Epochs: Autoreject cleaning
Autoreject was run to produce cleaner epochs before fitting ICA. 77 epochs were rejected because more than {'eeg': 2} channels were bad (cross-validated n_interpolate limit; excluding globally bad and non-data channels, shown in white). Note that none of the blue segments were actually interpolated before submitting the data to ICA. This is following the recommended approach for ICA described in the the Autoreject documentation.
Number of events 83
Events stimulus/car/normal: 48
stimulus/face/normal: 35
Time range -0.203 – 0.797 s
Baseline off
Epoch #
event_name
response/correct
response/error
stimulus/car/normal
stimulus/car/scrambled
stimulus/face/normal
stimulus/face/scrambled
Loading, please wait
Epoch #
event_name
response/correct
response/error
stimulus/car/normal
stimulus/car/scrambled
stimulus/face/normal
stimulus/face/scrambled
0stimulus/car/normal0.5620.000
1stimulus/car/normal0.4380.000
3stimulus/face/normal0.3750.000
5stimulus/face/normal0.3280.000
7stimulus/car/normal0.5310.000
9stimulus/car/normal0.4060.000
14stimulus/car/normal0.5160.000
20stimulus/car/normal0.5230.000
22stimulus/face/normal0.4220.000
23stimulus/face/normal0.3750.000
25stimulus/car/normal0.4840.000
26stimulus/car/normal0.4450.000
28stimulus/face/normal0.4690.000
30stimulus/face/normal0.3520.000
32stimulus/car/normal0.5230.000
35stimulus/car/normal0.3750.000
38stimulus/car/normal0.3200.000
39stimulus/car/normal0.2730.000
43stimulus/car/normal0.5310.000
45stimulus/car/normal0.4920.000
46stimulus/face/normal0.4530.000
48stimulus/car/normal0.5620.000
49stimulus/face/normal0.5310.000
50stimulus/car/normal0.6330.000
51stimulus/face/normal0.5550.000
54stimulus/face/normal0.4450.000
57stimulus/car/normal0.4140.000
58stimulus/face/normal0.3980.000
60stimulus/face/normal0.3590.000
61stimulus/car/normal0.3910.000
63stimulus/car/normal0.3050.000
64stimulus/car/normal0.2810.000
66stimulus/car/normal0.5390.000
68stimulus/face/normal0.3910.000
69stimulus/car/normal0.4770.000
74stimulus/car/normal0.5860.000
75stimulus/car/normal0.5700.000
78stimulus/car/normal0.5620.000
79stimulus/face/normal0.4380.000
80stimulus/face/normal0.3750.000
81stimulus/car/normal0.5470.000
82stimulus/car/normal0.4610.000
84stimulus/car/normal0.6020.000
85stimulus/car/normal0.4300.000
87stimulus/face/normal0.4300.000
89stimulus/car/normal0.3280.000
90stimulus/car/normal0.4220.000
93stimulus/face/normal0.3520.000
94stimulus/car/normal0.3910.000
97stimulus/car/normal0.3980.000
98stimulus/face/normal0.4610.000
100stimulus/face/normal0.2890.000
101stimulus/car/normal0.3830.000
103stimulus/face/normal0.4920.000
104stimulus/face/normal0.4530.000
106stimulus/face/normal0.6020.000
107stimulus/face/normal0.4220.000
108stimulus/face/normal0.4220.000
109stimulus/face/normal0.3830.000
110stimulus/face/normal0.4690.000
112stimulus/face/normal0.4690.000
113stimulus/car/normal0.3670.000
114stimulus/face/normal0.6170.000
115stimulus/car/normal0.5310.000
116stimulus/face/normal0.5780.000
118stimulus/face/normal0.2500.000
119stimulus/face/normal0.3360.000
123stimulus/car/normal0.5080.000
126stimulus/car/normal0.4690.000
127stimulus/car/normal0.4220.000
128stimulus/car/normal0.2890.000
129stimulus/face/normal0.4220.000
130stimulus/car/normal0.5000.000
131stimulus/car/normal0.3590.000
132stimulus/car/normal0.3280.000
134stimulus/face/normal0.3830.000
136stimulus/car/normal0.3750.000
139stimulus/car/normal0.4140.000
144stimulus/car/normal0.4450.000
145stimulus/car/normal0.5860.000
147stimulus/face/normal0.3980.000
150stimulus/car/normal0.3750.000
152stimulus/face/normal0.4770.000

83 rows × 8 columns

Drop log
PSD
PSD calculated from 83 epochs (83.0 s).
  """ERP CORE.

