compute_tfr: Compute Time-Frequency representation of EEG data

In craddm/eegUtils: Utilities for Electroencephalographic (EEG) Analysis

Compute Time-Frequency representation of EEG data

Description

This function creates a time frequency representation of EEG time series data. Currently, it is possible to use either Morlet wavelets or Hanning tapers during the decomposition, which uses convolution in the frequency domain.

Usage

compute_tfr(data, ...)

## Default S3 method:
compute_tfr(data, ...)

## S3 method for class 'eeg_epochs'
compute_tfr(
  data,
  method = "morlet",
  foi,
  n_freq,
  spacing = "linear",
  n_cycles = 7,
  keep_trials = FALSE,
  output = "power",
  downsample = 1,
  verbose = TRUE,
  ...
)

## S3 method for class 'eeg_evoked'
compute_tfr(
  data,
  method = "morlet",
  foi,
  n_freq,
  spacing = "linear",
  n_cycles = 7,
  keep_trials = FALSE,
  output = "power",
  downsample = 1,
  verbose = TRUE,
  ...
)

Arguments

data	An object of class eeg_epochs.
...	Further TFR parameters
method	Time-frequency analysis method. Defaults to "morlet".
foi	Frequencies of interest. Scalar or character vector of the lowest and highest frequency to resolve.
n_freq	Number of frequencies to be resolved. Must be an integer number of frequencies.
spacing	Use "linear" or "log" spacing for the frequency vector and number of cycles.
n_cycles	Number of cycles at each frequency. If a single integer, use a constant number of cycles at each frequency. If a character vector of length 2, the number of cycles will scale with frequency from the minimum to the maximum.
keep_trials	Keep single trials or average over them before returning. Defaults to FALSE.
output	Sets whether output is power, phase, or fourier coefficients.
downsample	Downsampling factor. Integer. Selects every n samples after performing time-frequency analysis on the full sampling rate data.
verbose	Print informative messages in console.

Value

An object of class eeg_tfr

Methods (by class)

• default: Default method for compute_tfr

• eeg_epochs: Default method for compute_tfr

• eeg_evoked: Method for eeg_evoked objects.

Author(s)

Matt Craddock [e-mail]

Examples

out <- compute_tfr(demo_epochs, method = "morlet", foi = c(4, 30), n_freq = 10, n_cycles = 3)
out
out$freq_info$morlet_resolution
out <- compute_tfr(demo_epochs, method = "morlet", foi = c(4, 30), n_freq = 10, n_cycles = c(3, 10))
out$freq_info$morlet_resolution

craddm/eegUtils documentation built on March 24, 2022, 9:17 a.m.