Laurent,

Thanks for this. scoring it is a bit complicated, and I'll have to dive
into the code to understand it, but I'll try to work out a fix for it.

Shane

On Wed, 2016-03-30 at 11:30 +0000, Schüpbach Laurent wrote:
> Dear Shane and Chris,
> 
> 
> 
> The problem is with these lines :
> 
>    corr1 <- Nth(des,1)==0
> 
>    corr2 <- Nth(des,2)==0
> 
> 
> 
> Strangely, the variable saved will be according to the previous test
> (last trial). Printing corr1 for each trial before and after the if
> made it clear. We don’t see this pattern for the hits as this test is
> overwritten during the “if”. Another peculiar behavior is that
> sometimes, corr on the very first trial is 0. As it isn’t systematic,
> I have a really hard time figuring what happens. I have been using
> such y <- x==0 in other scripts and it seems to work as expected for
> me.
> 
> 
> 
> Shane, this is highly disturbing. If statements like y <- x==0 are
> not always giving the same behavior, then it’s important to know we
> shouldn’t use them anymore. Or at least know in which circumstances
> it happens.
> 
> 
> 
> Best regards,
> 
> Laurent Schüpbach
> 
> 
> 
> 
> 
> From: Shane Mueller<mailto:[email protected]>
> Sent: Friday, March 25, 2016 04:26
> To: Christopher Higginson<mailto:[email protected]>; 
> [email protected]<mailto:
> [email protected]>
> Subject: Re: [Pebl-list] n-Back
> 
> 
> 
> Chris,
> I believe that somebody reported that to me recently, so I don't
> think
> you are incorrect.  I haven't verified this yet, but it if it isn't
> correct, then columns may be mislabeled.
> 
> 
> Here is the logic--maybe you have a fix?  the variable corr1 (and
> corr2) starts out with a default value that should be correct if no
> response were made:
> 
>    corr1 <- Nth(des,1)==0
>    corr2 <- Nth(des,2)==0
> 
> 
> It sounds like this
> 
> Then there are two blocks of code that record whether each of two
> possible responses are correct.  The second one is this, which would
> be
> the code that gets tested when the second response times out after
> the
> first response is complete:
> 
>      if(resp2 == "<lshift>")
>        {
>           Show(respLeft)
>           corr1 <- Nth(des,1)
>        }elseif(resp2=="<rshift>")
>        {
>           Show(respRight)
>           corr2 <- Nth(des,2)
>        }
> 
> Here, neither of these would run if there were a timeout, and so
> corr1
> would default to the original value above. So maybe the original
> default score is backward?  I think in des, it is marked a 0 if there
> is no n-back match and 1 if there is, so if no response is made, it
> should be marked as correct in that case; so I'm a bit puzzled what
> the
> happening.  If you can send me an example data file and tell me what
> the incorrect coding is, that would be helpful.
> 
> Shane
> 
> On Tue, 2016-03-22 at 15:36 +0000, Christopher Higginson wrote:
> > Dear PEBL Project,
> > I'm in the process of revising the n-Back task from the battery to
> > better meet my needs. I will certainly pass along my edited version
> > of the task when it is complete. I'm concerned about a potential
> > error in the data file that is created. The column "corr1" (which
> > appears to indicate whether the response is correct or not) seems
> > to
> > generate a score of "0" when a correct "timeout" (i.e., no
> > response)
> > response is made immediately after a "lshift" response. Has anyone
> > else noticed this or am I missing something? Any assistance would
> > be
> > much appreciated!
> > Chris
> > 
> > 
> > 
> > Christopher I. Higginson, Ph.D.
> > Associate Professor
> > Department of Psychology
> > Loyola University Maryland
> > 4501 North Charles Street
> > Baltimore, MD 21210
> > 410-617-2461
> > 
> > Disclaimer: This electronic message may contain information that is
> > proprietary, confidential, or legally privileged or protected. It
> > is
> > intended only for the use of the individual(s) and entity named in
> > the message. If you are not an intended recipient of this message,
> > please notify the sender immediately and delete the material from
> > your computer. Do not deliver, distribute or copy this message and
> > do
> > not disclose its contents or take any action in reliance on the
> > information it contains.
> > -------------------------------------------------------------------
> > --
> > ---------
> > Transform Data into Opportunity.
> > Accelerate data analysis in your applications with
> > Intel Data Analytics Acceleration Library.
> > Click to learn more.
> > http://pubads.g.doubleclick.net/gampad/clk?id=278785351&iu=/4140
> > _______________________________________________
> > Pebl-list mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/pebl-list
> 
> ---------------------------------------------------------------------
> ---------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785351&iu=/4140
> _______________________________________________
> Pebl-list mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/pebl-list
> 
> ---------------------------------------------------------------------
> ---------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140
> _______________________________________________
> Pebl-list mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/pebl-list

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140
_______________________________________________
Pebl-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/pebl-list

Reply via email to