We welcome contributions on all aspects of word processing (submit by 23 Jan 2026)!

Very happy to announce these two fantastic keynotes:
- Davide Crepaldi @davidecrepaldi.bsky.social
- Jana Reifegerste

maybe she's a proponent of the bootstrap-based method... 🤔

Instead, better tone identification was predicted by pitch acuity. This suggests that domain-general auditory processing―not universal phonological principles―is a more important determinant.

Preprint here: osf.io/preprints/ps...
(includes some goodies in the supp. materials, such as tips on speeding up model fitting.)

It's an intro to the foundational principles of Bayesian stats AND a practical tutorial on estimation and hypothesis testing using brms. I build up from simple regression to mixed-effects models.

Interesting... Tried mgcv::rmvn() and mvnfast::rmvn(). Both seems to produce the same random numbers (with and without the perturbations in the cov matrix).

right! mgcv can fit a binomial AR(1) model, but it's a little particular... you have to use bam() with discrete=T, for example (which, in turn, this precludes using something other than fREML)

package in the works! for now, you can source "OnsetDetection_Functions_0.4.7.R" from here and see the get_onset() examples in the analyses of Studies 1 and 2: osf.io/9ethq/files/... (note that the function still has some limitations, e.g., it only handles 1 random factor)

Interesting! We do fit a logistic GAMM with (temporal) autocorrelation... but we don't address the spatial autocorrelation at all. In any case, I think it's compatible? Our innovation is using posterior simulations to obtain a temporal CI, but there is some freedom in terms of the model building.

cool! we'd be super interested in hearing how well it works webcam VWP!

We provide a user-friendly R function for researchers to implement our method! Get it here: osf.io/9ethq/files

We formally evaluate a popular bootstrap-based method (Stone et al., 2021) and find that it can produce delayed estimates and inflated type I error rates. We propose a novel GAMM-based method that allows more accurate comparison of onsets between experimental conditions and/or speaker groups.