Deceleration Assistance Mitigated the Trade-off Between Sense of Agency and Driving Performance

Wen Wen, Sonmin Yun, Atsushi Yamashita, Brandon D. Northcutt, Hajime AsamaFrontiers in Psychology 2021 · DOI

Abstract

Driving assistance technology has gained traction in recent years and is becoming more widely used in vehicles. However, drivers usually experience a reduced sense of agency when driving assistance is active even though automated assistance improves driving performance by reducing human error and ensuring quick reactions. The present study examined whether driving assistance can maintain human sense of agency during early deceleration in the face of collision risk, compared with manual deceleration. In the experimental task, participants decelerate their vehicle in a driving simulator to avoid collision with a vehicle that suddenly cut in front of them and decelerated. In the assisted condition, the system performed deceleration 100 ms after the cut-in. Participants were instructed to decelerate their vehicle and follow the vehicle that cut-in. This design ensured that the deceleration assistance applied a similar control to the vehicle as the drivers intended to, only faster and smoother. Participants rated their sense of agency and their driving performance. The results showed that drivers maintained their sense of agency and improved driving performance under driving assistance. The findings provided insights into designing driving assistance that can maintain drivers' sense of agency while improving future driving performance. It is important to establish a mode of joint-control in which the system shares the intention of human drivers and provides improved execution of control.

Introduction

As vehicles move toward higher levels of automation, engineers face a significant psychological hurdle: the trade-off between system performance and driver engagement. Typically, as a system intervenes more to correct human error or improve reaction times, the driver's "Sense of Agency" (SoA)—the subjective feeling that their own actions caused an external event—diminishes. This feeling of being "in the loop" is crucial; if SoA drops too low, drivers may become disengaged, leading to safety risks if human intervention is suddenly required. This study investigates whether "joint-control," where a machine executes a human's intent more efficiently, can maintain SoA while improving safety.

Methods

The researchers conducted a controlled experiment using a high-fidelity driving simulator (CarSim and MATLAB Simulink) with 19 healthy male participants (mean age 22.4).

Figure 1: The experimental driving simulator setup (A) and the timeline of the trial (B) showing the onset of overtaking and the deceleration assistance window.

In the experimental scenario, a "cut-in" vehicle suddenly moved in front of the participant and decelerated. The study utilized a 2x2 factorial design, varying the driving mode (Manual vs. Assisted) and the risk level (High vs. Low risk). In the "Assisted" condition, the system automatically applied brakes 100ms after the lead vehicle slowed down—significantly faster than the typical human reaction time of ~700ms. Crucially, participants were not informed when the assistance was active, and the system was designed to mimic the driver's likely intention to slow down rather than override it.

Key Results

The study measured both subjective ratings and objective performance metrics, including the Distance Coefficient of Variation (DCV) and Speed Coefficient of Variation (SCV).

  • Agency Maintained: There was no significant difference in SoA ratings between manual and assisted driving. Drivers felt equally in control in both modes.
  • Performance Boost: Objective performance improved significantly in the assisted condition, with a lower SCV ($p < .001$), indicating a smoother transition to the following speed.
  • Subjective Perception: Drivers rated their own performance significantly higher in assisted trials ($p = .026$). Interestingly, they perceived the improved handling but attributed it to their own skill rather than the hidden assistance.
  • Risk Resilience: While higher risk levels made maintaining distance (DCV) more difficult, they did not negatively impact the sense of agency when assistance was provided.

Discussion and Conclusion

The findings suggest that "shared intention" is the key to maintaining a driver's sense of agency. Because the braking assistance executed exactly what the drivers intended—only faster and smoother—the drivers "claimed" the action as their own. This suggests that the conflict between human intent and machine action, rather than the assistance itself, is what typically reduces agency.

While the study is limited by a small, gender-specific participant pool and the use of a fixed-base simulator, it provides a roadmap for future driving assistance. By designing systems that act as an extension of human intent, engineers can improve vehicle safety and performance without making the driver feel like a passive passenger.

Summary updated

Abstract

Driving assistance technology has gained traction in recent years and is becoming more widely used in vehicles. However, drivers usually experience a reduced sense of agency when driving assistance is active even though automated assistance improves driving performance by reducing human error and ensuring quick reactions. The present study examined whether driving assistance can maintain human sense of agency during early deceleration in the face of collision risk, compared with manual deceleration. In the experimental task, participants decelerate their vehicle in a driving simulator to avoid collision with a vehicle that suddenly cut in front of them and decelerated. In the assisted condition, the system performed deceleration 100 ms after the cut-in. Participants were instructed to decelerate their vehicle and follow the vehicle that cut-in. This design ensured that the deceleration assistance applied a similar control to the vehicle as the drivers intended to, only faster and smoother. Participants rated their sense of agency and their driving performance. The results showed that drivers maintained their sense of agency and improved driving performance under driving assistance. The findings provided insights into designing driving assistance that can maintain drivers' sense of agency while improving future driving performance. It is important to establish a mode of joint-control in which the system shares the intention of human drivers and provides improved execution of control.
BibTeX
@inproceedings{wen2021deceleration,
  author = {Wen Wen and Sonmin Yun and Atsushi Yamashita and Brandon D. Northcutt and Hajime Asama},
  title = {{Deceleration Assistance Mitigated the Trade-off Between Sense of Agency and Driving Performance}},
  booktitle = {Frontiers in Psychology},
  year = {2021},
  doi = {10.3389/fpsyg.2021.643516},
}
Figure