Can robots lie? A posthumanist approach to robotic animals and deceptive practices in dementia care.

Robotic animals are designed to resemble real, living animals, but at the same time, dementia care guidelines and policies often emphasize the value of transparency in relation to robots-people should not be led to believe that robots have capacities that they in fact lack. However, it is not obvious how to separate truth from lies in everyday care practice. Based on participant observations and interviews with certified assistant nurses and nursing assistants in Swedish nursing homes for people with dementia, this article studies how robotic animals become "real" in care practice. The article takes a posthumanist approach to the co-constitution of aging, care, and technology-a perspective that recognizes that not only care staff and nursing home residents, but also robots and other material things, can take an active part in shaping care practice. The analysis results in four typical situations out of which robotic animals emerge as real, living animals: the cuddle, with its simple but dynamic embodied actions; the comfort, where the robot is used as a resource for distraction and emotional support; the conversation, by which robotic animals' agency is both established and challenged; and the adoption, through which narratives and props are used to establish the robot as a pet. Robots cannot lie, at least not by themselves; instead, robots' deceptive potential is enabled by a network of actors, which is why it is often difficult to draw a clear line between lying and care workers' empathic following.