[Eeglablist] time window fractional area latency

[Kathrin Sadus]

Hi everyone,

I am trying to determinethe P300 peak latency scores in three different 
tasks. However, when using the fractional area latency (50%), I am not 
sure how to definethe measurement window. I looked into various studies 
and found different approches, but it seems that most researchers would 
simply take the grand average and determine the window by appearance. 
However, as I found different approaches I wanted to ask if there is 
some sort of gold standard.

Specifically I have two questions:

1) Is it usual or correct to set the window in the peak of the preceding 
component (the N2) or should one rather try to set the window so that it 
is on the flanks of the P3?

2) In some studies the onset of the area is determined by calculating 
50% of the peak-to-peak amplitude (and the offset is set to the same 
level).  I wonder if anyone has experience with whether this approach 
works for more complex tasks where the signal-to-noise ratio is lower 
(Nback, Shifting Task etc).

Thanks in advance.

Best,

Kathrin