Plot single trial activity, grouped by ROI and sorted by RT

This will produce what is sometimes called an event related potential / field (ERP/ERF) image.

The EEGLAB example file - containing an experiment with button press responses to simple visual stimuli - is read in and response times are calculated. Regions of Interest are determined by the channel types (in 10/20 channel notation, even channels are right, odd are left, and ‘z’ are central). The median and the Global Field Power within each channel group is calculated, and the trials are plotted, sorting by response time.

# Authors: Jona Sassenhagen <jona.sassenhagen@gmail.com>
#
# License: BSD (3-clause)

import mne
from mne.event import define_target_events
from mne.channels import make_1020_channel_selections

print(__doc__)

Out:


Load EEGLAB example data (a small EEG dataset)

data_path = mne.datasets.testing.data_path()
fname = data_path + "/EEGLAB/test_raw.set"
montage = data_path + "/EEGLAB/test_chans.locs"

event_id = {"rt": 1, "square": 2}  # must be specified for str events
eog = {"FPz", "EOG1", "EOG2"}
raw = mne.io.read_raw_eeglab(fname, eog=eog, montage=montage,
                             stim_channel=False)
events = mne.events_from_annotations(raw, event_id)[0]

Out:

Reading /home/circleci/mne_data/MNE-testing-data/EEGLAB/test_raw.fdt
Used Annotations descriptions: ['rt', 'square']

Create Epochs

# define target events:
# 1. find response times: distance between "square" and "rt" events
# 2. extract A. "square" events B. followed by a button press within 700 msec
tmax = .7
sfreq = raw.info["sfreq"]
reference_id, target_id = 2, 1
new_events, rts = define_target_events(events, reference_id, target_id, sfreq,
                                       tmin=0., tmax=tmax, new_id=2)

epochs = mne.Epochs(raw, events=new_events, tmax=tmax + .1,
                    event_id={"square": 2})

Out:

73 matching events found
Applying baseline correction (mode: mean)
Not setting metadata
0 projection items activated

Plot using GFP

# Parameters for plotting
order = rts.argsort()  # sorting from fast to slow trials

selections = make_1020_channel_selections(epochs.info, midline="12z")

# The actual plots (GFP)
epochs.plot_image(group_by=selections, order=order, sigma=1.5,
                  overlay_times=rts / 1000., combine='gfp',
                  ts_args=dict(vlines=[0, rts.mean() / 1000.]))
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_001.png
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_002.png
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_003.png

Out:

Loading data for 73 events and 129 original time points ...
0 bad epochs dropped
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped

Plot using median

epochs.plot_image(group_by=selections, order=order, sigma=1.5,
                  overlay_times=rts / 1000., combine='median',
                  ts_args=dict(vlines=[0, rts.mean() / 1000.]))
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_004.png
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_005.png
  • ../../_images/sphx_glr_plot_roi_erpimage_by_rt_006.png

Out:

Loading data for 73 events and 129 original time points ...
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped
73 matching events found
No baseline correction applied
Not setting metadata
0 projection items activated
0 bad epochs dropped

Total running time of the script: ( 0 minutes 3.857 seconds)

Estimated memory usage: 8 MB

Gallery generated by Sphinx-Gallery