Maximal log likelihood ratios (base e) across subjects of models relative to causal inference model (mean��s.e.m., see methods for details).

For the last two entries, we used the prior proposed by Roach et al. and Bresciani et al. together with correct inference (see text for more detail). All of the maximal likelihood ratios in the table are considered decisive evidence in favor of the causal inference prior, even when correcting for the number of parameters using the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC) [28]. These criteria are methods for enabling fair comparison between models. Models with more parameters always fit data better than models with fewer parameters. AIC and BIC are ways of correcting for this bias.

对于最后两项，我们采用了Roach等人与Bresciani等人提出的先验（prior）结合正确的推断方法（详细说明参见正文）。本表中所有最大似然比均被视为支持该因果推断先验的决定性证据，即便使用赤池信息准则（Akaike Information Criterion, AIC）或贝叶斯信息准则（Bayesian Information Criterion, BIC）[28]对参数数量进行校正后亦是如此。上述准则均为实现模型间公平比较的方法。参数更多的模型总是比参数更少的模型对数据的拟合效果更优，而AIC与BIC正是用于校正这一偏差的手段。