Figure 3
From: Intact predictive motor sequence learning in autism spectrum disorder
Bayesian statistical testing. The Bayes factor (BF01) indicates the evidence for the null hypothesis of no difference between groups. The density distribution displays the prior and posterior distribution for the population effect size, with the median effect size estimated, and a 95% credible interval which contains the median effect size. (a) Sequence learning – easy: BF01 = 3.193, meaning that the data are over three times more likely under H0 and provide moderate support for null hypothesis of no difference between groups (b) Sequence learning – difficult: BF01 = 2.371, meaning that the data are over two times more likely under H0 and provide anecdotal support for the null hypothesis of no difference between groups. (c) Bayesian paired t-test for IES scores in the predictable difficult condition compared with the unpredictable condition. BF01 = 0.625, meaning that the data are more likely under H1, providing weak support for the alternative hypothesis of a difference between conditions. (d) Surprise-related slowing – easy condition. BF01 = 3.778, meaning that the data are over three times more likely under H0 and provide moderate support for null hypothesis of no difference between groups.
