No Repeat Trials, Selecting Across Blocks


Author
Message
Dave
Dave
Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)
Group: Administrators
Posts: 13K, Visits: 104K

<block WhiteTrial>
/ trials = [1=noreplacenorepeat(WM1B, WM2B, WM3B, WM4B, WM5B, WM6B, WM7B, WM8B, WM9B, WM10B, WM11B, WM12B, WM13B, WM14B, WM15B, WM16B, WM17B, WM18B, WM19B, WM20B)]
</block>


Using noreplacenorepeat makes no sense when you are only drawing a single sample.


It is not possible to keep /trials selection pools intact across blocks.


Instead you need to re-work your script. There should be no reason to have a separate <trial> for each and every face. One trial for black faces and one for whites will suffice. Then you need to work with proper /resetinterval settings in your stimulus elements.


st2pa
st2pa
Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)Distinguished Member (2.8K reputation)
Group: Forum Members
Posts: 36, Visits: 19

Hello,


I am having some trouble in not allowing repeat trials in my study.  Here is my issue.  I would like participants to do ten trials with White faces, ten trials with Black faces.  Each trial involves presenting an image at first, then three seconds of a blank screen, and then a test screen where they see 12 versions of that face and they choose which one they saw.  This is what a test trial looks like in the code:


<trial BM1B>
/ stimulustimes =[
0=BM1L;
3000=eraseall;
6000=BM1B1, BM1B2, BM1B3, BM1B4, BM1B5, BM1B6, BM1B7, BM1B8, BM1B9, BM1B10, BM1B11, BM1B12]
/ inputdevice = mouse
/ validresponse = (BM1B1, BM1B2, BM1B3, BM1B4, BM1B5, BM1B6, BM1B7, BM1B8, BM1B9, BM1B10, BM1B11, BM1B12)
/ posttrialpause = 750
</trial>


The "BM" refers to a Black face trial.  So, I have twenty of these Black trials and twenty White trials.  Wanting an equal proportion to occur in the study, I then made blocks for each race:


<block WhiteTrial>
/ trials = [1=noreplacenorepeat(WM1B, WM2B, WM3B, WM4B, WM5B, WM6B, WM7B, WM8B, WM9B, WM10B, WM11B, WM12B, WM13B, WM14B, WM15B, WM16B, WM17B, WM18B, WM19B, WM20B)]
</block>

<block BlackTrial>
/ trials = [1=noreplacenorepeat(BM1B, BM2B, BM3B, BM4B, BM5B, BM6B, BM7B, BM8B, BM9B, BM10B, BM11B, BM12B, BM13B, BM14B, BM15B, BM16B, BM17B, BM18B, BM19B, BM20B)]
</block>


In the study experiment portion, my code looks like this:


<expt>
/blocks =  [1-20=noreplacenorepeat(WhiteTrial, BlackTrial)]
</expt>


The issue is that while this setup gives me 10 White and 10 Black trials, there are repeats within those Black and White trials.  For example, they could do WM15B twice as part of their 10 White trials.  Does anyone know how to fix this?  I tried playing with the 'resetinterval' attribute, but that has not made a difference, as it appears that only works for counters and this study does not have a counter.  Or, if I should be using a counter, do you know how I could implement it?  I am having trouble hooking up a counter to select from a set of trials, rather than a set of images.


Thank you for your help.






GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search