Table 1 Summary of statistical models.

From: Amplitude and frequency of human gait synchronization with a machine oscillator system

Model description

Outcome variables

Fixed Effects variables

β

Lower CL

Upper CL

p

(1) Time-varying Amplitude Experiment (N = 56, R2 = 0.545)

\(A_{m}^{*}\)

Intercept

4.026

1.015

7.037

 0.010*

\(\left| {{\Delta }f_{m} } \right|\)

1.711

1.238

2.183

 < 0.001*

\(\left| {{\Delta }f_{m} } \right|{ }x{ }sgn\left[ {{\Delta }f_{m} } \right]\)

0.363

0.105

0.621

0.007*

(2) Time-varying Frequency Experiment (N = 114, R2 = 0.511)

\({\Delta }f_{m}^{*}\)

\(A_{m}^{0.31}\)

2.909

2.261

3.557

 < 0.001*

Directiona

0.156

− 0.709

1.021

0.722*

\(A_{m}^{0.31} { }x{ }sgn\left[ {{\Delta }f_{m} } \right]\)

− 0.635

− 0.820

− 0.451

 < 0.001*

  1. Linear mixed effects models. Fixed effects for the TVA Experiment include magnitude of motor frequency (\(\left| {{\Delta }f_{m} } \right|\)) and an interaction between motor frequency and its sign (i.e., higher or lower than baseline; \(\left| {{\Delta }f_{m} } \right| x sgn\left[ {{\Delta }f_{m} } \right]\)) to assess effects on the synchronization sensitivity amplitude (\(A_{m}^{*}\)). Fixed effects for the TVF Experiment include non-linear motor amplitude (\(A_{m}^{0.31}\)), direction of motor frequency drift and an interaction term between motor amplitude and sign of motor frequency (\(A_{m}^{0.31} x sgn\left[ {{\Delta }f_{m} } \right]\)) to assess effects on synchronization frequency range (\({\Delta }f_{m}^{*}\)). Coefficients (β), 95% confidence limits (CL), and p-values are shown, marked with asterisks to indicate significance after controlling for multiple testing.
  2. aDirection refers to motor frequency drifting away or toward baseline when synchronization first ceased or was reinitiated, respectively.
Back to article page