Response triggers with accurate timing

[David McFarlane]

Then something else is going on.  As explained in Chapter 3 of the 
User's Guide that came with E-Prime, EP will fail to log data 
whenever your program reaches the end of the Procedure before it gets 
a response, and in some specific situations a PreRelease will 
exacerbate that.  (If you look at the generated code you will see why 
-- EP puts the requested c.SetAttrib commands and a c.Log just before 
the End Sub of a Procedure, so once your program runs past that it is 
too late.)  But the code under discussion specifically holds up the 
code until it gets a response, so, PreRelease notwithstanding, there 
is no way that it could fail to log the response unless you are doing 
something wacky with your attribute logging.

As also suggested in that example you could set Duration to 0, but 
you say you have also tried that and it does not work.  All I can 
tell you is that the posted code has worked flawlessly for us, 
exactly as used & described in that thread.  So at this point you 
need to submit a request to PST Web Support at 
http://support.pstnet.com/e%2Dprime/support/login.asp and have them 
help you sort it out (and then please post back here with their response).


BTW, I trust that the line

writeport, &H37, 0

in your post is a typo, because if that really is in your code then 
you would likely get no 0 signal at all out to your device, because 
the proper line would be

writeport, &H378, 0

-- David McFarlane, Professional Faultfinder


>I had been working from that post but it doesn't work - at least not
>when I try it. In order to initiate the inline immediately you need to
>set a large prerelease time (as suggested in the thread) but this
>interferes with data logging in Eprime. So I can get accurate response
>triggers using this method but for at least 30% (but sometimes up to
>70%) of trials no data is logged by Eprime. : (
>
>
>On Dec 13, 7:03 pm, David McFarlane <mcfar... at msu.edu> wrote:
> > Did you see the thread 
> athttp://groups.google.com/group/e-prime/browse_thread/thread/cfc3d0307...?
> >
> > -- David McFarlane, Professional Faultfinder
> >
> > At 12/13/2010 08:25 AM Monday, you wrote:
> >
> >
> >
> > >I have been struggling to send triggers to Actiview at the exact time
> > >a participant makes a response.
> >
> > >I'm running a memory experiment where images are presented in a slide
> > >object and participants respond by pressing 1,2,3 or 4. I would like
> > >triggers to be sent a.) at the onset of the slide and b.) at the exact
> > >point a participant makes a response. I want the slide to run for
> > >2500ms but responses to be collected for 3700ms in total. (Slide1 is
> > >followed by a blank screen for 1200ms).
> >
> > >The first issue is fine. I use ...
> >
> > >slide1.onsetsignalenabled = true
> > >slide1.onsetsignalport = &H378
> > >slide1.onsetsignaldata = &H05 'send 5 so the value does not clash with
> > >a possible response.
> > >sleep (10)
> > >writeport, &H37, 0
> >
> > >However, I can't get the response trigger to send accurately. I have
> > >been trying to use slide1.inputmasks.Ispending to collect the
> > >participant's response and writeport it while the slide is presented.
> > >Unfortunaltely, this sends the trigger late (looks like it is being
> > >sent after slide1 has finished running). This remains the case even if
> > >I set the slide1 duration to 0 . I have also tried setting the slide
> > >duration in properties to 2500 and using a large prerelease on the
> > >slide object. This seems to send the response triggers on time
> > >although I'm only judging this by eye at the moment. However, it
> > >causes problems with data logging in eprime. I.e Eprime only logs
> > >about 30% of the responses unless I reduce the prerelease. Also, no
> > >matter what I try I seem to either get stimulus onset triggers OR
> > >response triggers but not both. (I am sending 0s after each trigger to
> > >clear everything but still having problems).
> >
> > >I would appreciate any help with this issue - I have honestly been
> > >trying for days to sort this by myself. I have found previous
> > >discussions useful and have looked at some of the sample experiments
> > >but I still haven't gotten close to solving this problem. I am able to
> > >achieve most of the subtasks but cannot put them together in a way
> > >that works.- Hide quoted text -
> >
> > - Show quoted text -
>
>--
>You received this message because you are subscribed to the Google 
>Groups "E-Prime" group.
>To post to this group, send email to e-prime at googlegroups.com.
>To unsubscribe from this group, send email to 
>e-prime+unsubscribe at googlegroups.com.
>For more options, visit this group at 
>http://groups.google.com/group/e-prime?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To post to this group, send email to e-prime at googlegroups.com.
To unsubscribe from this group, send email to e-prime+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/e-prime?hl=en.