convolve: Convolve a term with a hemodynamic response function

In bbuchsbaum/fmrireg: R package for the anlysis of fmri data

Convolve a term with a hemodynamic response function

Description

This function convolves an event sequence with a hemodynamic response function (HRF) over a specified time series grid. The convolution models the expected BOLD response to the events. For event-related designs, each event is convolved with the HRF and the results are summed. For block designs, the duration of each event is taken into account during convolution.

Usage

convolve(x, hrf, sampling_frame, ...)

## S3 method for class 'event_term'
convolve(
  x,
  hrf,
  sampling_frame,
  drop.empty = TRUE,
  summate = TRUE,
  precision = 0.3,
  ...
)

Arguments

x	The event sequence (typically an event_term or event_model)
hrf	The hemodynamic response function to use for convolution
sampling_frame	The time series grid over which to sample the convolved function
...	Additional arguments passed to methods
drop.empty	Logical; if TRUE, empty events are dropped (default: TRUE)
summate	Logical; if TRUE, sum the convolved HRF over event durations (default: TRUE)
precision	Numeric; precision of HRF sampling (default: 0.3)

Value

A tibble containing the convolved design matrix, with columns for each condition

See Also

fmrihrf::HRF_SPMG1(), event_term(), sampling_frame()

Examples

# Create a simple event-related design
event_data <- data.frame(
  condition = factor(c("A", "B", "A", "B")),
  onsets = c(1, 10, 20, 30),
  run = c(1, 1, 1, 1)
)

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

# Create an event term
eterm <- event_term(
  list(condition = event_data$condition),
  onsets = event_data$onsets,
  blockids = event_data$run
)

# Convolve with canonical HRF
convolved <- convolve(eterm, HRF_SPMG1, sframe)

# Convolve with multiple basis functions
convolved_fourier <- convolve(
  eterm, 
  fmrihrf::gen_hrf("fourier", nbasis = 2),
  sframe
)

bbuchsbaum/fmrireg documentation built on June 10, 2025, 8:18 p.m.