The Validity of Ultra-Short-Term Heart Rate Variability during Physical Activity and Cognitive Load

As mixed reality (MR) environments gain popularity
for enhancing human performance, there is a growing need for
real-time monitoring to personalize training programs and to
allow human-machine interaction. Heart rate variability (HRV)
is a valuable metric for assessing physical and cognitive load,
but traditional short-term (ST) HRV analysis requires prolonged
measurement periods, limiting its practicality in dynamic MR
scenarios. To support real-time adaptation in MR environment,
it is crucial to assess whether ultra-short-term (UST) HRV
analysis can reliably capture physiological responses during
mixed physical and cognitive tasks. ECG signals were collected at
baseline (squats), and dual-task (squats combined with cognitive
load) conditions from seven young and healthy subjects. Sixteen
time domain variables were extracted from HRV excerpts of
300, 240, 180, 120, 60, and 30 seconds. Significant differences
between baseline and dual-task conditions were investigated
using analysis of variance (ANOVA). Additionally, Pearson’s
correlation, Bland-Altman, and trends analysis were used to
assess the robustness of a subset of UST HRV variables as reliable
alternatives to ST ones. Six out of sixteen UST HRV variables
(MeanNN, CVSD, MedianNN, Prc20NN, Prc80NN, and pNN50)
were identified as reliable variables across all window lengths and
conditions. Three additional HRV variables (RMSSD, SDSD, and
SDRMSSD) showed a close to significant trend across conditions.
This preliminary analysis shows that HRV time-domain UST
variables can be potentially used as a reliable alternative of
standard ST ones, even in the presence of physical activity
combined with cognitive load.