Mapping Meaning and Purpose in Human-Robot Teams: Anthropomorphic Agents in Military Operations

We spontaneously tend to project animacy and sensitivity to inanimate objects and sometimes we attribute distinctively human features like intelligence, goals, and reasons to certain artificial devices. This phenomenon is called “anthropomorphism” and has been long studied by researchers in human-robot interaction and social-robotics. These studies are particularly important from the perspective of recent developments in military technology, as autonomous systems controlled by AI are expected to play a greater and greater role in the future of warfare. Anthropomorphistic effects can play a critical role in tactical operations involving hybrid human-robot teams, where service members and autonomous agents need to quickly coordinate relying almost exclusively on fast, cognitively parsimonious, natural forms of communication. These forms rely importantly on anthropomorphism to allow human soldiers read the behavior of machines in terms of goals and intentions. Understanding the cognitive mechanisms that underpin anthropomorphistic attributions is hence potentially crucial to increase the accuracy and efficacy of human-machine interaction in military operations. However, this question is largely philosophical, as numerous models compete in the space of social cognition theory to explain behavior reading and mental states attribution. This paper aims to offer an initial exploration of these mechanisms from a perspective of philosophical psychology and cognitive philosophy, reviewing the theories in social cognition that are most promising to explain anthropomorphism and predict how it can enable and improve natural communication between soldiers and autonomous military technologies.