HRV Device Index Rises Late in Tunnel Operators’ Shift

A device-generated heart rate variability index was higher during an end-of-shift measurement block than during a mid-shift block among 30 male tunnel control room operators, but the pilot study found that the result cannot be attributed specifically to accumulated shift strain. The operators worked 12-hour day shifts at a single road-tunnel control centre, where they continuously supervised technological and safety systems, interpreted operational deviations and prepared to coordinate emergency responses. Researchers combined a questionnaire on perceived stress, 13 measures of coping behaviour, a timed visual-attention test and approximately five-minute recordings from a finger-clip sensor. Most participants reported moderate perceived stress, and the device index rose by an average of 9.37 points between the two assessment occasions. The results offer a preliminary snapshot of human-factor conditions in a safety-critical occupation, while also illustrating the difficulty of separating genuine physiological change from time-of-day, meal-related and workload-related influences.

The field study was conducted between October 2025 and January 2026 at one of three road-tunnel control centres considered for participation. All 30 employees available at the selected site who met the inclusion criteria completed the protocol. They were exclusively men aged 34 to 65 years, with a mean age of 51.00 years and a standard deviation of 9.26 years. Every participant had at least three years of tunnel-operations experience and previous shift-work experience, although their exact years of service were not systematically recorded. Recruitment was non-probability based, and the sample was deliberately small to test the applicability of the instruments and procedures in a real operating environment. The authors said the results were not intended to represent the entire occupational population.

Perceived stress was assessed with the 10-item Perceived Stress Scale, which asks how often respondents have experienced life as unpredictable, uncontrollable and overwhelming during the preceding month. After reverse-scoring positively worded items, total scores range from 0 to 40, with higher values indicating greater perceived stress. The mean total was 20.80, with a standard deviation of 4.35, a median of 21 and a range of 8 to 27. Using published descriptive bands, two participants, or 6.67%, were classified as having low perceived stress; 26, or 86.67%, as moderate; and two, or 6.67%, as high. The mean score for the Perceived Helplessness subscale was 10.00 out of 24, while the mean for Lack of Self-Efficacy was 10.80 out of 16. These cutoffs were used for description rather than diagnosis, and the scale’s one-month reference period differs from the much shorter physiological recording windows.

For coping, the researchers used the 78-item Stressverarbeitungsfragebogen, or SVF-78, which distinguishes relatively stable ways of responding to stressful situations. Situation Control had the highest sample mean at 19.23, followed by Reaction Control at 16.97 and Positive Self-Instructions at 16.90. In descriptive terms, this profile suggested that operators tended to rely particularly on appraising and managing a situation, regulating their reactions and using reassuring self-direction. Exploratory differences from published raw-score means for men were largest for Avoidance, at plus 5.16 points, Undervaluation, at plus 2.63, and Situation Control, at plus 2.39. Perseveration was the only measure below its corresponding male normative mean, by 0.38 points, while Self-Blame was only 0.19 points higher. Because the researchers did not conduct inferential comparisons with the normative group, these figures cannot show that tunnel operators have a coping profile distinct from other workers.

The device-generated index came from the Max Pulse system, a 2013 pulse-wave device operated through MaxPulse Software and fitted with a finger sensor on the left index finger. The system produced a composite numerical output on a 100-point scale, with higher values representing a higher device-index reading. Operators were seated and asked to remain still and silent for each approximately five-minute recording, while signal quality was monitored. No recording required repetition because of poor signal, and the investigators reported no ectopic beats in the analysed records. Because the calculation algorithm and independent validation of this particular Max Pulse output were unavailable, the researchers treated the measure as an exploratory, device-specific composite rather than a diagnostic indicator of physiological stress.

Unlike the attentional test, the physiological recordings were not collected at the beginning and end of the entire shift. The mid-shift block began at approximately 11:00, five hours after the 06:00 start, and the end-shift block began around 16:00, two hours before the scheduled 18:00 finish. Participants were assessed sequentially, and recording all 30 operators required about 150 minutes at each occasion. A third physiological block at shift start was omitted because it would have overlapped with the pre-shift attention test and operational duties. Mean device-index values were 45.50 at mid-shift, with a standard deviation of 18.83, and 54.87 at end-shift, with a standard deviation of 16.75. The end-shift reading exceeded the mid-shift value by an average of 9.37 points, with a 95% confidence interval of 4.44 to 14.29.

That difference was statistically significant in a paired-samples t-test, with t(29) = 3.89, p < 0.001 and Cohen’s dz = 0.71. The authors described the result as temporal variability in a device-generated output under real working conditions, not proof of rising physiological stress. No start-of-shift baseline was available, and the two blocks occurred at different times of day. Meal timing and food intake were neither standardised nor systematically recorded, participants had no separate standardised rest period before measurement, and they continued their usual duties beforehand. Time-of-day variation, post-meal autonomic effects, fatigue, operational workload and other unmeasured factors could therefore have contributed to the difference. Heart-rate variability measures can respond to a broad range of physiological and contextual conditions, while the proprietary composite index adds an additional layer of uncertainty.

Selective attention was evaluated with the d2 Test of Attention, whose Concentration Performance index is the number of correctly marked items minus errors. It was administered near the beginning and end of the full 12-hour shift. The mean pre-shift score was 444.17, with a standard deviation of 115.44 and a median of 522, while the mean post-shift score was 414.03, with a standard deviation of 96.90 and a median of 392.50. Although the post-shift average was lower, the paired change of minus 30.13 points was not statistically significant: t(29) = −1.18, p = 0.248, with Cohen’s dz = −0.22 and a 95% confidence interval of minus 82.43 to 22.16. The study therefore did not demonstrate a systematic within-person decline in selective attention. Repeating the same test may also have introduced a practice effect that interacted with fatigue, workload or time of day.

Several limitations constrain what can be inferred. The sample was small, non-random and drawn from one control centre, and it included only male operators working 12-hour day shifts, limiting generalisation to women, other workplaces and night operations. Sleep quality, caffeine use, physical activity, traffic intensity and operational incidents were not systematically assessed, and the study did not record workplace errors, safety events or system reliability. The measures also covered different periods: the perceived-stress scale referred to the previous month, the attention test sampled two points in one shift, and the physiological index was based on short recordings. The result is best understood as a methodologically useful but preliminary observation from a naturally occurring workplace, not as evidence of cognitive deterioration, increased physiological stress or declining tunnel safety. Larger multisite studies with start-of-shift baselines, transparent and independently validated physiological measures, standardised recording conditions and systematic monitoring of contextual variables will be needed before operational conclusions can be drawn.

Subject of Research: Pilot assessment of perceived stress, coping, device-generated HRV and attention in tunnel control room operators during 12-hour day shifts.

Article Title: Perceived stress, coping strategies, HRV-based stress index, and attentional performance in tunnel control room operators: a pilot descriptive study

Article References: Šimůnková, E., Kozáková, E., & Hon, Z. (2026). Perceived stress, coping strategies, HRV-based stress index, and attentional performance in tunnel control room operators: a pilot descriptive study. Frontiers in Psychology, 17, Article 1876583. https://doi.org/10.3389/fpsyg.2026.1876583

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1876583

Keywords: tunnel operators, occupational stress, heart rate variability, selective attention, coping strategies, safety-critical work, human factors, shift work, pilot study, workplace physiology, Perceived, stress