This example demonstrate how to process 5 participants from the
[ERP CORE](https://erpinfo.org/erp-core) dataset. It shows how to obtain 7 ERP
components from a total of 6 experimental tasks:

- N170 (face perception)
- MMN (passive auditory oddball)
- N2pc (visual search)
- N400 (word pair judgment)
- P3b (active visual oddball)
- LRP and ERN (flankers task)

## Dataset information

- **Authors:** Emily S. Kappenman, Jaclyn L. Farrens, Wendy Zhang,
                       Andrew X. Stewart, and Steven J. Luck
- **License:** CC-BY-4.0
- **URL:** [https://erpinfo.org/erp-core](https://erpinfo.org/erp-core)
- **Citation:** Kappenman, E., Farrens, J., Zhang, W., Stewart, A. X.,
                & Luck, S. J. (2021). ERP CORE: An open resource for human
                event-related potential research. *NeuroImage* 225: 117465.
                [https://doi.org/10.1016/j.neuroimage.2020.117465](https://doi.org/10.1016/j.neuroimage.2020.117465)
"""

import argparse
import sys

import mne

bids_root = "~/mne_data/ERP_CORE"
deriv_root = "~/mne_data/derivatives/mne-bids-pipeline/ERP_CORE"

# Find the --task option
args = [arg for arg in sys.argv if arg.startswith("--task") or not arg.startswith("-")]
parser = argparse.ArgumentParser()
parser.add_argument("ignored", nargs="*")
parser.add_argument(
    "--task", choices=("N400", "ERN", "LRP", "MMN", "N2pc", "N170", "P3"), required=True
)
task = parser.parse_args(args).task
sessions = [task]

subjects = ["015", "016", "017", "018", "019"]

ch_types = ["eeg"]
interactive = False

raw_resample_sfreq = 128
# Suppress "Data file name in EEG.data (sub-019_task-ERN_eeg.fdt) is incorrect..."
read_raw_bids_verbose = "error"

eeg_template_montage = mne.channels.make_standard_montage("standard_1005")
eeg_bipolar_channels = {
    "HEOG": ("HEOG_left", "HEOG_right"),
    "VEOG": ("VEOG_lower", "FP2"),
}
drop_channels = ["HEOG_left", "HEOG_right", "VEOG_lower"]
eog_channels = ["HEOG", "VEOG"]

l_freq = 0.1
h_freq = None
notch_freq = 60

decode = True
decoding_time_generalization = True
decoding_time_generalization_decim = 2

find_breaks = True
min_break_duration = 10
t_break_annot_start_after_previous_event = 3.0
t_break_annot_stop_before_next_event = 1.5

if task == "N400":  # test autoreject local without ICA
    spatial_filter = None
    reject = "autoreject_local"
    autoreject_n_interpolate = [2, 4]
elif task == "N170":  # test autoreject local before ICA
    spatial_filter = "ica"
    ica_reject = "autoreject_local"
    reject = "autoreject_global"
    autoreject_n_interpolate = [2, 4]
else:
    spatial_filter = "ica"
    ica_reject = dict(eeg=350e-6, eog=500e-6)
    reject = "autoreject_global"

# These settings are only used for the cases where spatial_filter="ica"
ica_max_iterations = 1000
ica_eog_threshold = 2
ica_decim = 2  # speed up ICA fitting

run_source_estimation = False
on_rename_missing_events = "ignore"

parallel_backend = "dask"
dask_worker_memory_limit = "2.5G"
n_jobs = 4

if task == "N400":
    dask_open_dashboard = True

    rename_events = {
        "response/201": "response/correct",
        "response/202": "response/error",
        "stimulus/111": "stimulus/prime/related",
        "stimulus/112": "stimulus/prime/related",
        "stimulus/121": "stimulus/prime/unrelated",
        "stimulus/122": "stimulus/prime/unrelated",
        "stimulus/211": "stimulus/target/related",
        "stimulus/212": "stimulus/target/related",
        "stimulus/221": "stimulus/target/unrelated",
        "stimulus/222": "stimulus/target/unrelated",
    }

    eeg_reference = ["P9", "P10"]
    epochs_tmin = -0.2
    epochs_tmax = 0.8
    epochs_metadata_tmin = 0
    epochs_metadata_tmax = 1.5
    epochs_metadata_keep_first = ["stimulus/target", "response"]
    baseline = (None, 0)

