
Compare observed and predicted EEG time courses across sensors
Source:R/plot_sensor_trajectories.R
plot_sensor_trajectories.RdPlots observed sensor-wise means (black) against posterior median model predictions (blue), optionally with credible ribbons. Trajectories can be placed in approximate scalp positions, sharing one amplitude scale, or displayed as ordinary facets. Detected clusters can appear as translucent rectangles or as thicker, opaque portions of the *predicted* trajectories. Cluster segments respect sign, cluster ID and gaps in the time grid.
Usage
plot_sensor_trajectories(
results,
observed_data,
observed_data_b = NULL,
outcome_id = "voltage",
sensor_id = "channel",
time_id = "time",
spatial_id = c("xproj", "yproj"),
participant_id = NULL,
layout = c("scalp", "facets"),
cluster_style = c("rectangles", "segments", "none"),
cluster_alpha = 0.2,
highlighted_linewidth = 1,
max_gap = NULL,
show_ci = TRUE,
ncol = 8L,
free_y = FALSE,
sensors = NULL,
sensor_order = NULL,
mini_width = NULL,
mini_height = NULL,
amplitude_limits = NULL,
show_labels = TRUE,
head_outline = TRUE
)Arguments
- results
Output from [analyse_bgam_spatiotemporal()].
- observed_data
Original long-form observations, or condition A.
- observed_data_b
Observed condition B for A - B contrasts.
- outcome_id
Observed outcome column.
- sensor_id
Sensor identifier column.
- time_id
Numeric time column.
- spatial_id
Two projected sensor coordinates.
- participant_id
Optional participant ID for an equal-participant- weighted observed average. With `NULL`, observations are pooled.
- layout
`"scalp"` for mini-trajectories at electrode positions or `"facets"` for one independent panel per sensor.
- cluster_style
`"rectangles"` for transparent backgrounds, `"segments"` for highlighted prediction lines, or `"none"`.
- cluster_alpha
Opacity of cluster rectangle backgrounds.
- highlighted_linewidth
Width of cluster-highlighted segments.
- max_gap
Maximum time gap for joining cluster members. Defaults to 1.5 times the median sampling interval.
- show_ci
Plot posterior credible interval ribbons.
- ncol
Number of facet columns when `layout = "facets"`.
- free_y
Use independent amplitude limits in faceted plots.
- sensors
Optional vector of sensors to display.
- sensor_order
Optional vector specifying the order of facet panels.
- mini_width
Width of each scalp mini-trace in projected units; defaults to 80 percent of median nearest-sensor spacing.
- mini_height
Height of each scalp mini-trace in projected units; defaults to 75 percent of median nearest-sensor spacing.
- amplitude_limits
Optional common amplitude range for scalp traces.
- show_labels
Display sensor labels in scalp layout.
- head_outline
Add a schematic head circle and nose in scalp layout.
Author
Ladislas Nalborczyk ladislas.nalborczyk@cnrs.fr.
Examples
if (FALSE) { # \dontrun{
# Not run: generating `results_st` requires
# computationally intensive Bayesian model fitting.
# `eeg_data2` contains the corresponding observations.
plot_sensor_trajectories(
results_st, eeg_data2, outcome_id = "voltage",
sensor_id = "channel", layout = "scalp",
cluster_style = "segments"
)
} # }