Fig. 12.

Bayesian model comparison about polynomial degree within the normal aging group with N = 140. Top row shows log Bayes factors comparing a model with random intercepts and random slopes to a model with only random intercepts (D = 1 vs. D = 0). Middle row shows the comparison of a random intercept and random slope model to a random intercept model with fixed effects slope (D = 1 vs. D = 0, D_f = 1). Bottom row shows log Bayes factors comparing a full second order random effects model compared to a linear random effects model with random slopes and intercepts (D = 2 vs. D = 2). Columns show model comparisons separately for gray matter, white matter and CSF. Higher mixed-effect degrees were estimated but are not shown because of lower model evidence and limitations of space.