Skip to main content
. Author manuscript; available in PMC: 2019 Nov 1.
Published in final edited form as: J Phon. 2018 Aug 29;71:147–161. doi: 10.1016/j.wocn.2018.07.008

Figure 7:

Figure 7:

An artificial example of how a mismatch between the model assumptions and the data can lead to poor posterior predictive fits. Here, 5% of the Mandarin VOT values were randomly replaced with 0 ms values and the same model as the one for research question 1 was fit. Now the posterior predictive check shows that the lack of fit between the data and the predicted values.