Figure 1
Example Stimuli of Experiments 1 and 2. Each item was presented either intact or shifted—the bottom half was displaced to the left or to the right. The Spanish word “sangre” is the English for blood.
Figure 2
Mean response times (RTs, in ms) for the experimental conditions in Experiment 1. The error bars represent the standard errors, and the number above the bars correspond to the accuracy. Note: We observed larger reading costs, more akin to the model’s coefficients, for the harmonic means: (1) high-frequency words (529 vs. 534 ms for the intact and shifted words, respectively); (2) low-frequency words (585 vs. 590 ms for the intact and shifted words, respectively); and (3) pseudoword (604 vs. 610 ms for the intact and shifted pseudowords, respectively).
Figure 3
Posterior distributions in the latency data of Experiment 1. The grey areas represent the 95% credible intervals of each parameter estimate.
Figure 4
Exploratory quantile-based analyses using delta plots by computing the .1, .3, .5, .7, and .9 quantiles per subject in Experiment 1. The Y-axis reflects the reading cost (shifted – intact).
Figure 5
Mean response times (RTs, in ms) for the experimental conditions in Experiment 2. The error bars represent the standard errors, and the number above the bars correspond to the accuracy. Note: The values of the harmonic means were as follows: (1) high-frequency words (532 vs. 545 ms for the intact and shifted words, respectively); (2) low-frequency words (592 vs. 601 ms for the intact and shifted words, respectively); and (3) pseudowords (632 vs. 646 ms for the intact and shifted pseudowords, respectively).
Figure 6
Posterior distributions in the latency data of Experiment 2. The grey areas represent the 95% credible intervals of each parameter estimate.
Figure 7
Exploratory quantile-based analyses using delta plots by computing the .1, .3, .5, .7, and .9 quantiles per subject in Experiment 2. The Y-axis reflects the reading cost (shifted – intact).