From Tools to Teammates: Conceptualizing Humans’ Perception of Machines as Teammates with a Systematic Literature Review

Abstract

The accelerating capabilities of systems brought about by advances in Artificial Intelligence challenge the traditional notion of systems as tools. Systems' increasingly agentic and collaborative character offers the potential for a new user-system interaction paradigm: Teaming replaces unidirectional system use. Yet, extant literature addresses the prerequisites for this new interaction paradigm inconsistently, often not even considering the foundations established in human teaming literature. To address this, this study utilizes a systematic literature review to conceptualize the drivers of the perception of systems as teammates instead of tools. Hereby, it integrates insights from the dispersed and interdisciplinary field of human-machine teaming with established human teaming principles. The creation of a team setting and a social entity, as well as specific configurations of the machine teammate's collaborative behaviors, are identified as main drivers of the formation of impactful human-machine teams.

Foundational concepts in person-machine teaming

“…The question of whether a machine can truly team with people (or even other non-human agents) is a source of significant debate, and the term "team" is frequently misused or misapplied, especially with respect to person-machine teams. Research engineers often apply the term "human-machine team" to any collection of people and robots or autonomous agents, regardless of whether they meet the criteria that define a team, such as the need for interdependence between members or common identity as a team [7][8][9]. Frontiers in Physics frontiersin.org A team is a set of two or more people who interact dynamically, interdependently, and adaptively, toward a common and valued goal, each member having specific roles or functions to perform, and a limited life-span of membership [10].…”

My synthetic wingman must understand me: modelling intent for future manned–unmanned teaming

“…With advances in artificial intelligence, machine learning, and cognitive modelling, hybrid (or mixed) teaming vernaculars, such as human-robot collaboration (Bauer et al 2008) and human-machine cooperation (Hoc 2000), transition from merely conceptual to increasingly practical and applicable (O'Neill et al 2020). As such, there is a possibility for a new interaction paradigm where embedded or embodied synthetic agents are perceived as teammates rather than tools (Rix 2022;Wynne and Lyons 2018).…”

Modelling intent for Manned-Unmanned Teaming : Exploring human-centric approaches for future combat aircraft systems

“…With advances in artificial intelligence, machine learning, and cognitive modelling, the concept of humans and synthetic agents working as teammates transitions from merely conceptual to increasingly practical and applicable (O'Neill, McNeese, Barron, & Schelble, 2020). Following this transition, it has been argued that we are entering a new interaction paradigm, where embedded and embodied synthetic agents are no longer perceived as tools-but as teammates (Rix, 2022). Such hybrid teaming envisions taking advantage of the emergent effects resulting from humans and synthetic agents working collaboratively (Lyons, Sycara, Lewis, & Capiola, 2021).…”

“…As noted in this definition, Human-Autonomy Teaming literature often emphasises that synthetic agents ought to be perceived as teammates rather than tools, assistants or sub-ordinates. For instance, in a review by Rix (2022), it is pointed out that the increased capabilities offered by recent advances challenge the traditional perception of systems as tools, as these advances have the potential for a new interaction paradigm where they have agentic and collaborative characteristics. Similarly, O'Neill et al (2020) argue that synthetic agents should be perceived as having a degree of agency, expressed through independence, self-governance, and proactivity.…”