    conditions = {
        "related": '`first_stimulus/target` == "related" and '
        'first_response == "correct"',
        "unrelated": '`first_stimulus/target` == "unrelated" and '
        'first_response == "correct"',
    }
    contrasts = [("unrelated", "related")]
    cluster_forming_t_threshold = 1.5  # Only for testing!
    cluster_permutation_p_threshold = 0.2  # Only for testing!
elif task == "ERN":
    rename_events = {
        "stimulus/11": "compatible/left",
        "stimulus/12": "compatible/right",
        "stimulus/21": "incompatible/left",
        "stimulus/22": "incompatible/right",
        "response/111": "response/correct",
        "response/112": "response/incorrect",
        "response/121": "response/correct",
        "response/122": "response/incorrect",
        "response/211": "response/incorrect",
        "response/212": "response/correct",
        "response/221": "response/incorrect",
        "response/222": "response/correct",
    }

    eeg_reference = ["P9", "P10"]
    epochs_tmin = -0.6
    epochs_tmax = 0.4
    baseline = (-0.4, -0.2)
    conditions = ["response/correct", "response/incorrect"]
    contrasts = [("response/incorrect", "response/correct")]
    cluster_forming_t_threshold = 5  # Only for testing!
    cluster_permutation_p_threshold = 0.2  # Only for testing!
    decoding_csp = True
    decoding_csp_freqs = {
        "theta": [4, 7],
        "alpha": [8, 12],
        "beta": [13, 20, 30],
        "gamma": [50, 63],
    }
    decoding_csp_times = [-0.2, 0.0, 0.2, 0.4]
elif task == "LRP":
    rename_events = {
        "stimulus/11": "compatible/left",
        "stimulus/12": "compatible/right",
        "stimulus/21": "incompatible/left",
        "stimulus/22": "incompatible/right",
        "response/111": "response/left/correct",
        "response/112": "response/left/incorrect",
        "response/121": "response/left/correct",
        "response/122": "response/left/incorrect",
        "response/211": "response/right/incorrect",
        "response/212": "response/right/correct",
        "response/221": "response/right/incorrect",
        "response/222": "response/right/correct",
    }

    eeg_reference = ["P9", "P10"]
    epochs_tmin = -0.8
    epochs_tmax = 0.2
    baseline = (None, -0.6)
    conditions = ["response/left", "response/right"]
    contrasts = [("response/right", "response/left")]  # contralateral vs ipsi
elif task == "MMN":
    rename_events = {
        "stimulus/70": "stimulus/deviant",
        "stimulus/80": "stimulus/standard",
    }

    eeg_reference = ["P9", "P10"]
    epochs_tmin = -0.2
    epochs_tmax = 0.8
    baseline = (None, 0)
    conditions = ["stimulus/standard", "stimulus/deviant"]
    contrasts = [("stimulus/deviant", "stimulus/standard")]
elif task == "N2pc":
    rename_events = {
        "response/201": "response/correct",
        "response/202": "response/error",
        "stimulus/111": "stimulus/blue/left",
        "stimulus/112": "stimulus/blue/left",
        "stimulus/121": "stimulus/blue/right",
        "stimulus/122": "stimulus/blue/right",
        "stimulus/211": "stimulus/pink/left",
        "stimulus/212": "stimulus/pink/left",
        "stimulus/221": "stimulus/pink/right",
        "stimulus/222": "stimulus/pink/right",
    }

    eeg_reference = ["P9", "P10"]
    # Analyze all EEG channels -- we only specify the channels here for the purpose of
    # demonstration
    analyze_channels = [
        "FP1",
        "F3",
        "F7",
        "FC3",
        "C3",
        "C5",
        "P3",
        "P7",
        "P9",
        "PO7",
        "PO3",
        "O1",
        "Oz",
        "Pz",
        "CPz",
        "FP2",
        "Fz",
        "F4",
        "F8",
        "FC4",
        "FCz",
        "Cz",
        "C4",
        "C6",
        "P4",
        "P8",
        "P10",
        "PO8",
        "PO4",
        "O2",
    ]

    epochs_tmin = -0.2
    epochs_tmax = 0.8
    baseline = (None, 0)
    conditions = ["stimulus/right", "stimulus/left"]
    contrasts = [("stimulus/right", "stimulus/left")]  # Contralteral vs ipsi
elif task == "N170":
    rename_events = {
        "response/201": "response/correct",
        "response/202": "response/error",
    }

    eeg_reference = "average"
    # Analyze all EEG channels -- we only specify the channels here for the purpose of
    # demonstration
    analyze_channels = [
        "FP1",
        "F3",
        "F7",
        "FC3",
        "C3",
        "C5",
        "P3",
        "P7",
        "P9",
        "PO7",
        "PO3",
        "O1",
        "Oz",
        "Pz",
        "CPz",
        "FP2",
        "Fz",
        "F4",
        "F8",
        "FC4",
        "FCz",
        "Cz",
        "C4",
        "C6",
        "P4",
        "P8",
        "P10",
        "PO8",
        "PO4",
        "O2",
    ]

