| T O P I C R E V I E W |
| Dichter |
Posted - Jul 29 2007 : 4:19:17 PM
Sorry if this was covered elsewhere.... I am trying to analyze block design data with fsl, so the trick is to get the 3-column behavioral file to reflect the duration of the blocks appropriately (e.g., my blocks are 30 seconds long, but the output of xcede_extract_schedules.pl indicates they were one second long because I coded my Cigal presentation files to indicate only the onset of a block).
Is there an option in xcede_extract_schedules.pl to indicate lengths of events? Thanks. |
| 9 L A T E S T R E P L I E S (Newest First) |
| syam.gadde |
Posted - Aug 01 2007 : 08:17:59 AM
Oops. Sorry about that. You can replace the $3 with $2 in my script. |
| Dichter |
Posted - Jul 31 2007 : 10:42:50 PM
Sorry: just confirming that its the *second* column (not the third) of the 3-column file that reflects duration: http://www.fmrib.ox.ac.uk/fslfaq/ |
| jim.voyvodic |
Posted - Jul 30 2007 : 1:32:14 PM
If you put ID codes on your stimuli then you can generate FSL 3-column "schedule" files by adding a 'fslscheds' string list of condition names and which ID codes go with that condition. For example, putting the following line into the top section of a CIGAL input file will create 2 FSL timing files ("sch_encode.txt" and "sch_retrieval.txt"):
fslscheds = {
encode 6 66 7 77
retrieval 104 204 144 244
}
The sch_encode.txt file will list the timing of "encode" conditions consisting of all stimuli that have ID codes of 6, 66, 7, or 77. A run of multiple stimuli with any combination of those codes willl be lumped together as a single timing block. If you only code the onset of a condition the procedure is slightly different.
Using this approach involves putting sensible ID codes on the stimuli, but once it's set up in the input files the appropriate output files are generated automatically.
Jim
Jim |
| Dichter |
Posted - Jul 30 2007 : 1:15:43 PM
Yes, please! |
| jim.voyvodic |
Posted - Jul 30 2007 : 1:14:31 PM
The latest version of CIGAL can also create 3 column FSL files for you automatically if you want. It's a fairly simple change to the stimulus input file. Let me know if you're interested in learning more.
Jim |
| syam.gadde |
Posted - Jul 30 2007 : 12:48:28 PM
We've used examples of both (space and tab delimited) and it worked fine.
This doesn't apply for showplay2xml, but if you're converting event data using eprime2xml, presentation2xml or eventstable2xml, Chris' suggestion is a good one (to nip it in the bud from the beginning). |
| Dichter |
Posted - Jul 30 2007 : 12:39:48 PM
Thanks, guys. Syam's fix did the trick given that all blocks are the same length. It looks like, however, the original file is tab-delimited and the new one is space-delimited. Don't know if that will cause a problem down the pipeline... |
| petty |
Posted - Jul 30 2007 : 11:36:05 AM
another option would be to add a block specific field when you convert your data to XML and have duration set to 30secs.
just add an extra line to your parsing file:
event $type=@block $onset=Time:msec $duration=@30000:msec
you can add more fields if you match other info
$MATCHEQUAL='Label'@'emotional' blockname='Label'
-- here i had a column "Label" and i am matching where the fields are labeled "emotional" |
| syam.gadde |
Posted - Jul 30 2007 : 09:42:08 AM
Sorry there isn't. You could either edit the XML files or FSL files manually, or if all your blocks are the same length, you could try
the following command:awk '{$2=30; print}' old_timing_file.txt >new_timing_file.txtThat will change the second column in all lines to 30. (Don't forget to then use the "new" timing file)
(Edited to change third column to second column) |
