Hi Helmut, Thanks so much for your informative comments. They make a lot of sense and will help a lot~~~ All my best, Chunliang ------------------ 原始邮件 ------------------ 发件人: "H. Nebl";<[log in to unmask]>; 发送时间: 2015年7月23日(星期四) 晚上9:56 收件人: "SPM"<[log in to unmask]>; 主题: Re: [SPM] followed-up t-tests after 2 by 2 ANOVA whole-brain analysis Oh yes, and turning to another test (be it a non-parametric or a parametric one) wouldn't solve the issue, as you would still rely on the originally detected, sig. clusters/voxels.