[Eeglablist] Epoch multi-word ERP

Stephen Politzer-Ahles spa268 at nyu.edu
Wed Apr 15 09:20:51 PDT 2015

Hello Isaiah,

As far as I know, variable length epochs are not possible in EEGLAB
proper, but they are possible in Fieldtrip, and might be possible in
some EEGLAB plugins (maybe ERPLAB)?

Another thing that is sometimes done is to just use a fixed epoch
length that is long enough to encompass even the longest


On 4/13/15, Isaiah Innis <isainnis at umail.iu.edu> wrote:
> Hello eeglablist:
> We are planning a language ERP study exploring object versus subject
> relative clauses. The plan is to examine both word-by-word and whole
> sentence length processing, following the example set in this paper: (
> http://kutaslab.ucsd.edu/people/kutas/pdfs/1995.JCN.376.pdf). Stimuli are
> sentences presented one word at a time, followed by a comprehension probe
> sentence (subjects have up to 6000 msec to respond).
> The problem is, that paper provides only a gross description of how they
> epoched their EEG data. This has left us with a few questions:
> 1. Is there a way to extract epochs from sentences with different lengths?
> Some trials have 12 word sentences, others 13 words etc. . .
> 2. Since the subject may take up to 6000msec in their response, is it
> possible to have variable epoch endpoints based on when they responded?
> In general, our questions are about how to deal with variance in the
> stimulus/trial length when epoching. The above paper apparently handled
>  such differences in test stimuli length, but does not describe how they
> did so.
> Any insight or suggestions would be appreciated.
> Thank you for your time,
> --
> Isaiah Innis
> Indiana University '13
> EEG Technician, IUB IRF


Stephen Politzer-Ahles
New York University, Abu Dhabi
Neuroscience of Language Lab

More information about the eeglablist mailing list