[Eeglablist] Filter roll-off (not just for Butterworth filters)

[Andreas Widmann]

Dear all,

Steve Luck wrote:
> In response to the question about filter order for a Butterworth  
> filter, it is important for EVERYONE doing EEG/ERP research to realize  
> that a steeper roll-off will mean that the filter will cause more  
> temporal distortion of the waveform.

In agreement with Prof. Luck I would like to note that even if steeper
filters indisputably spread filter distortions wider in the time domain
(temporal imprecision) they might still be preferred to more gradual
ones in particular situations, e.g. when they allow for wider passbands.

If, for example, highpass filtering of ERPs can not be avoided, steeper
filters allow for lower cutoff frequencies and might result in
distortions of smaller amplitude (but wider distribution).

I second the approach to pass known signals, simulating the data,
through the filter to see the resulting distortions (implied by any kind
of filtering).

Best regards,
Andreas