Print

Print
You need to compare it to neutral1 and neutral2 as the baseline values may change between session and the baseline values could change due to the scanner and not the brain.

If fear1/2 and neutral1/2 are different tasks, the comparison you are making will be confounded by the task. If the only difference between fear1/2 and neutral1/2 is the pictures used, then the comparison of fear-neutral at each timepoint is fine. 

Best Regards, 
Donald McLaren, PhD


On Wed, Oct 28, 2015 at 2:44 PM, Albert <[log in to unmask]> wrote:
Dear Donald,

sorry for being imprecise!

In more detail: Each session equates a different treatment and each participant underwent each treatment/session. In every session/treatment participants viewed pictures eliciting e.g. fear and neutral images (fear pictures presented in session 1 = fear1, fear pictures presented in session 2 = fear2, and so on). Aside from the flexiblefactorial approach, would it be valid to build a contrast (e.g. fear1 - fear2) in the first-level and forward it into a one-sample t-test or do we have to include also the neutral condition (fear1 - neutral1) - (fear2 - neutral2)?

Thank you very much for your insights!

Kind regards,
Albert