[Eeglablist] time dilation on Tucker-Davis-via-neuroscan CNT import

[Ross Fulham]

Hamish,

Neuroscan CNT files store sampling rates as a 16 bit integer quantity (in Hz). So the problem with sampling rates being rounded down is occurring when you convert the data to Neuroscan, not when you import the neuroscan file into EEGLAB.

Ross 




------------------------------------------------------------
Dr Ross Fulham
Research Officer 
Room 1025, Ph: 492 46636, Fax: 492 46608
Centre For Mental Health Studies
James Fletcher Hospital
Newcastle, NSW. Australia.

Functional Neuroimaging Laboratory,
Discipline of Psychology
School of Behavioural Sciences, 
The University of Newcastle.

e-mail Ross.Fulham at newcastle.edu.au
------------------------------------------------------------
>>> Hamish Innes-Brown <hinnesbrown at gmail.com> 02/15/08 4:37 PM >>>
Hi all, I am having a strange problem on importing continuous EEG where the
time of events (and the whole EEG) is stretched a little bit on importing
the data.

It's a bit of a convoluted process getting the data from a TDT system into
EEGlab, what we do is this:

1) Export the channel data from the TDT tanks (to CSV files)
2) Read the CSV data into Neuroscan and add the events from the TDT event
log
3) Load the CNT file in to EEGlab using pop_loadcnt

The weird thing is that the data is all slightly "slowed down" when read
into EEGlab - events and EEG features are delayed in time.  I think this
might be a problem with how a sampling rate that's not a multiple of one is
read from the CNT file.  The TDT sampling rate is 488.28125, and once read
into EEGlab, the EEG.srate value has dropped all the decimals and become
"488".

Has anyone else had problems importing data with such strange sampling
rates?

I'm aware there is a direct TDT import plugin but so far I haven't been able
to get that to work (I think our TDT installation is missing some vital
parts), and we also use the Neuroscan CNT files for another parallel
analysis...

Thanks.


-hamish-