bci2000.org BBS

Hi all,

I was wondering if anyone can provide a suggestion on an optimal ISI for experiments in detecting P300.

I have been reading various papers on P300, and I've noticed that this parameter seems to vary widely, from several seconds (2700ms in one paper) to 100-200ms in other cases. Is it acceptable for the sum of the ISI and the StimulusDuration to be smaller than the 700-800ms window in which I am detecting the P300 response?

Thanks,

Doug Davies

Successive stimuli can definitely overlap with the P300, i.e., the ISI can be less than 700 ms. In healthy subjects, you can go quite fast, up to perhaps 12 Hz (80 ms). In patients, the ISI should be longer.

I think there was a Sellers, Wolpaw et al., study on the effect of matrix size and ISI.

Gerv