pspm_sf
Description
pspm_sf is a wrapper function for analysis of skin conductance as a measure of tonic arousal. SF are analysed over time windows that typically last 60 s and should at least be 15 s long. PsPM implements 3 different models.
(1) Skin conductance level (SCL): this is the mean signal over the epoch.
(2) Area under the curve (AUC): this is the time-integral of the signal with the minimum value subtracted (to account for pre-epoch arousal), divided by epoch duration. This is designed to be independent from SCL and ideally represents the number x amplitude of spontaneous fluctuations (also termed non-specific SCR) in this epoch.
(3) Number of SF estimated by DCM: this is a non-linear estimation of the number and onset of SF, and is the most sensitive indicator of tonic arousal. For counting peaks, a threshold in mcS is applied; hence it is important that the data are provided in the correct units. Estimated SF onset is stored in the model and is expressed in CNS time, i.e. the time point at which an SF was generated in the CNS. Thus, it already takes into account the conduction delay from CNS into the periphery.
(4) Number of SF estimated by MP: This is the same model as in (3) but estimated with an approximative matching pursuit (MP) algorithm.
Format
[sts, sf] = pspm_sf(model, options)
Arguments
| Variable | Definition | |||
|---|---|---|---|---|
| model | See following fields. | |||
| model.datafile | one data filename or cell array of filenames. | |||
| model.modelfile | one data filename or cell array of filenames. | |||
| model.timing | can be one of the following - an SPM style onset file with two event types: onset & offset (names are ignored) - a .mat file with a variable ‘epochs’, see below - a two-column text file with on/offsets - e x 2 array of epoch on- and offsets, with e: number of epochs or cell array of any of these, for multiple files. | |||
| model.timeunits | seconds, samples, markers, whole (in the last case, ‘timing’ will be ignored and the entire file will be used). | |||
| model.method | [optional, string/cell_array] [string] accept ‘auc’, ‘scl’, ‘dcm’, or ‘mp’, default as ‘dcm’. [cell_array] a cell array of methods mentioned above. | |||
| model.filter | [optional] filter settings; modality specific default. | |||
| model.missing | [optional, string/cell_array] [default: no missing values] Allows to specify missing (e.g. artefact) epochs in the data file. See pspm_get_timing for epoch definition; specify a cell array for multiple input files. This must always be specified in SECONDS. | |||
| model.channel | [optional, integer, default: last SCR channel] Channel number. | options | See following fields. | |
| options.overwrite | [logical, default: determined by pspm_overwrite] Define whether to overwrite existing output files or not. | |||
| options.marker_chan_num | [integer] marker channel number if undefined or 0, first marker channel is used. | |||
| options.threshold | [numeric, default: 0.1] [unit: mcS] Threshold for counting estimated SN peaks (default 0.1 mcS). (Used for DCM and MP only.). | |||
| options.theta | [vector default: read from pspm_sf_theta] A (1 x 5) vector of theta values for f_SF. (Used for DCM and MP only.). | |||
| options.fresp | [numeric, default: 0.5] (Maximum) frequency (in Hz) of responses in the model. (Used for DCM and MP only.). | |||
| options.dispwin | [logical, default: 1] Display progress plot (DCM) or result plot (MP). | |||
| options.dispsmallwin | [logical, default: 0] Display intermediate progress windows. (Used for DCM only.). | |||
| options.missingthresh | [numeric, default: 2] [unit: second] Threshold value for controlling missing epochs. (Used for DCM only). |
References
[1] DCM for SF: Bach DR, Daunizeau J, Kuelzow N, Friston KJ, Dolan RJ (2010). Dynamic causal modelling of spontaneous fluctuations in skin conductance. Psychophysiology, 48, 252-257.
[2] MP approximation: Bach DR, Staib M (2015). A matching pursuit algorithm for inferring tonic sympathetic arousal from spontaneous skin conductance fluctuations. Psychophysiology, 52, 1106-12.
[3] AUC for SF: Bach DR, Friston KJ, Dolan RJ (2010). Analytic measures for the quantification of arousal from spontanaeous skin conductance fluctuations. International Journal of Psychophysiology, 76, 52-55.