Hello Torsten,

In the HCP protocol, we collect two runs of each task for each participant. 
Each E-Prime run corresponds to a specific stimulus list. E-Prime runs were not 
randomized across participants out of concern that variability related to the 
use of differing stimulus lists across subjects might interfere with the 
ability to detect individual differences. Consequently, our intent was (if 
everything went as planned in a session) that all participants would receive 
the same two E-Prime runs (run1 and run2) for each task.

However, we are unable to achieve that goal in all subjects, due to a competing 
desire to avoid possible practice effects in situations where “issues” arise in 
a given scan.  For example, due to technical problems (e.g,. scanner or E-Prime 
crash) or participant issues (e.g., squeezed the “panic” bulb), a given run 
might be aborted. To avoid repetition and practice effects on the specific 
stimuli from the prior runs, a new task list (run3 or run4) was used if the 
participant had seen more than 25% of the previous task run. 

The specific script used for a given task scan is recorded in the "Eprime 
Script Number" variable in the ConnectomeDB, as well as in the filename of the 
TAB.txt file in the packages (e.g., 
100307/MNINonLinear/Results/tfMRI_WM_LR/WM_run2_TAB.txt).

[As an aside, the E-Prime script for WM has 8 runs because we initially planned 
to utilize runs 5-8 in our 7T scan protocol. However, that was cancelled in 
favor of using different tasks in the 7T protocol.]

--Greg

____________________________________________________________________
Greg Burgess, Ph.D.
Staff Scientist, Human Connectome Project
Washington University School of Medicine
Department of Anatomy and Neurobiology
Phone: 314-362-7864
Email: [email protected]

On Aug 4, 2014, at 2:59 AM, Torsten Ruest <[email protected]> wrote:

> Hi there,
> 
> we'd like to use the HCP tasks and reading the Barch et al. paper, 2 runs of 
> each were delivered. When testing the tasks however, we noticed that there 
> are some for which there are more than 2 runs available (gambling (4), 
> language (4), relational (4), social (4), wm (8)). 
> 
> Runs were probably randomised, but could someone confirm?
> 
> Thank you very much in advance!
> 
> Best wishes,
> 
> Torsten
> 
> 
> 
> 
> _______________________________________________
> HCP-Users mailing list
> [email protected]
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
> 


_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to