Hi all
We're running a study using auditory stimuli. The latency between the
trigger code and actual auditory onset is rather long and somewhat
variable, so we have a device that sends a '1' trigger when the sound is
actually played. Thus in our event log file we have the unique trigger
codes for each condition, followed ~100 ms later by the 1 that marks the
actual stimulus event. I would like to recode each '1' according to the
condition code that preceded it.
Thus far, I have the following:
# trigger codes in the event log (note there are actually 32 unique entries
in the dict; have shortened here for brevity)
event_id = {'Match/Det/F/con':10, 'Match/N/F/con':30, 'Match/Det/M/con':20,
'Match/N/M/con':40}
# new trigger codes to replace the 1s
new_event_id = {'Match/Det/F/con':110, 'Match/N/F/con':130,
'Match/Det/M/con':120, 'Match/N/M/con':140}
raw = mne.io.read_raw_eeglab(raw_fname, eog=eog_inds, event_id=event_id,
preload=True)
events = mne.find_events(raw, stim_channel='STI 014')
code_tmin = -.15
code_tmax = -.01
sound_onset = 1 # code for actual sound onset
for key, value in event_id.items():
events_tmp, lag_tmp = mne.event.define_target_events(events,
sound_onset, value, raw.info['sfreq'], code_tmin, code_tmax,
new_id=new_event_id[key], fill_na=None)
...where I'm stuck is figuring out how to either modify the original
"events" array to replace the 1s with the new codes after eery call to
define_target_events in the loop, or alternatively concatenate the
events_tmp arrays I create inside the for loop as I go. In the latter case,
I'm not sure if the resulting concatenated array would also need to be
sorted by column 0 so that it reflects the true temporal sequence of the
events?
The define_target_events example in the gallery was great as far as it
goes, but it seems to assume that one would only ever want to do this kind
of replacement once/for one condition prior to epoching. I need to do it 32
times.
thanks in advance!
Aaron