.png){kind=icon}
.png){kind=icon}
Return
A human-centered taxonomy of driver responses to cyber-attacks in automated vehicles
.png){kind=icon}
PRE
AI
G
K
U
M
DOI：10.1016/j.trf.2026.103595.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Automated vehicles (AVs) are increasingly vulnerable to cybersecurity threats, yet the human-factors dimensions of such threats remain under-explored. This study develops a human-centered taxonomy of driver responses to AV cyber-attacks, grounded in semi-structured interviews with thirteen domain experts. Using an iterative coding process grounded in theory, we identified four sequential stages of driver responses: (1) Perception, (2) Comprehension, (3) Decision-Making, and (4) Action, along with cross-cutting Integrative Factors. Within each stage, findings are organized around (a) envisioned scenarios, (b) limitations & barriers, (c) interventions & design factors, and (d) implications & responsibilities. Results highlight challenges in detecting subtle anomalies, interpreting irregular cues, and making timely decisions under uncertainty, with trust dynamics, workload, and preparedness shaping outcomes across stages. The taxonomy advances theoretical understanding by extending established models of situation awareness and trust calibration into adversarial contexts. Practically, it offers a structured roadmap for designing alerts, training programs, and policy measures that support calibrated human interventions and strengthen resilience in automated driving.
Keywords:
ADAPTIVE CRUISE CONTROL
TRUST
MODEL
PERFORMANCE
BEHAVIOR
Journal
IF:
4.4
Papers:
3.3K
Citations:
1.3W
Organization
Cited Papers