Compute transparent signal-quality indicators — compute_gazepoint_signal

Computes conservative, rule-based descriptive quality indicators for one or more Gazepoint biometric signal columns. The function does not interpret the physiological meaning of the signal and does not exclude data.

Usage

compute_gazepoint_signal_quality(
  data,
  signal_cols,
  group_cols = NULL,
  flatline_tolerance = 0,
  long_missing_run_threshold = 10,
  long_constant_run_threshold = 10,
  spike_z = 4,
  extreme_z = 4
)

Arguments

data: A data frame.
signal_cols: Character vector of numeric signal columns to evaluate.
group_cols: Optional character vector of grouping columns, such as participant, trial, condition, session, window, or segment identifiers.
flatline_tolerance: Numeric tolerance used when detecting adjacent constant values. Defaults to 0.
long_missing_run_threshold: Integer threshold used to count whether a segment contains a long missing run. The maximum run length is always returned regardless of this threshold.
long_constant_run_threshold: Integer threshold used to count whether a segment contains a long constant run. The maximum run length is always returned regardless of this threshold.
spike_z: Numeric z-score threshold for adjacent-change spikes.
extreme_z: Numeric z-score threshold for extreme standardized values.

Value

A data frame with class gazepoint_signal_quality.