Continuous Performance Task

[David McFarlane]

Years ago we programmed a Continuous Performance Task - Identical 
Pairs (CPT-IP), and had a devil of a time getting E-Prime to 
randomize the trials without generating spurious target trials.  We 
finally resorted to just constructing one good random order outside 
E-Prime and then running that in a List set to Sequential order 
(which in fact duplicates just what the original researcher did in 
the 1980s using a slide projector).

So if I were to program this, I might well do just what you already 
suggested.  Using a spreadsheet program (e.g., MS Excel) I could 
probably easily generate the 250 rows, then copy & paste that over to E-Prime.

I guess your problem comes when you inlcude "_ O" and "X _" as 
stimulus pairs (using "_" to indicate a placeholder).  In that case 
you could randomly get those two pairs one after the other, which 
would indeed cause a spurious target.  But if you just never used one 
or the other of those two stimulus pairs, I think that you could 
safely let E-Prime do the randomization without fear of generating 
spurious targets.  Or am I missing something?  To be sure, you would 
want to do several test runs to check for this, perhaps combined with 
code in E-Prime to at least detect and log any spurious targets 
(something we did while troubleshooting that CPT-IP several years ago).

BTW, I checked -- oddly, we have not discussed CPT much on this 
Group, and I can find no example programs at STEP, Paul Groot's site, 
or the PST Samples area.

Best,
-----
David McFarlane
E-Prime training 
online:  http://psychology.msu.edu/Workshops_Courses/eprime.aspx
Twitter:  @EPrimeMaster (https://twitter.com/EPrimeMaster)


At 9/11/2014 10:02 AM Thursday, richard at wermes.de wrote:
>I would like to implement a Continuous Performance Task (CTP). With 
>this task, single letters are presented and subjects are asked to 
>react whenever they detect an X, appearing after an O appeared. I 
>need about 500 trials for that and lets say 50 target-combinations.
>
>The only idea that come to my mind was creating a list including the 
>variables stimulus1 and stimulus2, including a procedure showing two 
>textdisplays in a row (stimulus1 and stimulus2). My problem with 
>that: I would need 250 rows in that list. If I had less rows and 
>random cycles, i might get additional unwanted target-combinations..
>
>Any ideas for different, more ellegant approches?

-- 
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To unsubscribe from this group and stop receiving emails from it, send an email to e-prime+unsubscribe at googlegroups.com.
To post to this group, send email to e-prime at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/e-prime/5411cb66.1412320a.643e.3388SMTPIN_ADDED_MISSING%40gmr-mx.google.com.
For more options, visit https://groups.google.com/d/optout.