Thanks Keith! I am currently working with the Movie Task fMRI 2mm/32k FIX-Denoised (Compact) dataset, and I noticed the issues below:
1) Two subjects (126931 and 745555) do not have MSMAll-registered time-series data. 2) Five subjects (181636, 473952, 536647, 552241, and 973770) have two runs (MOVIE1 and MOVIE2). 3) One subject (585256) has three runs (MOVIE1, MOVIE3, and MOVIE4). Best, Reza On Mon, Jun 3, 2019 at 8:33 PM Keith Jamison <kjami...@umn.edu> wrote: > Hi Reza, > > Your interpretation of the timing is correct. The validation segment in > each movie scan was shifted by 40-200ms for the "v2" version. This was done > in order to make that final 83 second clip begin precisely at the start of > the next TR for all 4 movie sessions. > > Given HRF variability, ignoring this change will probably not adversely > impact most analyses. > > -Keith > > > On Fri, May 31, 2019 at 9:05 PM Reza Rajimehr <rajim...@gmail.com> wrote: > >> Hi, >> >> The HCP S1200 Reference Manual says that there are two versions of 4 >> movie files. Some subjects have been scanned with one version, and the >> remaining subjects have been scanned with another version. Version 2 >> includes these changes: >> >> 7T_MOVIE1_CC1_v2 remains unchanged. >> >> 7T_ MOVIE2_HO1_v2 removed 1 frame of rest before the validation clip. >> >> 7T_MOVIE3_CC2_v2 removed 5 frames of rest before the validation clip. >> >> 7T_MOVIE4_HO2_v2 added 4 frames of rest before the validation clip. >> >> Here *frame* is a frame of the movie, right? >> >> Assuming that the movies are ~25 frames per second, the deviations are >> between 0 and ~200 ms in each movie file. Considering the TR of one second >> and the slow hemodynamic BOLD responses, these deviations are negligible, >> and we can possibly ignore them when concatenating time-series data across >> all subjects. Would you agree? >> >> Thanks, >> Reza >> >> _______________________________________________ >> HCP-Users mailing list >> HCP-Users@humanconnectome.org >> http://lists.humanconnectome.org/mailman/listinfo/hcp-users >> > _______________________________________________ HCP-Users mailing list HCP-Users@humanconnectome.org http://lists.humanconnectome.org/mailman/listinfo/hcp-users
