Empirical validation programmes in eyeprocess

In eyeprocess: Harmonize Eye-Tracking, Pupillometry, Biometrics, and Psychometric Process Data

knitr::opts_chunk$set(collapse = TRUE, comment = "#>")

eyeprocess 0.9 treats validation as an explicit research object rather than an implicit property of an estimator. process_validation_design() declares the measurement regimes to study; run_process_validation() records recovery, interval coverage, convergence, warnings, and failures; and freeze_validation_reference() supports regression-style evidence freezing.

d <- process_validation_design(n_persons=c(50,150), n_trials=c(10,30), missingness=c(0,.15), sampling_rate_hz=c(60,120), replications=100)
x <- run_process_validation(d)
validation_recovery_table(x)
validation_coverage_table(x)
validation_failure_profile(x)
plot(x, type="recovery")

The built-in simulator is a software-validation fixture with known truth. It is not a substantive psychological theory. Recovery under a supplied data-generating process does not establish external validity.

eyeprocess documentation built on Sept. 28, 2026, 5:08 p.m.