EegFun

Appearance

Plot ERP Image ​

This demo shows how to create ERP image plots for visualizing single-trial EEG activity.

What is an ERP Image? ​

An ERP image displays single-trial data as a 2D heatmap:

  • Y-axis: Individual trials/epochs

  • X-axis: Time points

  • Color: Amplitude at each time-trial coordinate

This reveals trial-to-trial variability that is hidden when viewing only averaged ERPs.

Why Use ERP Images? ​

See variability beyond the average:

Traditional ERPs show the average across trials, but ERP images show:

  • Trial-to-trial amplitude fluctuations

  • Temporal jitter in component latency

  • Phase consistency vs. phase resetting

Quality control:

Quickly spot:

  • Artifact-contaminated trials (bright/dark streaks)

  • Drift across the experiment

  • Individual outlier trials

Understand components:

Distinguish between:

  • Time-locked activity: Consistent latency across trials (vertical bands)

  • Phase-locked activity: Consistent phase but variable amplitude

  • Induced activity: Not phase-locked to stimulus

Layout Options ​

The demo shows three layout modes:

LayoutDescription
:singleAll selected channels in one column
:gridChannels arranged in a grid
:topoChannels positioned by scalp location

Visualization Features ​

Boxcar averaging:

Smooth the image by averaging across neighboring trials:

julia
plot_erp_image(epochs, boxcar_average = 20)

Reduces noise while preserving overall patterns.

Colorbar control:

julia
plot_erp_image(epochs, colorbar_plot = false)  # Hide colorbar
plot_erp_image(epochs, colorrange = (-50, 50))  # Custom range

Optional ERP overlay:

julia
plot_erp_image(epochs, plot_erp = false)  # Hide the averaged ERP

By default, the averaged ERP is plotted above the image for reference.

Interpretation ​

Vertical bands:

Strong vertical alignment indicates time-locked activity with consistent latency across trials.

Gradual color changes:

Smooth transitions suggest sustained activity or slow baseline drift.

Random speckles:

High-frequency noise or lack of consistent time-locked activity.

Outlier rows:

Trials with extreme values may indicate artifacts (blinks, movements, poor contact).

Workflow Summary ​

This demo demonstrates:

  1. Load and preprocess data (rereferencing, filtering)

  2. Extract epochs (-2 to 4 seconds)

  3. Plot ERP images with different layouts

  4. Apply boxcar averaging to smooth trial-to-trial noise

  5. Customize visualization with colorbars and color ranges

ERP images are a powerful complement to traditional ERP plots, revealing the underlying trial structure that averages can obscure.

Code Examples ​

Show Code
julia
# Demo: ERP Image Plots
# Shows trial-by-trial ERP activity as image/heatmap plots.

using EegFun
# Note: EegFun.example_path() resolves bundled example data paths.
# When using your own data, simply pass the file path directly, e.g.:
# dat = EegFun.read_raw_data("/path/to/your/data.bdf")

# read raw data
dat = EegFun.read_raw_data(EegFun.example_path("data/bdf/example1.bdf"));

# read and prepare layout file
layout = EegFun.read_layout(EegFun.example_path("layouts/biosemi/biosemi72.csv"));
EegFun.polar_to_cartesian_xy!(layout)

dat = EegFun.create_eegfun_data(dat, layout)

# minimal preprocessing
EegFun.rereference!(dat, :avg)
EegFun.highpass_filter!(dat, 1)

# EPOCHS
epoch_cfg = EegFun.EpochCondition(name = "ExampleEpoch1", trigger_sequences = [[1]])
epochs = EegFun.extract_epochs(dat, epoch_cfg, (-2, 4))

EegFun.plot_erp_image(epochs, layout = :single)  # average of all channels
EegFun.plot_erp_image(epochs, layout = :single, channel_selection = EegFun.channels([:PO7, :PO8]))
EegFun.plot_erp_image(epochs, layout = :single, channel_selection = EegFun.channels([:PO7, :PO8]), interpolate = true)

EegFun.plot_erp_image(epochs, layout = :grid, colorbar_plot = false)
EegFun.plot_erp_image(epochs, layout = :grid, colorbar_plot = true)

EegFun.plot_erp_image(epochs, layout = :topo)
EegFun.plot_erp_image(epochs, channel_selection = EegFun.channels([:PO7]))
EegFun.plot_erp_image(epochs, channel_selection = EegFun.channels([:Fp1]), plot_erp = false)


EegFun.plot_erp_image(epochs, layout = :single)

EegFun.plot_erp_image(
    epochs,
    channel_selection = EegFun.channels([:PO7, :PO8]),
    layout = :single,
    boxcar_average = 2,
    colorrange = (-50, 50),
)

EegFun.plot_erp_image(epochs, layout = :topo, boxcar_average = 20, colorrange = (-50, 50))

See Also ​