Social robots serve people as tutors, caretakers, receptionists, companions, and other social agents. People know that the robots are mechanical artifacts, yet they interact with them as if they were actual agents. How is this possible? The proposal here is that people construe social robots not as social agents per se, but as depictions of social agents. They interpret them much as they interpret ventriloquist dummies, hand puppets, virtual assistants, and other interactive depictions of people and animals. Depictions as a class consist of three physical scenes with part-by-part mappings between them: (a) a base scene (the raw physical artifact), (b) the depiction proper (the artifact construed as a depiction), and (c) the scene depicted (the scene people are to imagine). With social robots, evidence shows people form the same three scenes plus mappings: They perceive the raw machinery of a robot, construe it as a depiction of a character, and, using the depiction as a guide, engage in the pretense that they are interacting with the character depicted. With social robots, people also recognize three classes of agents – the characters depicted, the intended recipients of the depictions (those who view or interact with the robots), and the authorities responsible for the robots (the designers, makers, and owners). Construing social robots as depictions, we argue, accounts for many phenomena not covered by alternative models.

Theories are an integral part of the scientific endeavour. The target article proposes interesting ideas for a theory on human–robot interaction but lacks specificity that would enable us to properly test this theory. No empirical data are yet available to determine its predictive power.

Clark and Fischer (C&F) claim that trait attribution has major limitations in explaining human–robot interactions. We argue that the trait attribution approach can explain the three issues posited by C&F. We also argue that the trait attribution approach is parsimonious, as it assumes that the same mechanisms of social cognition apply to human–robot interaction.

Following the depiction theory by Clark and Fischer we would expect people interacting with robots to experience fictional emotions akin to those toward films or novels. However, some people's emotional reactions toward robots display the motivational force typical to non-fictional emotions. We discuss this incongruity and offer two suggestions on how to explain it while maintaining the depiction theory.

Clark and Fischer analyze social robots as interactive depictions, presenting characters that people can interact with in social settings. Unlike other types of depictions, the props for social robot depictions depend on emerging interactive technologies. This raises questions about how such depictions depict: They conflate character and prop in ways that delight, confuse, mistreat, and may become ordinary human–technology interactions.

Clark and Fischer's dismissal of extant human–robot interaction research approaches limits opportunities to understand major variables shaping people's engagement with social robots. Instead, this endeavour categorically requires multidisciplinary approaches. We refute the assumption that people cannot (correctly or incorrectly) represent robots as autonomous social agents. This contradicts available empirical evidence, and will become increasingly tenuous as robot automation improves.

Clark and Fischer's three levels of depiction of social robots can be conceptualized as cognitive schemas. When interacting with social robots, humans shift between schemas similarly to how they shift between identity category schemas when interacting with other humans. Perception of mind, context cues, and individual differences underlie perceptions of which level of depiction is most situationally relevant.

When people interact with social robots, they treat them as real social agents. How people depict robots is fun to consider, but when people are confronted with embodied entities that move and talk – whether humans or robots – they interact with them as authentic social agents with minds, and not as mere representations.

We outline two points of criticism. Firstly, we argue that robots do constitute a separate category of beings in people's minds rather than being mere depictions of non-robotic characters. Secondly, we find that (semi-)automatic processes underpinning communicative interaction play a greater role in shaping robot-directed speech than Clark and Fischer's theory of social robots as depictions indicate.

Do people hold robots responsible for their actions? While Clark and Fischer present a useful framework for interpreting social robots, we argue that they fail to account for people's willingness to assign responsibility to robots in certain contexts, such as when a robot performs actions not predictable by its user or programmer.

The target article proposes that people perceive social robots as depictions rather than as genuine social agents. We suggest that people might instead view social robots as social agents, albeit agents with more restricted capacities and moral rights than humans. We discuss why social robots, unlike other kinds of depictions, present a special challenge for testing the depiction hypothesis.

Clark and Fischer (C&F) discuss how people interact with social robots in the context of a general analysis of interaction with characters. I suggest that a consideration of aesthetic illusion would add nuance to this analysis. In addition, I illustrate how people's experiences with other depictions of characters require adjustments to C&F's claims.

We suggest that as people move to construe robots as social agents, interact with them, and treat them as capable of social ties, they might develop (close) relationships with them. We then ask what kind of relationships can people form with bots, what functions can bots fulfill, and what are the societal and moral implications of such relationships.

Clark and Fischer argue that people see social robots as depictions of social agents. However, people's interactions with virtual assistants may change their beliefs about social robots. Children and adults with exposure to virtual assistants may view social robots not as depictions of social agents, but as social agents belonging to a unique ontological category.

In the target article, Clark and Fischer argue that little is known about children's perceptions of social robots. By reviewing the existing literature we demonstrate that infants and young children interact with robots in the same ways they do with other social agents. Importantly, we conclude children's understanding that robots are artifacts (e.g., not alive) develops gradually during the preschool years.

Clark and Fischer propose that people interpret social robots not as social agents, but as interactive depictions. Drawing on research focusing on how children selectively learn from social others, we argue that children do not view social robots as interactive toys but instead treat them as social learning partners and critical sources of information.

Social robots have limited social competences. This leads us to view them as depictions of social agents rather than actual social agents. However, people also have limited social competences. We argue that all social interaction involves the depiction of social roles and that they originate in, and are defined by, their function in accounting for failures of social competence.

While we applaud the careful breakdown by Clark and Fischer of the representation of social robots held by the human user, we emphasise that a neurocognitive perspective is crucial to fully capture how people perceive and construe social robots at the behavioural and brain levels.

The authors at times propose that robots are mere depictions of social agents (a philosophical claim) and at other times that people conceive of social robots as depictions (an empirical psychological claim). We evaluate each claim's accuracy both now and in the future and, in doing so, we introduce two dangerous misperceptions people have, or will have, about social robots.

Clark and Fischer's depiction hypothesis is based on examples of western mimetic art. Yet social robots do not depict social interactions, but instead perform them. Similarly, dance and performance art do not rely on depiction. Kinematics and expressivity are better predictors of dance aesthetics and of effective social interactions. In this way, social robots are more like dancers than actors.