When is rPPG physiologically valid for stress and workload monitoring? A PRISMA-ScR scoping review
Remote photoplethysmography (rPPG) enables non-contact estimation of cardiovascular signals from ordinary cameras and has been increasingly applied to stress and mental workload assessment using heart rate (HR), heart rate variability (HRV), and pulse rate variability (PRV). However, reported performance varies substantially across studies, and stress-classification results are often presented without sufficient physiological validation, limiting interpretability and generalizability. This paper presents a PRISMA-ScR-guided scoping review of rPPG-based stress and mental workload research published between January 2010 and March 2026. Searches across Scopus, IEEE Xplore, PubMed, ACM Digital Library, and Google Scholar identified 20 empirical studies meeting predefined inclusion criteria (stress, mental workload, anxiety, and stress-related arousal outcomes); additional references were retained for background (methods, datasets/benchmarks, bias, and wearable comparisons) and were excluded from PRISMA counts. The literature is synthesized across laboratory, academic, online-interaction, and telehealth contexts, with emphasis on signal-processing pipelines, physiological targets, validation strategies, and deployment conditions. The reviewed evidence indicates that rPPG supports robust non-contact heart-rate estimation, while HRV/PRV reliability degrades under head motion, illumination variability, and short analysis windows commonly used for stress inference. Many studies rely on protocol labels or wearable references without agreement against electrocardiography or high-quality contact photoplethysmography, constraining physiological validity. Based on these findings, this review provides a validity-oriented synthesis and reporting recommendations to support reliable rPPG-based stress and workload assessment in ecologically relevant settings.