Personality Psychology! If you are interested in misinformation, political polarization, and/or computational social science, I'd love to hear from you. Feel free to reach out with any questions.

R code and output showing the new functionality: ``` r ## pak::pkg_install("quentingronau/bridgesampling#44") ## see: https://cran.r-project.org/web/packages/bridgesampling/vignettes/bridgesampling_example_stan.html library(bridgesampling) ### generate data ### set.seed(12345) mu <- 0 tau2 <- 0.5 sigma2 <- 1 n <- 20 theta <- rnorm(n, mu, sqrt(tau2)) y <- rnorm(n, theta, sqrt(sigma2)) ### set prior parameters ### mu0 <- 0 tau20 <- 1 alpha <- 1 beta <- 1 stancodeH0 <- 'data { int<lower=1> n; // number of observations vector[n] y; // observations real<lower=0> alpha; real<lower=0> beta; real<lower=0> sigma2; } parameters { real<lower=0> tau2; // group-level variance vector[n] theta; // participant effects } model { target += inv_gamma_lpdf(tau2 | alpha, beta); target += normal_lpdf(theta | 0, sqrt(tau2)); target += normal_lpdf(y | theta, sqrt(sigma2)); } ' tf <- withr::local_tempfile(fileext = ".stan") writeLines(stancodeH0, tf) mod <- cmdstanr::cmdstan_model(tf, quiet = TRUE, force_recompile = TRUE) fitH0 <- mod$sample( data = list(y = y, n = n, alpha = alpha, beta = beta, sigma2 = sigma2), seed = 202, chains = 4, parallel_chains = 4, iter_warmup = 1000, iter_sampling = 50000, refresh = 0 ) #> Running MCMC with 4 parallel chains... #> #> Chain 3 finished in 0.8 seconds. #> Chain 2 finished in 0.8 seconds. #> Chain 4 finished in 0.8 seconds. #> Chain 1 finished in 1.1 seconds. #> #> All 4 chains finished successfully. #> Mean chain execution time: 0.9 seconds. #> Total execution time: 1.2 seconds. H0.bridge <- bridge_sampler(fitH0, silent = TRUE) print(H0.bridge) #> Bridge sampling estimate of the log marginal likelihood: -37.73301 #> Estimate obtained in 8 iteration(s) via method "normal". #### Expected output: ## Bridge sampling estimate of the log marginal likelihood: -37.53183 ## Estimate obtained in 5 iteration(s) via method "normal". ```

Exciting #rstats news for Bayesian model comparison: bridgesampling is finally ready to support cmdstanr, see screenshot. Help us by installing the development version of bridgesampling and letting us know if it works for your model(s): pak::pkg_install("quentingronau/bridgesampling#44")

How do people choose between moral rules or cost-benefit moral reasoning when faced with dilemmas?
Research by @maxmaier.bsky.social et al suggests
metacognitive learning from *consequences*
shapes moral decision-making, which is surprisingly malleable:

buff.ly/6SREXa0

Our new research paper is out in NHB, which we apparently coauthored with acclaimed actor Ryan Reynolds. Critics are calling it his best work since Deadpool.

For all those at #CogSci2025, @glenspiteri.bsky.social will be presenting this work at 14:15 PDT (22:15 BST for those online and UK based)!

“Communicating Global Income Rank Increases Charitable Donations”

Session: ‘Talks 8: Social Cognition’

@cogscisociety.bsky.social

@fbartos.bsky.social @maxmaier.bsky.social‬ & EJ Wagenmakers have extended RoBMA with 3-level publication bias-adjusted model-averaged meta-regression models in R along with the usual code-free and intuitive @jaspstats.bsky.social implementation. Nerdy details are osf.io/preprints/ps...

In Deutschland wird leidenschaftlich über mögliche Änderungen an den Strukturen der wissenschaftlichen Zusammenarbeit diskutiert. Das Ziel ist, die vielfältigen Verführungen zum Machtmissbrauch, die das gegenwärtige deutsche System beinhaltet, stärker zu beschränken.

Incredibly grateful to my brilliant supervisors Adam Harris and @singmann.bsky.social, and all the other collaborators and mentors at UCL and other universities who supported my research and development - looking back, it feels like the PhD passed way too quickly!

