Print

Print
I suspect it will not be very different from what you use for thresholding but the best way to find out is to plot your single trial data and look at typical peak to peak ranges in trials with artefacts and without. An alternative way specific to eyeblinks would be to first do eyeblink detection in the 'Mark' mode on continuous data (see chapter 41 in the manual) either on an EOG channel if you have one or on a frontal channel with clear eyeblinks and then reject trials containing eyeblinks after epoching.

Best,

Vladimir

On Thu, Jun 9, 2016 at 4:41 PM, Emily Hird <[log in to unmask]> wrote:

Dear Vladimir,

 

Thanks for your response.

I’m looking to remove eyeblinks from my data. What sort of threshold would you suggest to remove eyeblinks using the ‘Peak to Peak amplitude’ method?

 

Thanks,

Emily

 

From: Vladimir Litvak [mailto:[log in to unmask]]
Sent: 09 June 2016 16:01
To: Emily Hird
Cc: [log in to unmask]
Subject: Re: [SPM] Bad channel threshold ERP data

 

Dear Emily,

 

It might be a better idea instead of 'Threshold channels' method use the 'Peak to Peak amplitude' method which is not sensitive to the absolute value of your data but just do differences within a trial. It all depends on what kind of artefacts you actually have so you might want to also look at your data first.

 

Best,

 

Vladimir

 

On Thu, Jun 9, 2016 at 3:24 PM, Emily Hird <[log in to unmask]> wrote:

Dear SPM users,

 

I am preprocessing ERPs to look for a slow-wave anticipatory potential, so I have removed the highpass filter from my pipeline to ensure this slow-wave activity isn’t cleaned out of the data.

 

However, this leads to more channels being marked as ‘bad’ during artefact rejection with a threshold of 0.2 (this threshold is suggested in the manual). Is it advisable to change the ‘bad channel’ threshold from 0.2 to a higher value, so fewer channels are marked as bad when looking at data which hasn’t been highpass filtered?

 

Many thanks in advance.

Emily Hird

PhD student

School of Psychological Sciences

Zochonis Building, Brunswick Street, Manchester M13 9PT