Analysis of the Sensitivity of Heart Rate Variability and Subjective Workload Measures in a Driving Simulator: The Case of Highway Work Zones

Heart rate variability and subjective workload measures are extensively used to determine workload during driving. However, the sensitivity of heart rate and workload measurements in simulated driving environments is mostly unknown and can significantly affect the experiment results. The objectives of this paper are to determine how heart rate variability and subjective workload are affected in simulated highway work zones and study the relationship between heart rate variability, subjective workload, and driving performance indicators in simulated driving environments. Conventional lane merge (CLM), joint lane merge (JLM) and a road with no work zone are modeled with high and low traffic densities in a full-size driving simulator. NASA-TLX subjective workload measures and heart rate variability measures of root mean square of successive heartbeat differences (RMSSD), low frequency (LF), high frequency (HF) and the ratio of low frequency to high frequency (LF/HF) are collected in 30 participants. Variability in steering angle, braking and speed are used as driving performance indicators. Results show that compared to no work zone, participants experienced higher mental, temporal, and overall workload in the CLM scenario and poorer driving performance ratings in the CLM and JLM scenarios. All workload measures except for performance were higher with high traffic density. However, heart rate variability measures were not sensitive to the differences in driving scenarios and traffic densities. Pearson correlation coefficients indicated an association between RMSSD and all the subjective workload measures (r > 0.21) except performance, and between LF, HF, and LF/HF ratio and mental workload (r > 0.21). Steering angle variability was slightly correlated with LF, HF, and LF/HF ratio (r > 0.16), but brake and speed variability were not associated with physiological outcomes. In conclusion, the subjective workload was higher in simulated work zones and under higher traffic density, but heart rate measures were largely unaffected.