The use of cardiac-based metrics to assess secondary task engagement during automated and manual driving: An experimental simulation study

Luigi Di Stasi, Leandro; Angioi, Francesco; Fernandes, Marcelo; De Cet, Giulia; Jesus Caurcel, M; Stojmenova, Kristina; Sodnik, Jaka; Prat, Christophe; Díaz Piedra, Carolina · 2023 · Crossref

DOI: 10.54941/ahfe1004334

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the viability of cardiac-based metrics as alternative indicators for driver monitoring systems (DMS), addressing the limitations of current camera-based solutions in future automated driving (AD) environments. As AD vehicle interiors may feature rotated seats or obstructed views, camera-based gaze and head-position tracking may become ineffective. The researchers aimed to determine how engaging in non-driving-related tasks (NDRT) affects heart rate (HR) and heart rate variability, specifically the standard deviation of R-R intervals (SDRR), during both manual driving (MD) and AD supervision. The experiment utilized a 2 (Driving Modality: AD vs. MD) × 2 (Task Modality: one-hand vs. two-hand NDRT) within-participants design. Thirty-two expert drivers participated in a semi-dynamic driving simulator, completing two highway scenarios of approximately 22 minutes each. During four 5-minute distraction periods per scenario, participants performed concurrent tasks on a tablet: a one-hand task involving arithmetic operations via SMS and a two-hand task requiring bimanual coordination using a mobile application. Cardiac activity was recorded via electrocardiogram (ECG) using wearable electrodes. Data analysis included driving performance metrics (speed, lateral position, speeding time), NDRT performance scores, subjective task complexity ratings, and physiological indices (HR and SDRR). Statistical analyses employed repeated measures ANOVA and dependent samples t-tests with Holm-Bonferroni corrections. The results demonstrated that cardiac metrics effectively differentiated task demands. The two-hand task induced significantly higher HR and SDRR compared to the one-hand task, regardless of driving modality. Driving modality significantly influenced HR, which was higher during MD than AD, but did not significantly affect SDRR. Behavioral and subjective data validated the experimental manipulation: the two-hand task during MD was the most disruptive, resulting in poorer driving control (higher speed variation and lateral deviation) and lower NDRT performance. Participants also reported higher subjective task complexity for the two-hand task, particularly during MD. Notably, drivers self-regulated by driving slower during the demanding two-hand MD condition, leading to less time spent speeding despite reduced control accuracy. These findings suggest that cardiac-based indices, particularly HR and SDRR, can accurately reflect driver engagement and cognitive load during multitasking. The study supports the development of non-camera-based DMS, such as radar-based sensors embedded in seats, which are compatible with future AD interior designs. By providing reliable physiological measures of driver state, these technologies could enhance road safety and facilitate effective driver-vehicle interactions in automated systems. The authors note that while HR showed sensitivity to both task and driving modality, SDRR was primarily sensitive to task complexity, highlighting the need for further research into optimal time windows and alternative ECG metrics for real-world application.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• workload measurement
• manual
• distraction detection algorithms
• temporal
• visual manual
• hands on hands off engagement

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data, behavioral performance data
• Methodological Resource: validation psychometrics