Michael Long

Can one perform an intentional action without successfully performing it? On the face of it, this seems like a silly question. One can’t intentionally do A without actually doing A. An underlying assumption that many philosophers of action would share would be that if we have an intentional action, we have an action. Thus, many philosophers of action engaged in experimental work compose vignettes that feature a certain action, and then directly ask the participants to rate their agreement with a statement classifying the action as intentional. Due to the common assumption above, this method is seen as unproblematic. However, if faced with evidence that ordinary English users are comfortable with saying that an agent ‘intentionally’ performed an action despite the agent’s failure to successfully perform it, what lesson should philosophers of action take? We designed a study to investigate whether there are instances in which participants would agree that an agent has intentionally performed an action, even though the agent’s performance was unsuccessful. The results of our study indicate that English users are comfortable with saying that an agent intentionally performed an action despite the fact that the action did not occur. We posit two possible conclusions: either non-specialists are in vast error about the conception of intentional action, or the results garnered from experimental methods that ask directly for intentionality ascriptions are unreliable due to the flexible nature of ordinary language. We argue that the latter is the more reasonable conclusion, and suggest that a popular methodology in experimental philosophy of action should be re-evaluated.

Effective human–robot teaming increasingly requires humans to work with intelligent, autonomous machines. However, novel features of intelligent autonomous systems such as social agency and incomprehensibility may influence the human’s trust in the machine. The human operator’s mental model for machine functioning is critical for trust. People may consider an intelligent machine partner as either an advanced tool or as a human-like teammate. This article reports a study that explored the role of individual differences in the mental model in a simulated environment. Multiple dispositional factors that may influence the dominant mental model were assessed. These included the Robot Threat Assessment, which measures the person’s propensity to apply tool and teammate models in security contexts. Participants were paired with an intelligent robot tasked with making threat assessments in an urban setting. A transparency manipulation was used to influence the dominant mental model. For half of the participants, threat assessment was described as physics-based ; the remainder received transparency information that described psychological cues. We expected that the physics-based transparency messages would guide the participant toward treating the robot as an advanced machine, while psychological messaging would encourage perceptions of the robot as acting like a human partner. We also manipulated situational danger cues present in the simulated environment. Participants rated their trust in the robot’s decision as well as threat and anxiety, for each of 24 urban scenes. They also completed the RoTA and additional individual-difference measures. Findings showed that trust assessments reflected the degree of congruence between the robot’s decision and situational danger cues, consistent with participants acting as Bayesian decision makers. Several scales, including the RoTA, were more predictive of trust when the robot was making psychology-based decisions, implying that trust reflected individual differences in the mental model of the robot as a teammate. These findings suggest scope for designing training that uncovers and mitigates the individual’s biases toward intelligent machines.