Individuals perceive phantom costs, such as ulterior motives and risks, when a person makes an unreasonably generous offer without sufficient explanation. Prior research relying exclusively on the Nao robot found similar effects, though smaller than with humans. To better understand these differences, we manipulated agent human-likeness across five robots and a human. Participants read a vignette in which the agent either offered a free parking spot (reasonable) or added an unjustified $10 incentive (unreasonably generous). They then decided whether to accept the offer, explained why they thought the agent made this offer, and rated the agent's anthropomorphism and perceived phantom costs. Results showed that unreasonably generous offers prompted participants to attribute more mind to the agents, increasing perceived phantom costs. Anthropomorphism also influenced phantom costs: Animacy and Disturbance increased them, while Intentionality and Sociability decreased them. This study advances our knowledge of phantom costs in HRI, suggesting that people adopt the intentional stance to explain a robot's behaviour—especially when it deviates from social norms—highlighting the need for careful anthropomorphic design of social robots to minimize phantom costs perception.

Recent studies found that people imagine phantom costs—bad intentions and risks—when a human or a robot makes an overly generous offer without sufficient explanations. However, these studies used a paradigm in which the agent justified a cookie with $2 offer by saying they had eaten cookies with friends, a scenario we think implausible for robots. The present study replicated this paradigm while measuring perceived plausibility of the agent's justification. Results indicated that, unlike humans, the justification of eating cookies with friends was perceived as implausible when said by a robot. This perceived implausibility increased perceived phantom costs, reduced trust in the robot, and decreased offer acceptance. This study suggests that phantom costs occur when explanations are both insufficient and implausible, highlighting the need for sufficient and plausible explanations to promote effective HRI.

A recent psychology study found that people are more likely to reject an overly generous offer from an agent (human or robot) when there is no clear justification. This “money backfire effect” is said to stem from perceived phantom costs (e.g., agent's bad intentions). However, their online version provided less clear results. To test its robustness, we conducted a high-powered direct replication study aiming to know whether money backfiring and phantom costs effects occur in human-robot interaction. A human or robot offered participants a cookie with or without money. Participants had to decide whether they would eat the cookie and answer some questions regarding the scenario. Results indicated that participants perceived more phantom costs when the offer included money, negatively influencing their decisions to eat the cookie. However, participant's likelihood to eat the cookie did not differ as a function of agent and offer. This study provides new insights regarding the validity of online studies. We advise scholars to adapt their study setting regarding the effect they want to explore.

The heuristic of sufficient explanation (HOSE) is a process by which people learn hidden information about other agents. If the publicly available reasons for behaviour seem insufficient to explain the agent's behaviour, people look for hidden reasons that would explain it. I will present evidence for HOSE across several contexts, including economic decision-making, belief in conspiracy theories, judgments of other agents’ bad intentions, and romantic attraction. I will discuss implications for Human-Agent Interaction more broadly, including for perceptions of non-human agents’ motives.

Robots that are capable of outperforming human beings on mental and physical tasks provoke perceptions of threat. In this article we propose that implicit self-theory (core beliefs about the malleability of self-attributes, such as intelligence) is a determinant of whether one person experiences threat perception to a greater degree than another. We test for this possibility in a novel experiment in which participants watched a video of an apparently autonomous intelligent robot defeating human quiz players in a general knowledge game. Following the video, participants received either social comparison feedback, improvement-oriented feedback, or no feedback, and were then given the opportunity to play against the robot. We show that those who adopt a malleable self-theory (incremental theorists) are more likely to play against a robot after imagining losing to it, as well as exhibit more favorable responses and less identity threats than entity theorists (those adopting a fixed self-theory). Moreover, entity theorists (vs. incremental theorists) perceive autonomous intelligent robots to be significantly more threatening (both in terms of realistic and identity threats). These findings offer novel theoretical and practical implications, in addition to enriching the HRI literature by demonstrating that implicit self-theory is, in fact, an influential variable underpinning perceived threat.