Fig. 4. Results for Hedge et al. data.
A model assuming additive practice (blocks) and conflict fixed effects, and random subject intercepts and conflict slopes. Blocks refer to cumulative sets of 48 trials and each row represents one task. In the first column, dashed lines show mean response time (RT) for congruent (lower line) and incongruent trials calculated directly from data and Bayes Factors (BF) compare the assumed model in the numerator to models with (1) no practice effect and (2) a practice × conflict interaction. In the remaining columns, solid lines represent median predictions from the model, and gray bands in all graphs show 95% credible intervals from fits of the assumed model. Values provided in columns 2–4 are seconds-scale equivalents for fits to all 432 trials for the conflict effect (CE) in column 2, and the trait and noise standard deviations (SD) in columns 3 and 4, respectively. Column 5 shows trait precision.