Skip to contents

correlation_map

Source: R/all_generic.R, R/correlation_map_methods.R
correlation_map.Rd

Generate a correlation heatmap showing the relationships between columns in a design matrix. This visualization helps identify potential collinearity between regressors in the model. For event models, it shows correlations between different conditions. For baseline models, it shows correlations between drift and nuisance terms.

These methods provide correlation heatmap visualizations for various model objects.

Usage

correlation_map(x, ...)

# S3 method for class 'baseline_model'
correlation_map(
  x,
  method = c("pearson", "spearman"),
  half_matrix = TRUE,
  absolute_limits = TRUE,
  within_run = TRUE,
  label_values = NULL,
  ...
)

Arguments

x

The model object (event_model, baseline_model, or fmri_model)

...

Additional arguments passed to methods. Common arguments include:

method

Correlation method: "pearson" (default) or "spearman"

half_matrix

Logical; if TRUE, show only lower triangle (default: FALSE)

absolute_limits

Logical; if TRUE, set color limits to [-1,1] (default: TRUE)

method

Correlation method: "pearson" (default) or "spearman"

half_matrix

Logical; if TRUE (default), show only the lower triangle

absolute_limits

Logical; if TRUE, set color limits to [-1,1] (default: TRUE)

within_run

Logical; centre columns within runs and drop run intercepts before correlating (default: TRUE)

label_values

Logical or NULL; print correlations in the cells (NULL: all cells when there are at most 12 columns)

Value

A ggplot2 object containing the correlation heatmap, where:

  • Rows and columns represent model terms

  • Colors indicate correlation strength (-1 to 1)

  • Darker colors indicate stronger correlations

A ggplot2 object containing the correlation heatmap visualization

Details

Create a correlation heatmap for an fMRI design matrix.

See also

event_model(), baseline_model()

Examples

# Create event data
event_data <- data.frame(
  condition = factor(c("face", "house", "face", "house")),
  rt = c(0.8, 1.2, 0.9, 1.1),
  onsets = c(1, 10, 20, 30),
  run = c(1, 1, 1, 1)
)

# Create sampling frame
sframe <- sampling_frame(blocklens = 50, TR = 2)

# Create event model
evmodel <- event_model(
  onsets ~ hrf(condition) + hrf(rt),
  data = event_data,
  block = ~run,
  sampling_frame = sframe
)

# Plot correlation map for event model
correlation_map(evmodel)


# Create baseline model
bmodel <- baseline_model(
  basis = "bs",
  degree = 3,
  sframe = sframe
)

# Plot correlation map for baseline model
correlation_map(bmodel)


# Note: To create a full fmri_model and plot combined correlations,
# you would need an fmri_frame object:
# fmodel <- fmri_model(evmodel, bmodel, dataset)
# correlation_map(fmodel, method = "pearson", half_matrix = TRUE)