Hi Jill, I found my mistake. For the interaction of PPI, I set make zero for PHYS for min, and it should be mean. What I set for PSY was centre, and it should be right. I revised my mistake, and attached the design in the mail. Could you please help me to check it? Another question is, this time I found the result was still negative, Do I need to try with different seed regions got from group results? Or do I need to try to separate the two EVs, not one EV for contrast between the two? I need your suggestion to get positive results about this. Many thanks.