Skip to contents

Extracts a hand-marker trajectory, computes tangential speed with computeVelocity() and computeSpeed(), and reports temporal, submovement, smoothness, endpoint, and optional trunk-compensation metrics.

Usage

reachingKinematics(
  pe,
  marker,
  target = NULL,
  assay_prefix = "position",
  sampling_rate = NULL,
  onset_threshold = 0.05,
  trunk_marker = NULL,
  shoulder_markers = NULL,
  ...
)

Arguments

pe

A PhysioExperiment with position_x, position_y, and optionally position_z assays.

marker

Reaching hand marker name or column index.

target

Optional target coordinate matching trajectory dimension.

assay_prefix

Position-assay prefix (default "position").

sampling_rate

Optional sampling frequency in Hz; defaults to PhysioExperiment::samplingRate().

onset_threshold

Movement threshold.

trunk_marker

Optional trunk marker name or column index.

shoulder_markers

Optional length-2 vector identifying right and left shoulder markers.

...

Additional arguments passed to sparc().

Value

A reaching_kinematics report with movement time, peak velocity, time to peak, movement units, SPARC, LDLJ, dimensionless jerk, movement bounds, sampling rate, marker, and optional endpoint/trunk results.

Examples

demo <- demoMoCapData(seed = 1)
rk <- reachingKinematics(demo$mocap, marker = "Toe_R")
names(rk)
#>  [1] "movement_time"      "peak_velocity"      "time_to_peak"      
#>  [4] "time_to_peak_frac"  "n_movement_units"   "sparc"             
#>  [7] "ldlj"               "dimensionless_jerk" "onset"             
#> [10] "offset"             "fs"                 "marker"