[Eeglablist] How to compute the CDA in EEGLAB (contralateral delay activity)

[Massimiliano Papera]

es, I am not sure whether or not those formulas are correct. I found this
on the ERP tutorial (see below). Now according to this bit of the tutorial
the formulae we used should be good to get Lateralised readiness potential
waveforms. As far as I know this are waveforms down to motor responsed. Not
sure if you they can be down to CDA. I mean is LRP equal to CDA? if so,
then the formula below has to be right to be used to compute CDA. If not,
then I ve got no clue how to compute the cda.

MP

Specifying subsets of channels

It is also possible to specify different subsets of channels for the
different bins being used in a given equation.  For example, if you are
trying to compute a lateralized readiness potential (LRP) waveform, you
will want to subtract trials with an ipsilateral response from trials with
a contralateral response, averaged across the left and right hemispheres.
That is, you want to make a contralateral average (left-hemisphere
electrode sites for right-hand responses averaged with right-hemisphere
electrode sites for left-hand responses) and an ipsilateral average
(left-hemisphere sites for left-hand responses averaged with
right-hemisphere sites for right-hand response) and compute the difference.

To do this in Bin Operations, you first define sets of electrodes
corresponding to the left and right hemispheres.  Imagine that the
electrode sites 1-6 are as follows 1=F3, 2=F4, 3=C3, 4=C4, 5=P3, 6=P4.  And
imagine that Bin 1 corresponds to left-hand responses whereas Bin 2
corresponds to right-hand responses.  The following shows how you would
define left- and right-hemisphere groups and use them to form a
contralateral waveform (new Bin 1), an ipsilateral waveform (new Bin 2),
and a contra-minus-ipsi difference wave (new Bin 3):

LH = [1 3 5]

RH = [2 4 6]

b1 = (b2 at LH + b1 at RH)/2 label Average Contra

b2 = (b1 at LH + b2 at RH/2 label Average Ipsi






b3 = ((b2 at LH + b1 at RH)/2) – ((b1 at LH + b2 at RH)/2) label Average Contra Minus
Ipsi


On Tue, Nov 6, 2012 at 9:49 AM, Gunseli, E. <e.gunseli at vu.nl> wrote:

>  Dear Massimiliano,****
>
> ** **
>
> Probably b1 includes the trials at which the memory item is presented on
> the left visual field and b2 is when it is presented on the right visual
> hemifield.****
>
> Then the CDA should be calculated with the below formula:****
>
> nbin1= ((b2 at LH + b1 at RH)/2 - (b1 at LH + b2 at RH)/2 ) label CDA  ****
>
> ** **
>
> I would be glad if some expert member would confirm the formula.****
>
> ** **
>
> By the way, including only the trials with correct behavioral response
> would be a better idea.****
>
> ** **
>
> Best regards,****
>
> Eren Gunseli ****
>
> ** **
>
> *From:* eeglablist-bounces at sccn.ucsd.edu [mailto:
> eeglablist-bounces at sccn.ucsd.edu] *On Behalf Of *Massimiliano Papera
> *Sent:* Monday, November 05, 2012 17:59
> *To:* eeglablist at sccn.ucsd.edu
> *Subject:* [Eeglablist] How to compute the CDA in EEGLAB (contralateral
> delay activity)****
>
> ** **
>
> Dear All,****
>
>  ****
>
> I am investigating the relationship between Attention and Memory and would
> like to use the Contralateral Delay Activity (CDA) as a correlate of the
> visual working memory. I am dealing with the bin operations function.****
>
> I would like to check through the Discussion list if anyone who is
> familiar with the CDA knows I am using the right coding to get my CDA
> component across three different conditions.****
>
> I am investigating 3 different pair of electrodes (p3-4, o1-2 and p7-8)
> for correct and incorrect responses. However I am not sure if I need to
> filter out the incorrect trials.****
>
>  ****
>
> Thank you****
>
> ERP = pop_binoperator( ERP, {'LH = [18 23 24]' 'RH = [14 28 27]' 'nb1=
> b1 at RH-b1@LH label CDA LVF20' 'nb2= b2 at LH-b2@RH label CDA RVF20' 'nb3=
> b3 at RH-b3@LH label CDA LVF40' 'nb4= b4 at LH-b4@RH label CDA RVF40' 'nb5=
> b5 at RH-b5@LH label CDA LVF22' 'nb6= b6 at LH-b6@RH label CDA RVF22'});****
>
>    ****
>
> This gives me 6 bins, 2 per each of the 3 conditions. Each pair gives me
> erp for either left visual field (LVF) or right visual field (RVF).****
>
>  ****
>
> To get the CDA I now subtract  the activity between left and right visual
> field trials for all of the 3 conditions:****
>
> ERP = pop_binoperator( ERP, {'nbin1= b1-b2 label CDA 2T0D' 'nbin2=b3-b4
> label CDA 4T0D' 'nbin3=b5-b6 label CDA 2T2D' }); ****
>
>  ****
>
> This gives me 3 CDA component, one for each condition.****
>
>  ****
>
> Does anyone know if this is correct to compute the CDA?****
>
>  ****
>
> Thank you ****
>
>  ****
>
> Thanks to everyone****
>
> Massimiliano Papera****
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20130524/20b18026/attachment.html>