Skip to contents

Ecosystem update — September 2026

Source: vignettes/ecosystem-update-2026-09.Rmd
ecosystem-update-2026-09.Rmd

eyeprocess 0.11.1 published on CRAN

eyeprocess 0.11.1 was published on CRAN on 28 September 2026. The canonical package page is https://CRAN.R-project.org/package=eyeprocess and the package DOI is https://doi.org/10.32614/CRAN.package.eyeprocess.

The release channels are intentionally distinct: CRAN currently provides 0.11.1, GitHub provides the newer stable v0.12.0 release, and master continues as the 0.12.0.9000 development line. Analyses should report the version actually installed rather than treating these channels as equivalent.

The related Python package gpbiometricspy now includes a fully exact-main-certified crossed participant–item Gaussian hierarchical location–scale model with one location random slope for each crossed factor.

This does not modify eyeprocess, its vendor-neutral measurement workflows, psychometric/process-analysis surface, or release record. The new gpbiometricspy method is a downstream modelling option for appropriately designed continuous repeated outcomes after acquisition, timing, validity, preprocessing, and construct-evidence questions have been handled explicitly.

Certified gpbiometricspy PR #129 is pinned to merge SHA d078e0366ace49c3ebeb2f6800bad6394d70631e: 14/14 exact-main push workflow families, 12/12 OS/Python matrix lanes, 782/782 tests, 14,015/14,015 statements, and 6,757/6,776 raw branches (99.7196%). The frozen gpbiometrics 2.0.0 R-parity surface remains 406/406 and is unchanged.

The scientific boundary remains important: crossed random slopes estimate conditional-association heterogeneity; they do not establish measurement validity, causal effects, participant traits, stimulus quality, or latent psychological states.