Skip to contents

This function runs a regional RSA analysis using a specified RSA model and region mask. The analysis can be customized to return model fits and performance measures.

Usage

# S3 method for rsa_model
run_regional(
  model_spec,
  region_mask,
  return_fits = FALSE,
  compute_performance = TRUE,
  regtype = c("pearson", "spearman", "lm", "rfit"),
  distmethod = c("pearson", "spearman"),
  coalesce_design_vars = FALSE,
  ...
)

Arguments

model_spec

An object of type rsa_model specifying the RSA model to be used.

region_mask

A mask representing different regions in the brain image.

return_fits

Whether to return model fit for every ROI (default is FALSE to save memory).

compute_performance

logical indicating whether to compute performance measures (e.g. Accuracy, AUC).

regtype

The regression method ("pearson", "spearman", "lm", or "rfit").

distmethod

The distance computing method ("pearson" or "spearman").

coalesce_design_vars

Concatenate additional design variables with output stored in `prediction_table`.

...

Additional arguments to be passed to the function.

Value

A list of type regional_mvpa_result containing a named list of NeuroVol objects, where each element contains a performance metric and is labeled according to the metric used (e.g. Accuracy, AUC).