Comparative sensitivity of continuous cardiovascular markers for detecting cognitive workload, time-on-task fatigue, and acute stress.
Abstract
Cardiovascular markers such as heart rate (HR), heart rate variability (HRV), and blood pressure (BP) are promising for low-intrusive monitoring of mental states relevant to safety-critical performance, yet evidence is mixed regarding which measures are most sensitive to cognitive workload, stress, and fatigue across different task types and time frames. Sixty adults (33 women; M age = 28.1 years, SD = 7.9) completed a 90-120 minute laboratory protocol including four cognitive workload tasks (Judgment of Line Angle and Position, Corsi Block-Tapping, Balloon Analogue Risk Task, Stroop), a modified Trier Social Stress Test (preparation, speech, arithmetic), and a fatigue condition operationalized as time-on-task during a prolonged Stroop (15 min) and Psychomotor Vigilance Task (10 min), each divided into three intervals. Continuous beat-to-beat HR, interbeat interval (IBI), and systolic/diastolic BP were recorded using the CareTaker device; HRV indices (RMSSD, high-frequency power) were derived from IBI using Kubios. Difference scores (task minus baseline) were analyzed using linear mixed models (random intercepts for participants; age and gender covariates), with Tukey-corrected pairwise comparisons to quantify sensitivity across conditions. HR and IBI consistently differentiated multiple workload and stress phases and distinguished Stroop-based fatigue from PVT intervals. HRV-RMSSD showed selective sensitivity, differentiating specific cognitive tasks (notably Stroop relative to other workload tasks), but only demonstrated a trend toward detecting fatigue-related differences at later time intervals. High-frequency HRV showed task-dependent effects only within the cognitive workload condition. Continuous BP did not reliably differentiate tasks, showing at most marginal (uncorrected) effects during TSST phase comparisons. Cardiovascular reactivity patterns depended on task type and temporal dynamics. HR and IBI provided the most robust sensitivity across workload, stress, and fatigue within a prolonged, multi-task protocol, while HRV indices contributed complementary but more task-specific information. The results support multi-metric monitoring approaches for real-time mental state transitions in high-risk operational settings.
Related
- Diagnostic accuracy of urine test strips for CSF analysis: high sensitivity in detecting blood, low sensitivity but high specificity in detecting inflammation
- Curve walking gait analysis using depth camera and machine learning for detecting cognitive impairment
- Adjudicator sensitivity of safety estimates in large language models for open-ended cardiovascular reasoning: a blinded multi-adjudicator benchmark study running title: adjudicator variability in LLM safety
- Distinct insulin sensitivity profiles revealed by clustering based on continuous glucose monitoring data in people living with type 1 diabetes
- Increased sensitivity of detecting variation in malaria endemicity by parasite density measurement in large surveys
BibTeX
@misc{kristofersson2026comparative,
title = {{Comparative sensitivity of continuous cardiovascular markers for detecting cognitive workload, time-on-task fatigue, and acute stress.}},
author = {P. Kristofersson and S. T. Sveinsdottir and S. M. Kristinardottir and S. H. Kristjansson and K. R. Johannsdottir},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.02.749015},
url = {https://www.biorxiv.org/content/10.64898/2026.09.02.749015v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Comparative sensitivity of continuous cardiovascular markers for detecting cognitive workload, time-on-task fatigue, and acute stress.
AU - P. Kristofersson
AU - S. T. Sveinsdottir
AU - S. M. Kristinardottir
AU - S. H. Kristjansson
AU - K. R. Johannsdottir
PY - 2026
DA - 2026/09/02
PB - bioRxiv
DO - 10.64898/2026.09.02.749015
UR - https://www.biorxiv.org/content/10.64898/2026.09.02.749015v1
AB - Cardiovascular markers such as heart rate (HR), heart rate variability (HRV), and blood pressure (BP) are promising for low-intrusive monitoring of mental states relevant to safety-critical performance, yet evidence is mixed regarding which measures are most sensitive to cognitive workload, stress, and fatigue across different task types and time frames. Sixty adults (33 women; M age = 28.1 years, SD = 7.9) completed a 90-120 minute laboratory protocol including four cognitive workload tasks (Judgment of Line Angle and Position, Corsi Block-Tapping, Balloon Analogue Risk Task, Stroop), a modified Trier Social Stress Test (preparation, speech, arithmetic), and a fatigue condition operationalized as time-on-task during a prolonged Stroop (15 min) and Psychomotor Vigilance Task (10 min), each divided into three intervals. Continuous beat-to-beat HR, interbeat interval (IBI), and systolic/diastolic BP were recorded using the CareTaker device; HRV indices (RMSSD, high-frequency power) were derived from IBI using Kubios. Difference scores (task minus baseline) were analyzed using linear mixed models (random intercepts for participants; age and gender covariates), with Tukey-corrected pairwise comparisons to quantify sensitivity across conditions. HR and IBI consistently differentiated multiple workload and stress phases and distinguished Stroop-based fatigue from PVT intervals. HRV-RMSSD showed selective sensitivity, differentiating specific cognitive tasks (notably Stroop relative to other workload tasks), but only demonstrated a trend toward detecting fatigue-related differences at later time intervals. High-frequency HRV showed task-dependent effects only within the cognitive workload condition. Continuous BP did not reliably differentiate tasks, showing at most marginal (uncorrected) effects during TSST phase comparisons. Cardiovascular reactivity patterns depended on task type and temporal dynamics. HR and IBI provided the most robust sensitivity across workload, stress, and fatigue within a prolonged, multi-task protocol, while HRV indices contributed complementary but more task-specific information. The results support multi-metric monitoring approaches for real-time mental state transitions in high-risk operational settings.
ER -