Gaze-informed signatures of trust and collaboration in human-autonomy teams
In the evolving landscape of human-autonomy teaming (HAT), fostering effective collaboration and trust between human and autonomous agents is increasingly important. To explore this, we used the game Overcooked AI to create dynamic teaming scenarios featuring varying agent behaviors (clumsy, rigid,...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Computers in Human Behavior: Artificial Humans |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2949882125000556 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850091508525432832 |
|---|---|
| author | Anthony J. Ries Stéphane Aroca-Ouellette Alessandro Roncone Ewart J. de Visser |
| author_facet | Anthony J. Ries Stéphane Aroca-Ouellette Alessandro Roncone Ewart J. de Visser |
| author_sort | Anthony J. Ries |
| collection | DOAJ |
| description | In the evolving landscape of human-autonomy teaming (HAT), fostering effective collaboration and trust between human and autonomous agents is increasingly important. To explore this, we used the game Overcooked AI to create dynamic teaming scenarios featuring varying agent behaviors (clumsy, rigid, adaptive) and environmental complexities (low, medium, high). Our objectives were to assess the performance of adaptive AI agents designed with hierarchical reinforcement learning for better teamwork and measure eye tracking signals related to changes in trust and collaboration. The results indicate that the adaptive agent was more effective in managing teaming and creating an equitable task distribution across environments compared to the other agents. Working with the adaptive agent resulted in better coordination, reduced collisions, more balanced task contributions, and higher trust ratings. Reduced gaze allocation, across all agents, was associated with higher trust levels, while blink count, scanpath length, agent revisits and trust were predictive of the human's contribution to the team. Notably, fixation revisits on the agent increased with environmental complexity and decreased with agent versatility, offering a unique metric for measuring teammate performance monitoring. This is one of the first studies to use gaze metrics such as revisits, gaze allocation, and scanpath length to predict not only trust, but also human contribution to teaming behavior in a real-time task with cooperative agents. These findings underscore the importance of designing autonomous teammates that not only excel in task performance but also enhance teamwork by being more predictable and reducing the cognitive load on human team members. Additionally, this study highlights the potential of eye-tracking as an unobtrusive measure for evaluating and improving human-autonomy teams, suggesting eye gaze could be used by agents to dynamically adapt their behaviors. |
| format | Article |
| id | doaj-art-d5b52ba3ae77453b8755fadb140e80db |
| institution | DOAJ |
| issn | 2949-8821 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Computers in Human Behavior: Artificial Humans |
| spelling | doaj-art-d5b52ba3ae77453b8755fadb140e80db2025-08-20T02:42:22ZengElsevierComputers in Human Behavior: Artificial Humans2949-88212025-08-01510017110.1016/j.chbah.2025.100171Gaze-informed signatures of trust and collaboration in human-autonomy teamsAnthony J. Ries0Stéphane Aroca-Ouellette1Alessandro Roncone2Ewart J. de Visser3U.S. Army DEVCOM, Army Research Laboratory, Aberdeen Proving Ground, MD, 21001, USA; Warfighter Effectiveness Research Center, U.S. Air Force Academy, CO, 80840, USA; Corresponding author. Warfighter Effectiveness Research Center 2354 Fairchild Drive, Suite 5L-29 USAF Academy, CO, 80840, USA.Department of Computer Science, University of Colorado Boulder, CO, 80309, USADepartment of Computer Science, University of Colorado Boulder, CO, 80309, USAWarfighter Effectiveness Research Center, U.S. Air Force Academy, CO, 80840, USAIn the evolving landscape of human-autonomy teaming (HAT), fostering effective collaboration and trust between human and autonomous agents is increasingly important. To explore this, we used the game Overcooked AI to create dynamic teaming scenarios featuring varying agent behaviors (clumsy, rigid, adaptive) and environmental complexities (low, medium, high). Our objectives were to assess the performance of adaptive AI agents designed with hierarchical reinforcement learning for better teamwork and measure eye tracking signals related to changes in trust and collaboration. The results indicate that the adaptive agent was more effective in managing teaming and creating an equitable task distribution across environments compared to the other agents. Working with the adaptive agent resulted in better coordination, reduced collisions, more balanced task contributions, and higher trust ratings. Reduced gaze allocation, across all agents, was associated with higher trust levels, while blink count, scanpath length, agent revisits and trust were predictive of the human's contribution to the team. Notably, fixation revisits on the agent increased with environmental complexity and decreased with agent versatility, offering a unique metric for measuring teammate performance monitoring. This is one of the first studies to use gaze metrics such as revisits, gaze allocation, and scanpath length to predict not only trust, but also human contribution to teaming behavior in a real-time task with cooperative agents. These findings underscore the importance of designing autonomous teammates that not only excel in task performance but also enhance teamwork by being more predictable and reducing the cognitive load on human team members. Additionally, this study highlights the potential of eye-tracking as an unobtrusive measure for evaluating and improving human-autonomy teams, suggesting eye gaze could be used by agents to dynamically adapt their behaviors.http://www.sciencedirect.com/science/article/pii/S2949882125000556 |
| spellingShingle | Anthony J. Ries Stéphane Aroca-Ouellette Alessandro Roncone Ewart J. de Visser Gaze-informed signatures of trust and collaboration in human-autonomy teams Computers in Human Behavior: Artificial Humans |
| title | Gaze-informed signatures of trust and collaboration in human-autonomy teams |
| title_full | Gaze-informed signatures of trust and collaboration in human-autonomy teams |
| title_fullStr | Gaze-informed signatures of trust and collaboration in human-autonomy teams |
| title_full_unstemmed | Gaze-informed signatures of trust and collaboration in human-autonomy teams |
| title_short | Gaze-informed signatures of trust and collaboration in human-autonomy teams |
| title_sort | gaze informed signatures of trust and collaboration in human autonomy teams |
| url | http://www.sciencedirect.com/science/article/pii/S2949882125000556 |
| work_keys_str_mv | AT anthonyjries gazeinformedsignaturesoftrustandcollaborationinhumanautonomyteams AT stephanearocaouellette gazeinformedsignaturesoftrustandcollaborationinhumanautonomyteams AT alessandroroncone gazeinformedsignaturesoftrustandcollaborationinhumanautonomyteams AT ewartjdevisser gazeinformedsignaturesoftrustandcollaborationinhumanautonomyteams |