    ica_n_components = 30 - 1
    for i in range(1, 180 + 1):
        orig_name = f"stimulus/{i}"

        if 1 <= i <= 40:
            new_name = "stimulus/face/normal"
        elif 41 <= i <= 80:
            new_name = "stimulus/car/normal"
        elif 101 <= i <= 140:
            new_name = "stimulus/face/scrambled"
        elif 141 <= i <= 180:
            new_name = "stimulus/car/scrambled"
        else:
            continue

        rename_events[orig_name] = new_name

    epochs_tmin = -0.2
    epochs_tmax = 0.8
    baseline = (None, 0)
    conditions = ["stimulus/face/normal", "stimulus/car/normal"]
    contrasts = [("stimulus/face/normal", "stimulus/car/normal")]
elif task == "P3":
    rename_events = {
        "response/201": "response/correct",
        "response/202": "response/incorrect",
        "stimulus/11": "stimulus/target/11",
        "stimulus/22": "stimulus/target/22",
        "stimulus/33": "stimulus/target/33",
        "stimulus/44": "stimulus/target/44",
        "stimulus/55": "stimulus/target/55",
        "stimulus/21": "stimulus/non-target/21",
        "stimulus/31": "stimulus/non-target/31",
        "stimulus/41": "stimulus/non-target/41",
        "stimulus/51": "stimulus/non-target/51",
        "stimulus/12": "stimulus/non-target/12",
        "stimulus/32": "stimulus/non-target/32",
        "stimulus/42": "stimulus/non-target/42",
        "stimulus/52": "stimulus/non-target/52",
        "stimulus/13": "stimulus/non-target/13",
        "stimulus/23": "stimulus/non-target/23",
        "stimulus/43": "stimulus/non-target/43",
        "stimulus/53": "stimulus/non-target/53",
        "stimulus/14": "stimulus/non-target/14",
        "stimulus/24": "stimulus/non-target/24",
        "stimulus/34": "stimulus/non-target/34",
        "stimulus/54": "stimulus/non-target/54",
        "stimulus/15": "stimulus/non-target/15",
        "stimulus/25": "stimulus/non-target/25",
        "stimulus/35": "stimulus/non-target/35",
        "stimulus/45": "stimulus/non-target/45",
    }

    eeg_reference = ["P9", "P10"]
    epochs_tmin = -0.2
    epochs_tmax = 0.8
    baseline = (None, 0)
    conditions = ["stimulus/target", "stimulus/non-target"]
    contrasts = [("stimulus/target", "stimulus/non-target")]
    cluster_forming_t_threshold = 0.8  # Only for testing!
    cluster_permutation_p_threshold = 0.2  # Only for testing!
else:
    raise RuntimeError(f"Task {task} not currently supported")

  Platform             Linux-5.15.0-1053-aws-x86_64-with-glibc2.35
Python               3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
Executable           /home/circleci/python_env/bin/python3.10
CPU                  x86_64 (36 cores)
Memory               68.6 GB

Core
├☑ mne               1.7.0.dev156+g415e7f68e (devel, latest release is 1.6.1)
├☑ numpy             1.26.4 (OpenBLAS 0.3.23.dev with 2 threads)
├☑ scipy             1.12.0
└☑ matplotlib        3.8.3 (backend=agg)

Numerical (optional)
├☑ sklearn           1.4.1.post1
├☑ numba             0.59.1
├☑ nibabel           5.2.1
├☑ pandas            2.2.1
└☐ unavailable       nilearn, dipy, openmeeg, cupy

Visualization (optional)
├☑ pyvista           0.43.4 (OpenGL 4.5 (Core Profile) Mesa 23.2.1-1ubuntu3.1~22.04.2 via llvmpipe (LLVM 15.0.7, 256 bits))
├☑ pyvistaqt         0.11.0
├☑ vtk               9.3.0
├☑ qtpy              2.4.1 (PyQt6=6.6.0)
└☐ unavailable       ipympl, pyqtgraph, mne-qt-browser, ipywidgets, trame_client, trame_server, trame_vtk, trame_vuetify

Ecosystem (optional)
├☑ mne-bids          0.15.0.dev43+g17d20c132
├☑ mne-bids-pipeline 1.8.0
└☐ unavailable       mne-nirs, mne-features, mne-connectivity, mne-icalabel, neo