Happy to share that I passed my viva yesterday without corrections and will start as Assistant Professor at Warwick Business School on 1st of August - I look forward to the next chapter at Warwick and to building new collaborations with the exceptional colleagues there!

Out in Cognitive Psychology, led by @maxmaier.bsky.social: www.sciencedirect.com/science/arti...
A new lottery task with choices that matter across trials; the risky option has a chance of going extinct, which ends the study. We derive optimal policies and develop a strategy-classification model.

Excited to share that this paper is now published in PNAS! (With @maxmaier.bsky.social & Falk Lieder) www.pnas.org/doi/10.1073/...

New preprint w/ @glenspiteri.bsky.social , @maxmaier.bsky.social & Falk Lieder: Telling people their global income rank (RBN) increased charitable donations. Also explored RBN’s effect on perceived global income distribution + whether asking how much others should donate increased giving (IDT)

🧵…

New paper with Tong Liu and Arndt Broeder, just accepted in Cognition. We test novel qualitative predictions from sampling-based models of probability estimation in an event ranking task. Results provide evidence for the idea that mental sampling underlies probability judgements.

First time on Blue Sky! Super happy this article is now out in an issue of Political Psychology! And honoured by the promo tweet with custom-made graphic! @leede-wit.bsky.social

Dave Lagnado is looking for a post doc to work on causal inference! The BR-UK team is doing some very cool stuff, so if you're currently looking for a job, check this out: www.ucl.ac.uk/work-at-ucl/...

I Have Written You A Book On Forensic Metascience Use it to cause trouble

Friends, I have written you a book on forensic metascience.

It is free. You can have it. Happy St. Valentine's Day.

If you wish to give me a gift back, you can use it to cause trouble - the greatest gift of all.

open.substack.com/pub/jamescla...

Overall, we find that participants are relatively good in terms of the qualitative strategies they employ. However, we also document the influence of loss chasing, scope insensitivity, and opportunity cost neglect, which increase deviation from optimality in certain conditions.

We derive optimal strategies for three different types of extinction and near-extinction events and compare them to participants' choices in three experiments. We also use a (dependent) mixture model to describe strategies on an individual level.

People often need to make decisions that involve risks with irrecoverable losses. To study these types of decisions, we developed a new task in which participants lose their (entire) bonus pay when drawing an extinction option.

New preprint "Decision Making Under Extinction Risk" with Adam Harris, David Kellen, and @singmann.bsky.social.

osf.io/preprints/ps...

We derive optimal strategies for three different types of extinction and near-extinction events and compare them to participants' choices in three experiments. We also use a (dependent) mixture model to describe strategies on an individual level.

We updated our preprint on moral decision-making in LLMs (osf.io/preprints/ps...) with a new study investigating sources of the yes-no framing bias and amplified omission bias. Results show that they likely arise from fine-tuning for chatbot applications. (w/ @maxmaier.bsky.social and Falk Lieder)

(1/5) Very excited to announce the publication of Bayesian Models of Cognition: Reverse Engineering the Mind. More than a decade in the making, it's a big (600+ pages) beautiful book covering both the basics and recent work: mitpress.mit.edu/978026204941...

To facilitate application, we also provide an implementation in the R package RoBTT and a JASP version.

Bayesian model-averaging for the t-test allows the data to guide the inference to be based most strongly on those models that predicted the data best. It thus increases robustness while alleviating many assumption checks.

We first introduce model averaging over equal and unequal variances to incorporate uncertainty about differences in variances in a model-averaged Bayesian t-test. We then extend the ensemble by model-averaging over t-likelihoods to accommodate outliers (robust model-averaged Bayesian t-test).

YouTube video by JASP Statistics Model Averaged Bayesian T-Test

New paper "Model-Averaged Bayesian t tests" in Psychonomic Bulletin and Review with Frantisek Bartos, @dsquintana.bsky.social, Fabian Dablander, Don van den Bergh, @maartenmarsman.bsky.social, Alexander Ly, & @ejwagenmakers.bsky.social

Paper: doi.org/10.3758/s134...
Tutorial video & summary below👇