Hi folks: Here's a question that hopefully makes sense to you guys. In going through the HowTo5 tutorial, we are curious about how to choose the min/max lag for 3dDeconvolve for IRFs and the specification of stimuli in the all_stim.1d file.
After reading the Beauchamp et al. design, it seems that each film was presented for 2.5 s (followed by a 0.5 s fixation) for an "effective" stimulus duration of 3 s (pg. 997). The TR was 2 s. Later, on pg. 998, it states that 34 3 s fixation periods were interspersed between the 56 2.5 s movies (i.e., 14 trials x 4 conditions). But, how does a .5 fixation period post-movie lead to a 3 s stimulus duration, and, how does a 2 s TR fit into this? Also, on pg. 998, it then says that the combination of a 3 s stimulus with a 2 s TR allows for an effective TR of 1 s -- we weren't clear on this point either.
Second, why was a min-max lag range of 15 TRs used for this study, when,with TR=2 s, this would be a 30 s time window, and the movies were only presented for 2.5 s? That is, why would the IRF for these conditions peak midway through this 15 TR segment, as indicated in the demo (graph included)? Wouldn't this be 15 s after onset of the movie???? Anyway, we realized we didn't really understand the demo as well as we should, and hope you could help us out. Thanks in advance..... mbradley

p.s. Here is a portion of all_stims.1d file included in the tutorial.
0 1 0 0
0 0 0 0
0 0 0 0
0 0 0 1
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 1
0 0 0 0
0 0 0 0
0 1 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
0 0 1 0
0 0 0 1