bci2000.org BBS

Hi all,

What is an appropriate epoch length to use for experiments in detecting p300? Further, what is an appropriate rationale for selecting the epoch length? Should it be longer than the maximal p300 time? (i.e. > 600ms)

Thanks,

Doug Davies

Hey,

You want to have an epoch that is long enough so that all important P300 components are in it. On the other hand, if you make the epoch longer, you are making it harder for the classifier to find good features. Overall, 800ms should be sufficient (I think this is the default in the P300Classifier). P300 latencies vary considerably, and you may have important information in periods prior to or after the standard P300.

Gerv