By JacintaS - 7/12/2016
Hi, I am using the script for ANT-R with a few changes. The current test goes for 30 minutes (4 blocks, 72 trials each:12 no cue, 12 double cue, 12 invalid spatial cue, 36 valid spatial cue). How do I reduce the time of the test to around 3-5 minutes (5 no cue, 5 double cue, 5 invalid spatial cue, 15 valid spatial cue)? Thanks!
|
By Dave - 7/12/2016
You will have to edit <list trialsequence> and its associated <list>s (CueTargetISI, FlankerCongruence, etc.) to only contain the number and types trials and associated values you want.
Once you've done that, you also need to adjust the /trials attribute in <block ANT_R> to run fewer trials in line with the above.
<block ANT_R>
/onblockbegin = [values.blockcount += 1]
/onblockbegin = [values.practice = 0]
/trials = [1 = fixation_start; 2-73 = list.trialsequence; 74 = rest]
</block>
|
By JacintaS - 8/11/2016
Thank you for the reply. I have tried to find some information however I am a bit confused as to what these lists are showing. Can I please have further clarification? Thanks
|
By JacintaS - 8/11/2016
Thank you for the reply. I have tried to find some information however I am a bit confused as to what these lists are showing. Can I please have further clarification? Thanks
|
By Dave - 8/11/2016
The purpose of those lists is detailed in the script's comments. It would be helpful if you could clarify what exactly remains unclear to you after reading through the code & comments. Then I can try to explain.
|
By JacintaS - 8/24/2016
I think what was confusing me was that in the scripts comments state that in one 72 block of trials there should be: 12 nocue; 12 doublecue; 12 invalid spatial cue; 36 valid spatial cues trials (split into 3 identical ones of 12 trials) per block. However instead there were; 72 trials with 14 no cue, 12 double cue, 11 invalid spatial cue, 14 valid spatial cue one, 12 valid spatial cue two and 10 valid spatial cue three.
My next question is that I was wanting the trials to occur in a randomised order rather than a predetermined sequence. How do I go about this and will randomising the trial sequence disrupt the other lists (cue target, flanker congruence, target location, target direction, location congruence, target cue).
Thanks!
|
By Dave - 8/24/2016
To achieve a random -- instead of sequential / fixed -- trial sequence, all you need to do is change
/ selectionmode = sequence
to
/ selectionmode = random
in <list trialsequence>.
The other lists are paired to list.trialsequence as per
/ selectionmode = list.trialsequence.currentindex
and will automatically retrieve the correct item -- regardless of whether <list trialsequence> is set to sequence or random.
|
By JacintaS - 8/25/2016
Thank you very much for all of your help, it is greatly appreciated!
I was just wondering how would it effect the test if the other lists were randomised as well?
|
By JacintaS - 8/25/2016
Also if the trial sequence is set to random will all trials be used only once in the test (i.e. trial is used and then not replaced). Thanks!
|
By Dave - 8/25/2016
> I was just wondering how would it effect the test if the other lists were randomised as well?
It would completely mess up the various conditions. <list trialsequence> contains trial elements. The various other lists contain *parameters* for the trial sampled from <list trialsequence> to use -- that's how the various conditions *within* trial types are realized: Different ISIs, congruent or incongruent flankers, target location, etc.
> Also if the trial sequence is set to random will all trials be used only once in the test (i.e. trial is used and then not replaced). Thanks!
Yes. Trials will be sampled without replacement from the list until there are no unsampled items left in the list. Then the list resets, i.e., all items become available again to be sampled (without replacement).
If you wanted to sample *with replacement*, you would set the list's /replace attribute to true (the default is false).
|
By JacintaS - 8/29/2016
Great! Thanks for all of your help!
|