Summarise tonic and phasic GSR/EDA components — summarise_gazepoint_gsr_tonic

Creates a simple descriptive tonic/phasic decomposition of a GSR/EDA signal. The tonic component is estimated with a rolling median, and the phasic component is the observed signal minus the rolling-median tonic estimate.

Usage

summarise_gazepoint_gsr_tonic_phasic(
  data,
  gsr_col = NULL,
  group_cols = NULL,
  time_col = NULL,
  window_n = 15L,
  peak_threshold = NULL,
  output_prefix = "gsr"
)

Arguments

data: A data frame.
gsr_col: Optional GSR/EDA column. If NULL, the function prefers GSR_US, then GSR, then other recognised GSR/EDA columns.
group_cols: Optional grouping columns. Tonic/phasic values are computed separately within each group.
time_col: Optional time column used to order rows within groups.
window_n: Rolling-median window size in samples.
peak_threshold: Optional phasic peak threshold. If NULL, a robust data-driven threshold is computed within each group as median(phasic) + 2 * MAD(phasic).
output_prefix: Prefix for generated columns.

Value

A list with data, summary, and settings.

Details

This is a lightweight descriptive helper, not a full skin-conductance-response deconvolution model. It should be used for quality checks, window summaries, and exploratory reporting unless a study requires a specialised EDA model.

Examples

df <- data.frame(
  CNT = 1:10,
  GSR_US = c(1, 1.1, 1.0, 1.2, 2.0, 1.3, 1.2, 1.1, 1.0, 1.1)
)
summarise_gazepoint_gsr_tonic_phasic(df, window_n = 3)
#> $data
#>    CNT GSR_US gsr_tonic gsr_phasic gsr_phasic_peak gsr_phasic_peak_threshold
#> 1    1    1.0      1.05      -0.05           FALSE                   0.14826
#> 2    2    1.1      1.00       0.10           FALSE                   0.14826
#> 3    3    1.0      1.10      -0.10           FALSE                   0.14826
#> 4    4    1.2      1.20       0.00           FALSE                   0.14826
#> 5    5    2.0      1.30       0.70            TRUE                   0.14826
#> 6    6    1.3      1.30       0.00           FALSE                   0.14826
#> 7    7    1.2      1.20       0.00           FALSE                   0.14826
#> 8    8    1.1      1.10       0.00           FALSE                   0.14826
#> 9    9    1.0      1.10      -0.10           FALSE                   0.14826
#> 10  10    1.1      1.05       0.05           FALSE                   0.14826
#> 
#> $summary
#>   group n_rows source_column n_signal_finite mean_signal median_signal
#> 1   all     10        GSR_US              10         1.2           1.1
#>   mean_tonic median_tonic mean_phasic median_phasic max_phasic min_phasic
#> 1       1.14          1.1        0.06             0        0.7       -0.1
#>   positive_phasic_sum n_phasic_peaks peak_threshold
#> 1                0.85              1        0.14826
#> 
#> $settings
#> $settings$gsr_col
#> [1] "GSR_US"
#> 
#> $settings$group_cols
#> NULL
#> 
#> $settings$time_col
#> NULL
#> 
#> $settings$window_n
#> [1] 3
#> 
#> $settings$peak_threshold
#> NULL
#> 
#> $settings$output_prefix
#> [1] "gsr"
#> 
#> $settings$tonic_col
#> [1] "gsr_tonic"
#> 
#> $settings$phasic_col
#> [1] "gsr_phasic"
#> 
#> $settings$peak_col
#> [1] "gsr_phasic_peak"
#> 
#> $settings$threshold_col
#> [1] "gsr_phasic_peak_threshold"
#> 
#> $settings$note
#> [1] "This is a descriptive rolling-median tonic/phasic decomposition, not a full EDA deconvolution model."
#> 
#>