I'm quite a noob to Matlab and SCR in general, and having worked through the tutorial in the manual, I'm still at a loss for how to format my data for import.
My experiment involved showing people subliminal pictures from 4 categories (positive, negative, neural, and threatening) while measuring continuous SCRs. I had one channel for EDA, and 4 channels for event markers. Participants experienced 16 trials of each category.
Given the length of the experiment (over 30 minutes) and the sampling rate, the files were very long. I initually used SAS to stack all participants files into one dataset, and I standardized within participants. This dataset is over 108million lines long. Then, I collapsed participant responses by creating an average of each category within a participant, so each participant's 16 trials were collapsed into one average trial. They had 4 of these corresponding to each category. This also effectively cut out all unnecessary data both before and after the trial, such as data during the fixation and the ISI.
So now I have a dataset where each participant has 4004 rows of data; 1001 rows for their average positive trial, 1001 rows for their average negative trial, etc...
As you can see in the attached .txt file, their subject id is designated "subj", the category of trial is "trial_type", the sample in each trial is "ms", which goes from 0 at trial onset to 1000 at trial offset; this repeats 4 times for each subject. Their eda value is "ave_eda_z", which is their averaged z-scored eda at each sample for each trial type.
I am wondering what more I need to do to get this into a format that is approrpiate for imoporting into PsPM and next steps to analysis. Talk to me like I'm an idiot, cause this process is literally making me feel like one!
PsPM expects continuous data. This is particularly relevant for EDA data where the responses are rather extended in time (> 90 s) and usually overlap with the next trial even if the inter trial interval is very long. The power of PsPM is that it takes all these data into account - in fact, ISI data are very informative.
Best practice would be to import one continuous file per participant. Using 4 marker channels will be a bit of a challenge in PsPM as it usually expects all information in one marker channel. But I'm happy to assist you using this information nevertheless.
Hope this helps
Dominik
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for the response. The reason I have been ignoring the ISI is because I had a task 4 seconds following the offset of the subliminal prime. I am also ignoring the 4 seconds prior to the prime because it appears as though participants are simply orienting to the fixation.
Is the ISI data really relevant when all I'm interested in is differential responses within the 4sec period just after prime presentation?
By continuous, I assume you mean not averaged together? The reason I averaged it was to deal with the unruly length of the file at over 108million lines long. When I tried to down-sample, I realized that because the even markers were only 10ms long, downsampling was actually erasing some of the markers, causing me to lose some trials. So I down sampled from 1000 samples per second to 250.
The data format right now does have all markers in one channel, but they are not in a format that PsPM would like.
Basically I'm wondering whether I can use the data as it currently is and if not, what I need to do to get it into a format that it would be usable.
I hope all that makes sense!
Thanks again,
David
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I'm quite a noob to Matlab and SCR in general, and having worked through the tutorial in the manual, I'm still at a loss for how to format my data for import.
My experiment involved showing people subliminal pictures from 4 categories (positive, negative, neural, and threatening) while measuring continuous SCRs. I had one channel for EDA, and 4 channels for event markers. Participants experienced 16 trials of each category.
Given the length of the experiment (over 30 minutes) and the sampling rate, the files were very long. I initually used SAS to stack all participants files into one dataset, and I standardized within participants. This dataset is over 108million lines long. Then, I collapsed participant responses by creating an average of each category within a participant, so each participant's 16 trials were collapsed into one average trial. They had 4 of these corresponding to each category. This also effectively cut out all unnecessary data both before and after the trial, such as data during the fixation and the ISI.
So now I have a dataset where each participant has 4004 rows of data; 1001 rows for their average positive trial, 1001 rows for their average negative trial, etc...
As you can see in the attached .txt file, their subject id is designated "subj", the category of trial is "trial_type", the sample in each trial is "ms", which goes from 0 at trial onset to 1000 at trial offset; this repeats 4 times for each subject. Their eda value is "ave_eda_z", which is their averaged z-scored eda at each sample for each trial type.
I am wondering what more I need to do to get this into a format that is approrpiate for imoporting into PsPM and next steps to analysis. Talk to me like I'm an idiot, cause this process is literally making me feel like one!
Thanks!
-David
Last edit: David S March 2018-08-03
Hi David
PsPM expects continuous data. This is particularly relevant for EDA data where the responses are rather extended in time (> 90 s) and usually overlap with the next trial even if the inter trial interval is very long. The power of PsPM is that it takes all these data into account - in fact, ISI data are very informative.
Best practice would be to import one continuous file per participant. Using 4 marker channels will be a bit of a challenge in PsPM as it usually expects all information in one marker channel. But I'm happy to assist you using this information nevertheless.
Hope this helps
Dominik
Hi Dominik,
Thanks for the response. The reason I have been ignoring the ISI is because I had a task 4 seconds following the offset of the subliminal prime. I am also ignoring the 4 seconds prior to the prime because it appears as though participants are simply orienting to the fixation.
Is the ISI data really relevant when all I'm interested in is differential responses within the 4sec period just after prime presentation?
By continuous, I assume you mean not averaged together? The reason I averaged it was to deal with the unruly length of the file at over 108million lines long. When I tried to down-sample, I realized that because the even markers were only 10ms long, downsampling was actually erasing some of the markers, causing me to lose some trials. So I down sampled from 1000 samples per second to 250.
The data format right now does have all markers in one channel, but they are not in a format that PsPM would like.
Basically I'm wondering whether I can use the data as it currently is and if not, what I need to do to get it into a format that it would be usable.
I hope all that makes sense!
Thanks again,
David