Autonomous vehicle experiences can meaningfully affect human health—not through direct biological injury, but by altering cognitive processing, perceived safety, and anxiety-related arousal. A core health concept relevant to public adoption of robotaxis is anxiety. Anxiety is a state characterized by heightened threat appraisal, increased physiological arousal, and cognitive symptoms such as worry, hypervigilance, and difficulty concentrating. In the context of driverless systems, anxiety may arise when users cannot predict system behavior, cannot assume control, or perceive uncertainty about safety outcomes.
Anxiety mechanisms involve several interacting pathways. First is cognitive appraisal: people interpret ambiguous cues as potentially dangerous. In robotaxi rides, ambiguity can include unpredictable routing, frequent topic changes in attention, braking patterns, or limited ability to intervene. This increases the likelihood of threat-oriented interpretations, especially in individuals with pre-existing anxiety disorders or higher intolerance of uncertainty. Second is attentional bias: anxious individuals tend to preferentially allocate attention to threat-relevant signals. In an autonomous setting, threat signals may include roadway events, system prompts, or the absence of a human driver. Third is physiological arousal through the autonomic nervous system and stress hormone signaling. Elevated arousal can produce symptoms such as tachycardia, dyspnea sensation, gastrointestinal discomfort, and dizziness—symptoms that can be misattributed to the transport experience and reinforce anxiety.
Cognitive load also plays a role. Robotaxis change the traditional task demands of driving. Users may shift from continuous control to monitoring and trust calibration. Monitoring requires sustained attention to system status and environmental context, which can lead to mental fatigue. If the user simultaneously tries to override uncertainty (e.g., by tracking controls, estimating stopping distance, or anticipating edge cases), working memory demands increase. When cognitive resources are exceeded, performance and comfort degrade, which can be experienced subjectively as anxiety or panic.
Risk perception is a second major determinant of anxiety. People evaluate risk using more than actuarial probability; they weigh controllability, transparency, and personal agency. Humans generally feel safer when they can intervene. In a robotaxi, perceived loss of control can increase anxious vigilance even when objective safety is high. Trust calibration is therefore central: overly low trust may intensify anxiety and cause avoidance, while overly high trust without adequate monitoring may create vulnerability to unsafe decision-making in novel scenarios. Clinically, this is analogous to miscalibrated threat beliefs and impaired safety learning.
From a public-health perspective, designing an “intelligent driving experience” should consider anxiety reduction as a safety feature. Evidence-based approaches include improving system transparency (clear explanations of what the vehicle is doing and why), providing predictable behavioral patterns (consistent acceleration and braking profiles), and offering user-controllable reassurance mechanisms (e.g., accessible support, well-designed emergency procedures). Interfaces that reduce uncertainty—such as real-time status indicators, confidence cues, and straightforward override options when appropriate—can lower threat appraisal.
If anxiety symptoms become severe, differential diagnosis matters. Panic disorder involves recurrent panic attacks with abrupt surges of fear and physical symptoms. Phobias may develop when individuals associate transport with fear cues, leading to avoidance and anticipatory anxiety. Generalized anxiety disorder features pervasive worry that is difficult to control across domains, including technology-mediated safety. A normal transient stress response to a new technology should be distinguished from persistent anxiety causing impairment.
Management strategies for anxiety in autonomous mobility environments can be practical. Psychoeducation about system capabilities and limitations reduces uncertainty. Gradual exposure—beginning with short rides, familiar routes, and supportive guidance—can attenuate anticipatory anxiety through extinction learning. Cognitive-behavioral techniques such as identifying catastrophic thoughts (e.g., “the vehicle will fail”) and replacing them with balanced appraisals (“the system has redundancies and follows defined operating constraints”) can reduce cognitive distortions. Mindfulness-based attention training can reduce hypervigilant monitoring by anchoring attention to non-threatening present cues.
In clinical research terms, outcomes to measure include validated anxiety scales, physiological markers (heart rate variability, skin conductance), subjective comfort ratings, and trust calibration metrics. Importantly, perceived safety and anxiety are modifiable through human factors engineering and user education. As robotaxis expand in real-world pilots, integrating behavioral science into technical deployment can support healthier user experiences and improve adoption while reducing anxiety-driven barriers.
Source: [GeelyGroup] (original post about CaoCao robotaxi pilot operations and an “Intelligent Driving Experience”).
Geely Holding Group: Spotted on the streets outside Geely. Unmanned robotaxis from Geely Holding’s green mobility service provider, CaoCao begun pilot operations in Hangzhou. Through CaoCao app, users can now try out an “Intelligent Driving Experience” when visiting Geely.. #breaking
— @GeelyGroup May 1, 2026
SHOP AMAZON BEST SELLERS, CLICK TO BUY FROM AMAZON.
SHOP AMAZON BEST SELLERS, CLICK TO BUY FROM AMAZON.
