Teammates Instead of Tools: The Impacts of Level of Autonomy on Mission Performance and Human–Agent Teaming Dynamics in Multi-Agent Distributed Teams
DOI: 10.3389/frobt.2022.782134
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 impact of varying Levels of Autonomy (LOAs) on mission performance, trust, and team dynamics in human–agent teaming (HAT) within multi-agent distributed teams. As unmanned aerial systems (UAS) evolve from simple tools to collaborative teammates, research has shifted from determining how many agents a human can manage to how humans and agents can coordinate effectively. The authors address the gap in understanding how different LOAs—specifically regarding decision-making authority—affect operator workload, stress, trust, and overall mission effectiveness in complex intelligence, surveillance, and reconnaissance (ISR) scenarios. The researchers employed a repeated-measures within-subjects design with 41 participants who completed simulated ISR missions using a custom-built multi-UAV simulator. Participants monitored four autonomous UAV agents tasked with detecting and classifying targets as friendly, enemy, or neutral. The independent variable was the LOA, manipulated across four conditions: Manual (Level 1, no assistance), Advice (Level 4, agent suggests classification), Consent (Level 5, agent classifies but requires human approval), and Veto (Level 7, agent classifies and executes, allowing human override). The agents operated with approximately 92% reliability in detection and classification. Dependent variables included target identification performance, operator stress and workload, trust in the agent, and perceived team effectiveness. The study found that the level of autonomy significantly influenced human–agent teaming dynamics. Specifically, the Consent condition (Level 5), where agents proposed decisions requiring human approval, yielded the highest performance gains compared to Manual or fully autonomous Veto conditions. This middle-ground LOA balanced the need for agent assistance with the necessity of keeping the human operator engaged in the decision-making loop. In contrast, the Veto condition, while reducing immediate cognitive load, risked operator disengagement and reduced situation awareness, potentially leading to overreliance on automation. The Manual condition resulted in higher operator workload and stress due to the lack of agent assistance. Trust levels varied by LOA, with the Consent condition fostering higher calibrated trust compared to conditions where agents acted without direct human confirmation. These findings suggest that "teammates instead of tools" requires careful calibration of autonomy to optimize human–agent collaboration. The results indicate that mixed-initiative systems, particularly those requiring human consent for agent actions, are superior for maintaining operator engagement and mission performance in multi-agent environments. This challenges the assumption that higher autonomy always reduces workload effectively, highlighting that removing humans from the decision loop can degrade team effectiveness. The study provides critical guidance for designing future HAT systems, emphasizing that optimal LOAs depend on balancing agent capability with human supervisory control to ensure trust, situational awareness, and operational success.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-08-09 |
| archive | success | canonical_url | — | — | 1 | 2026-08-09 |
| extract | success | pdftotext | — | — | 4 | 2026-08-10 |
| clean | success | clean | — | — | 2 | 2026-08-10 |
| chunk | success | chunk | — | — | 2 | 2026-08-10 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 2 | 2026-08-10 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.6-27b-nvidia | summ-v5 | 2 | 2026-08-10 |
| tag | success | vector_similarity | — | — | 17 | 2026-08-11 |
| verify | partial | — | — | — | 2 | 2026-08-10 |
Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
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: self report data