我想使用两个分类因素（conditionStimuli = 3 个级别；sequenceTrials = 2 个级别）创建一个关于反应时间的贝叶斯层次模型。最初，我使用默认先验运行模型：
maximal_RTs.bmodel = brm(RTs ~ conditionStimuli * sequenceTrials + (conditionStimuli * sequenceTrials | Num_part)
, data = data_RTs_go
, warmup = 500
, iter = 3000
, chains = 2
, inits = "random"
, cores = 2
)
summary() 函数报告这些：我是否应该通过查看 1-95%/u-95% 值来删除与 0 相同的交互？
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS
sd(Intercept) 77.13 9.37 61.46 98.27 1.00 852
sd(conditionStimulishapes) 43.58 5.82 33.83 56.31 1.00 1220
sd(conditionStimulileaves) 50.70 6.34 39.76 64.69 1.00 1034
sd(sequenceTrialsNGG) 24.97 4.51 16.77 34.49 1.00 1945
sd(conditionStimulishapes:sequenceTrialsNGG) 13.75 6.63 1.39 26.87 1.00 1397
sd(conditionStimulileaves:sequenceTrialsNGG) 15.35 7.69 1.42 30.65 1.00 782
cor(Intercept,conditionStimulishapes) -0.22 0.15 -0.50 0.09 1.00 1034
cor(Intercept,conditionStimulileaves) -0.23 0.15 -0.52 0.08 1.00 1084
cor(conditionStimulishapes,conditionStimulileaves) 0.31 0.15 -0.01 0.59 1.00 895
cor(Intercept,sequenceTrialsNGG) -0.19 0.18 -0.52 0.18 1.00 2101
cor(conditionStimulishapes,sequenceTrialsNGG) 0.02 0.19 -0.36 0.38 1.00 1955
cor(conditionStimulileaves,sequenceTrialsNGG) -0.00 0.19 -0.37 0.36 1.00 2042
cor(Intercept,conditionStimulishapes:sequenceTrialsNGG) -0.38 0.29 -0.83 0.28 1.00 3516
cor(conditionStimulishapes,conditionStimulishapes:sequenceTrialsNGG) -0.16 0.29 -0.68 0.45 1.00 4298
cor(conditionStimulileaves,conditionStimulishapes:sequenceTrialsNGG) 0.29 0.29 -0.35 0.76 1.00 3323
cor(sequenceTrialsNGG,conditionStimulishapes:sequenceTrialsNGG) 0.15 0.31 -0.47 0.72 1.00 3609
cor(Intercept,conditionStimulileaves:sequenceTrialsNGG) 0.24 0.28 -0.38 0.72 1.00 3656
cor(conditionStimulishapes,conditionStimulileaves:sequenceTrialsNGG) 0.03 0.29 -0.55 0.57 1.00 3728
cor(conditionStimulileaves,conditionStimulileaves:sequenceTrialsNGG) -0.41 0.27 -0.82 0.25 1.00 3180
cor(sequenceTrialsNGG,conditionStimulileaves:sequenceTrialsNGG) -0.10 0.31 -0.63 0.54 1.00 2882
cor(conditionStimulishapes:sequenceTrialsNGG,conditionStimulileaves:sequenceTrialsNGG) -0.14 0.35 -0.76 0.58 1.00 1750
还有这些问题：
是否有可能（或有意义）在结果变量上设置先验（反应时间：前高斯）？
在预测变量上，什么类型的先验是合适的？
谢谢，
数据如下所示：
Num_part trial_type Go_type conditionStimuli ITI_ms response RTs correctResponse order_pres sequenceTrials sdt
2 1 Go Bent leaves 819 1 301 1 1 NGG 1
3 1 Go Bent leaves 771 1 237 1 1 GG 1
4 1 Go Bent leaves 1086 1 393 1 1 GG 1
5 1 Go Straight leaves 652 1 331 1 1 GG 1
7 1 Go Bent leaves 919 1 372 1 1 NGG 1
9 1 Go Straight leaves 802 1 359 1 1 NGG 1