scoring a nonresponse as a correct answer

[David McFarlane]

Jim,

>I think I understand Mich's response now. Essentially, there needs to
>be a gap between the end of the response interval of a given stimulus
>and the presentation of the next instance of the same stimulus? What
>would be the least disruptive way to change my task so that non-
>responses are correctly logged (e.g., should I change the response
>duration to 1249ms)?

Hmm, even if that worked I think there is a better way.

>  Also just to make sure I'm clear on this, Is this
>problem related to the fact that my objects have 100ms of prerelease
>or is this a totally separate issue?

Quite likely related, the more I understand your program.

>I read the timing chapter twice and the prerelease section more than
>that; unfortunately I am still very confused. Your general guidelines
>help. From your description, it really seems like one should not use
>prerelease if they are logging responses. Or I'm just misunderstanding
>you :)

No, you got that right.  I would start with all PreRelease set to 0, 
and then add PreRelease only as needed.  And instead of doing this in 
my actual experiment program, I would start with a lot of little demo 
programs as exercises to first test my understanding.  Cumulative 
mode should do everthing you need to make sure that your task program 
stays more or less in sync with external equipment (brain scan, 
etc.), even in the presence of large OnsetDelays (in fact I have 
sometimes deliberately taken advantage of that fact); PreRelease 
contributes nothing to synchronizing with external equipment, and 
comes into play only when you need to reduce individual OnsetDelays, 
and even then some cases require techniques that go beyond PreRelease 
(preloading/caching, etc.).  So leave PreRelease at 0 until you know 
you need it, and know *why* and *where* you need it.

>I tried setting all objects to cumulative
>mode. Timing was relatively repeatable accross runs (give or take a
>millisecond); however, when I responded to the stimuli instead of just
>letting the task run without responses, the timing was not repeatable
>(30-50ms off).

That's just weird.  Unless... unless you have End Action = (none) or 
Jump anywhere, then all bets are off.  But then you should also get 
off by much more than 30-50 ms.  Or unless you have some inline code 
messing with TargetOnsetTime, or skipping past some objects, but 
short of that, with Cumulative Timing everything should Just Work 
regardless of responses.  So maybe it's time for you to take this to 
a local expert who can look at the program, or send it back to your 
collaborator and have their developer fix this.

-- David McFarlane, Professional Faultfinder

-- 
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.