[Eeglablist] Contralateral and Ipsilateral waveforms

[Dwight Peterson]

Hi all,


        For those of you familiar with the CDA, please help if possible.
 I'm working on a project exploring the CDA for visual working memory and
my data thus far show shockingly similar waveforms for contralateral and
ipsilateral activity (my understanding is contralateral should be much
higher in amplitude and negative going).  I have 4 different conditions in
my experiment and I'm am including in my analysis only activity associated
with CORRECT trials.  I have 7 channels with 6 electrode sites (Channel 1
corresponds to a photo-diode that I use to mark the onset of my
stimulus/trial).  I have these few sites b/c I'm working with an 8 channel
"home-brew" sort of EEG system.   Ch 2 = P7, Ch 3 = P8, Ch 4 = PO7, Ch 5 =
PO8, Ch 6 = P1, Ch 7 = P2.

         I've created an event list that creates 8 bins that correspond to
Left Hemifield trials and Right Hemifield trials for each of my 4
conditions that correspond to my event type (unique identifier values) as
shown below:

   70       "ss_two_r"     1   "two_objects_RVHF"

   71       "ss_two_l"     2   "two_objects_LVHF"

   80     "ss_three_r"     3 "three_objects_RVHF"

   81     "ss_three_l"     4 "three_objects_LVHF"

   90 "ss_three_apart_r"     5 "three_objects_apart_RVHF"

   91 "ss_three_apart_l"     6 "three_objects_apart_LVHF"

   94 "ss_three_prox_r"     7 "three_objects_prox_RVHF"

   95 "ss_three_prox_l"     8 "three_objects_prox_LVHF"

Then after epoching, running artifact detection, and computing averaged ERP
waveforms for each of these bins, I created a new ERPset from Bin
Operations using the Independent Transformations "mode" with the following
formulas to extract the contralateral and ipsilateral waveforms:

LH = [2 4 6]

RH = [3 5 7]

nb1 = 0.5*b1 at LH + 0.5*b2 at RH label average contra two objects

nb2 = 0.5*b1 at RH + 0.5*b2 at LH label average ipsi two objects

nb3 = 0.5*b3 at LH + 0.5*b4 at RH label average contra three objects

nb4 = 0.5*b3 at RH + 0.5*b4 at LH label average ipsi three objects

nb5 = 0.5*b5 at LH + 0.5*b6 at RH label average contra three apart

nb6 = 0.5*b5 at RH + 0.5*b6 at LH label average ipsi three apart

nb7 = 0.5*b7 at LH + 0.5*b8 at RH label average contra three prox

nb8 = 0.5*b7 at RH + 0.5*b8 at LH label average ipsi three prox


Then I get the CDA difference waveforms for each condition by recursively
updating the new Contra Ipsi ERPset by using the following formula

b9 = b1-b2 label CDA Two Objects

b10 = b3-b4 label CDA Three Objects

b11 = b5-b6 label CDA Three Apart

b12 = b7-b8 label CDA Three Prox

       Perhaps I am just seeing an extremely NULL result and lack of a
difference between contra and ipsi waveforms for each condition, but I've
tried a variety of things and double checked these formulas (which seem to
be correct given my event types and bins, etc.).    Impedances are below 5 KΩ
 and aside from the lack of a difference between contra and ipsi, the data
looks pretty clean.

For anybody who has done this type of research before, do you have any
thoughts?

Thanks so much!!!


--Dwight--

-- 
Dwight J. Peterson, M.A.
Graduate Assistant
Cognitive and Brain Sciences
Department of Psychology/296
Memory and Brain Laboratory
University of Nevada, Reno
Reno, NV 89557
Cellular: 319-231-8596
Laboratory: 775-682-8667
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20130307/e0549b2d/attachment.html>