Shared control versus traded control in driving: a debate around automation pitfalls

Winter, Joost de; Petermeijer, Sebastiaan M.; Abbink, David A. · 2022 · openalex

DOI: 10.1080/00140139.2022.2153175

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the critical design dilemma in human-automation interaction: whether driving functions should be "traded" (alternating control between human and automation) or "shared" (simultaneous, congruent control). Motivated by the increasing viability of automation in unstructured environments like driving, the authors present a structured debate reflecting on the pitfalls of automation and the merits of each control strategy. The discussion is framed around six established pitfalls of automation—loss of situation/mode awareness, deskilling, unbalanced mental workload, behavioral adaptation, misuse, and disuse—which have historically been associated with traded control systems. The methodology is theoretical and reflective rather than empirical. The authors synthesize existing literature and insights from a decade of classroom debates to construct opposing arguments. One author argues that haptic shared control, where the driver receives guiding torques while maintaining direct input coupling, remedies automation pitfalls by keeping the human "in the loop." A second author rebuts this, arguing that traded control is superior for improving road safety. The paper defines these concepts sharply, distinguishing shared control (congruent perception-action cycles) from traded control (supervisory monitoring with intermittent takeover), and clarifies that many systems labeled as shared in literature are actually traded. The findings highlight that neither approach is universally superior; rather, their effectiveness depends on environmental complexity. The authors conclude that shared control outperforms traded control in environments with medium complexity, as it maintains driver engagement and mitigates issues like deskilling and loss of situation awareness. Conversely, traded control is more effective in low-complexity environments, where automation can safely handle the task without requiring constant human input. The paper details how traded control leads to specific risks: low workload leading to sudden terror during failures (unbalanced workload), skill erosion due to lack of practice (deskilling), and behavioral adaptation where users compensate for safety features by taking greater risks. It also notes that shared control does not necessarily improve performance metrics compared to manual control but serves to preserve manual skills and awareness. The significance of this work lies in providing a nuanced framework for designers of automated driving systems. By moving beyond the binary view of automation as either fully manual or fully autonomous, the paper offers evidence-based guidance on function allocation. It suggests that the choice between sharing and trading control should be context-dependent, tailored to the specific operational design domain and complexity of the driving environment. This reflection aims to stimulate further research and help guide the development of safer human-automation interfaces, addressing long-standing concerns about automation pitfalls such as mode errors and overtrust.

Key finding

Shared control and traded control each outperform the other depending on environmental complexity, with shared control being advantageous in medium complexity and traded control in low complexity.

Methodology

review

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. Discovered via openalex_abstract on 2026-05-08.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

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

• adas effectiveness