Extended reaction and accuracy

[liwenna]

Oh dear, it seems I answered too fast.

(Of course, we
took care that we use Slide1.Acc only when the time limit for reaction
is over).

^ I take it that that is the answer to my question already, sorry for
that.

Still the do-while-loop suggestions stands, I think.

There is one thing that I do not understand though: if I understood
you correctly, the extended reaction time for slide1 overlaps with the
display time of slide2? Yet you wish to distinguish between responses
(either correct or false) made during slide1 and slide2. Did I
understand that correctly? (probably not?).



On 11 feb, 20:32, liwenna <liwe... at gmail.com> wrote:
> Thanks for the more complicated set-up info. Hmmmz  Could it be that
> the inline is placed directly following slide1? And that therefore the
> inline script is already executed before the end of the extended
> reaction period, thereby not taking into account responses made during
> the extended response time? (rather than the .acc and .resp attributes
> being set at the end of the trial, I have no e-prime at home to fiddle
> with and I'd like to take you on your word, but it sounds a bit 'un-e-
> primey' to me)
>
> Is it needed that the trigger is sent immediatly following the correct
> response? (since you mention the option of a false response followed
> by a correct response, would you also take into account a correct
> response correct by a false response, in which case I suppose no
> trigger should be sent?) If false corrections do not need to be taken
> into account, you could perhaps do something with a 'do while' loop
> that continuously monitors slide1.resp and initiates the trigger the
> moment that slide1.resp = slide1.cresp?
>
> On 11 feb, 14:25, Maria <maria.richter.m... at googlemail.com> wrote:
>
> > Thanks liwenna for answering.
>
> > Of course, if you set the property „Time Limit“ to “same as duration”
> > of both slides an combine the accuracy of Slide1.acc and Slide2.acc it
> > works fine. The problem is that we distinguish between “correct
> > response”, “false response” and “no response” and even with only 2
> > Slides during the subject can react; there are 9 different
> > combinations of possible reactions we have to regard (e.g. the subject
> > could react wrong during Slide1 and try to undo it and press the right
> > key at slide2…). The more slides where the response can be given, the
> > more possibilities and if the duration of the slides are variable, it
> > is not possible (or very complicate) to calculate the accuracy if the
> > time for reaction should be a fix interval.
>
> > The easiest way is to set the property “time limit” of Slide1 longer
> > than the presentation of Slide1. And it works if you look at the data
> > of the results-file, the accuracies (Slide1.acc) and given responses
> > (Slide1.resp) are correct. But if we tried to send a trigger depending
> > on Slide1.acc, it doesn’t work if the reaction time is longer than the
> > duration (but in the result file Slide1.acc is correct even if the
> > reaction time is longer than the duration of Slide1). (Of course, we
> > took care that we use Slide1.Acc only when the time limit for reaction
> > is over).
>
> > Our hypothesis is that if you use a time limit for reactions longer
> > than the presentation of a slide (and the reaction is given after the
> > duration of the slide but during the allowed time for reaction), then
> > the properties Slide1.acc and Slide1.resp will be set at the end of a
> > trial into the list. So Slide1.acc is not available during the trial
> > (and so we can’t use it for sending a trigger). What would help us is
> > some code that set the properties Slide1.acc and Slide1.resp into the
> > list before ending a trial or a code that we can use the properties
> > Slide1.acc and Slide1.resp of the previous trial (then we would sent
> > the current accuracy during the next trial). And perhaps there is also
> > a solution we not thought about…
>
> > Thanks for any help,
> > Max
>
> > On 10 Feb., 13:27, Maria <maria.richter.m... at googlemail.com> wrote:
>
> > > Good morning!
>
> > > We have the following strange problem:
>
> > > We sat the "time limit" for reactions (=2000ms) longer than the
> > > presentation time of the slide1(1000ms) (at which should be reacted
> > > and which was followed by a slide2 (1000ms)) and asked in a later
> > > inline
>
> > > -----------------
>
> > > if c.getattrib("Slide1.acc") = 1 then
>
> > > l=3
>
> > > end if
>
> > > if c.getattrib("Slide1.acc") = 0 then
>
> > > l=4
>
> > > end if
> > > debug.print l
>
> > > ---------------------------
>
> > > In trials, where the reaction time was longer than the presentation
> > > time of the first slide (reaction during slide2) the "l" was set 4,
> > > even in correct cases.
>
> > > We need the accuracyinformation at this point of the experiment to
> > > send triggers.
>
> > > Is there anyone who had the same problem before?
> > > Thanks for any answer, we have no idea...
>
> > > Best wishes,
>
> > > Max

-- 
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.