Dear all, 

 to the best of my understanding, the number of bins in Marsbar analysis should correspond to a certain length of time. For example, 13 bins, with 2 sec for each bin = 26 sec. 

In my experiment, I broke down each trial to several time frames. So if I put in the bin box 13 but the length of my trial is shorter, the time course that I'm getting should show me the activation during the first time frame, and the activation of the next time frame. However, when I'm looking at the activation during the second time frame, the percent signal change (PSC) do not correspond to the PSC of the corresponding time in the first-time frame analysis. 

My suspicion is that it has something to do with the choice of baseline, but I'm not sure. Furthermore I'm not sure what is the implication to that. Should I choose a different number of bins for each trial, depending on it's length? if so, why does Marsbar offers a default of 13?

Would appreciate your suggestions, thank you very much,

David