The paper discusses the analysis of reaction times (RTs) in experimental psychology, emphasizing the use of mixed-effects modeling to predict individual responses rather than aggregated data. The authors advocate for empirical flexibility in choosing transformations for RTs and minimal a-priori data trimming, combined with model criticism. They also address the importance of considering trial-by-trial longitudinal dependencies and the evaluation of interactions involving fixed-effect factors that partition the levels sampled by random-effect factors. The study highlights the variability in RT distributions and the need to transform RTs to meet the assumptions of linear models. The authors provide examples of how to handle outliers and temporal dependencies in RT data, using a large dataset with a complex random-effects structure. The mixed-modeling approach is shown to be effective in capturing the individual differences in RTs and improving the fit of the model.The paper discusses the analysis of reaction times (RTs) in experimental psychology, emphasizing the use of mixed-effects modeling to predict individual responses rather than aggregated data. The authors advocate for empirical flexibility in choosing transformations for RTs and minimal a-priori data trimming, combined with model criticism. They also address the importance of considering trial-by-trial longitudinal dependencies and the evaluation of interactions involving fixed-effect factors that partition the levels sampled by random-effect factors. The study highlights the variability in RT distributions and the need to transform RTs to meet the assumptions of linear models. The authors provide examples of how to handle outliers and temporal dependencies in RT data, using a large dataset with a complex random-effects structure. The mixed-modeling approach is shown to be effective in capturing the individual differences in RTs and improving the fit of the model.