Page 260 - Demo

Appendices258Figure 5 – Predictive posterior distribution of 50 simulated datasets based on the model (in grey) with the distribution of the original data plotted on top (in black) for the preference test model.PSISFor all dot-probe models and the preference test model, all k-values were < 0.5, indicating that no extremely influential observations were found.Chapter 7Experiment 1Posterior predictive checkFor experiment 1, posterior predictive checks showed no indication of model misspecification for both the first fixation- and the total fixation duration-model. For experiment 2, posterior predictive checks showed no indiviation of model misspecifications for the first fixation- and the total fixation duration-models (Figure 6).Tom Roth.indd 258 08-01-2024 10:42