[Eeglablist] Will problems arise with using non-integer Event latencies? (UNCLASSIFIED)

[Hairston, W. Dave (Civ, ARL/HRED)]

Classification:  UNCLASSIFIED 
Caveats: NONE

I see that when data containing event latency information is imported, it is
automatically changed to being referenced by the sample #
(EEG.event.latency) rather than the original time value (in ms).
It's stored as a whole-number integer.
In some cases, this means that it is rounded (b/c the system actually
recorded/reports with millisecond level accuracy, not the system sample
rate) and some accuracy is lost, especially with low-sample rate systems.

Is there any harm/potential problem with using non-integer values in the
EEG.event.latency structure? 
This is especially important when trying to account for temporal drift
between the original EEG system and an external device used for the event
logging.
We have run a handful of ERP analyses (using the ERPlab plugin) with
non-integer values for the event latencies, and they seem to work OK -
however I don't know if there are other, specific applications where it
could be a problem.

Thanks,
Dave

-------------------------------
Dr. W. David Hairston

Multi-Sensory Research Team, 
Translational Neuroscience, HRED
Army Research Lab
Aberdeen Proving Ground, MD 21005

Office: (410) 278-5925
Fax:    (410) 278-8828
Classification:  UNCLASSIFIED 
Caveats: NONE

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 6438 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20110512/dd9e2357/attachment.bin>