Table 1.

GLM predicting total HRV (GLM-HRV)

Regressorβ EstimateSEMtp
Number of artifacts5.071.493.410.0014
Mean heart rate−1.170.47−2.480.017
Trait anxiety−0.570.63−0.910.37
  • Results from a GLM with possible determinants of HRV (represented as untransformed values of SDNN in milliseconds): the number of artifacts corrected in the dataset and mean heart rate (after artifact correction by deletion). To assess whether trait anxiety explains additional variance beyond an increase in mean heart rate, we added a regressor with the trait anxiety score as measured by the Spielberger State-Trait-Anxiety Inventory, so that mean heart rate and trait anxiety compete for variance in the same model. When doing so, mean heart rate accounts for a decrease in HRV, but trait anxiety does not explain further variance. The results above hold when excluding the HRV outlier from this model.