I'm writing with hopes of a better understanding of making sense out of 2 different analysis for which I surprisingly get the same result.

My first analysis is:

1) A first level analysis per subject with a contrast that Late trials > Early trials. I have 10 trials, and take the first three as Early trials and the last three as Late trials. So my contrast is: -[1 -1 -1 0 0 0 0 1 1 1]

2) At a second level, I do a one-sample t-test with F contrast [1]

Here, I get a single significant cluster at -46 -64 28.

My second analysis is:

1) A first level analysis per subject with contrast Late trials>Early trials, my contrast is [1 1 1 0 0 0 0 -1 -1 -1].

2) A second level analysis, one-sample t-test with F contrast [1].

I am surprised that here I get the same significant cluster as in my first analysis: -46 -64 28

I'm wondering why two different analysis, which in essence are looking at opposite things (Late>Early and Early>Late) give me the same cluster.

Thanks in advance,

Joelle