run_regional.rsa_model: Run regional RSA analysis on a specified RSA model

In bbuchsbaum/rMVPA: Multivoxel Pattern Analysis in R

Run regional RSA analysis on a specified RSA model

Description

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 class '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).

bbuchsbaum/rMVPA documentation built on April 28, 2024, 6:30 a.m.