 |
 |
Dear Keri, In case you think of something like this http://www.brainvoyager.de/BV2000OnlineHelp/BrainVoyagerWebHelp/Deconvolution_analysis_of_rapid_event-related_designs.htm you would have to choose the FIR option during model specification. Ideally, the onsets of the trials should not occur randomly relative to TR onsets, as the specified onsets are rounded to the nearest TR onset (or time bin onset), which means loss of information, which could be problematic in case of rapid event-related designs. See a previous message https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1510&L=spm&D=0&P=72177 The last time I ran some FIR models there was an issue with orthogonalization of the different time bin regressors within conditions in case of "overlap" of trials, which one would not want to occur. See https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=spm;f525092b.1508 for more details. Not sure whether the code has been changed since then, but you can just comment these lines out when running the FIR models. Best Helmut