Jakub Złotowski

Anthropomorphism: Opportunities and Challenges in Human-Robot Interaction

Anthropomorphism is a phenomenon that describes the human tendency to see human-like shapes in the environment. It has considerable consequences for people’s choices and beliefs. With the increased presence of robots, it is important to investigate the optimal design for this technology. In this paper we discuss the potential benefits and challenges of building anthropomorphic robots, from both a philosophical perspective and from the viewpoint of empirical research in the fields of human–robot interaction and social psychology. We believe that this broad investigation of anthropomorphism will not only help us to understand the phenomenon better, but can also indicate solutions for facilitating the integration of human-like machines in the real world.

Dimensions of anthropomorphism: from humanness to humanlikeness

In HRI anthropomorphism has been considered to be a unidimensional construct. However, social psychological studies of the potentially reverse process to anthropomorphisation - known as dehumanization - indicate that there are two distinct senses of humanness with different consequences for people who are dehumanized by deprivation of some of the aspects of these dimensions. These attributes are crucial for perception of others as humans. Therefore, we hypothesized that the same attributes could be used to anthropomorphize a robot in HRI and only a two-dimensional measures would be suitable to distinguish between different forms of making a robot more humanlike. In a study where participants played a quiz based on the TV show “Jeopardy!” we manipulated a NAO robot’s intelligence and emotionality. The results suggest that only emotionality, not intelligence, makes robots be perceived as more humanlike. Furthermore, we found some evidence that anthropomorphism is a multidimensional phenomenon.Categories and Subject DescriptorsI.2.9 [Computing Methodologies]: ARTIFICIAL INTELLIGENCE—Robotics

Persistence of the uncanny valley: the influence of repeated interactions and a robot's attitude on its perception

The uncanny valley theory proposed by Mori has been heavily investigated in the recent years by researchers from various fields. However, the videos and images used in these studies did not permit any human interaction with the uncanny objects. Therefore, in the field of human-robot interaction it is still unclear what, if any, impact an uncanny-looking robot will have in the context of an interaction. In this paper we describe an exploratory empirical study using a live interaction paradigm that involved repeated interactions with robots that differed in embodiment and their attitude toward a human. We found that both investigated components of the uncanniness (likeability and eeriness) can be affected by an interaction with a robot. Likeability of a robot was mainly affected by its attitude and this effect was especially prominent for a machine-like robot. On the other hand, merely repeating interactions was sufficient to reduce eeriness irrespective of a robots embodiment. As a result we urge other researchers to investigate Moris theory in studies that involve actual human-robot interaction in order to fully understand the changing nature of this phenomenon.

Can we control it? Autonomous robots threaten human identity, uniqueness, safety, and resources

Abstract Emergence of autonomous machines is a hotly debated topic in mass-media. However, previous research has not empirically investigated whether the perceived autonomy of robots affects their social acceptance. In this study we examined the impact of perceived robot autonomy on realistic threats (threats to human jobs, resources and safety) and identity threats (threats to human identity and distinctiveness), attitude toward robots, and support for robotics research. US based participants watched a video of robots performing various tasks – these robots were presented as either autonomous and capable of disregarding human commands or non-autonomous and only capable of following human commands. Participants who watched videos of supposedly autonomous robots perceived robots in general to be significantly more threatening to humans (both realistic and identity threats) than those who watched a video of non-autonomous robots. Furthermore, exposure to autonomous robots evoked stronger negative attitude towards robots in general and more opposition to robotics research than exposure to non-autonomous robots. Both realistic and identity threats mediated the increase in negative attitudes toward robots and opposition to robotics research, although realistic threats were often the stronger mediator of the two. Our findings have practical implications for research on AI and open new questions on the relationship between robot autonomy and their social impact.

The inversion effect in HRI: are robots perceived more like humans or objects?

The inversion effect describes a phenomenon in which certain types of images are harder to recognize when they are presented upside down compared to when they are shown upright. Images of human faces and bodies suffer from the inversion effect whereas images of objects do not. The effect may be caused by the configural processing of faces and body postures, which is dependent on the perception of spatial relations between different parts of the stimuli. We investigated if the inversion effect applies to images of robots in the hope of using it as a measurement tool for robots anthropomorphism. The results suggest that robots, similarly to humans, are subject to the inversion effect. Furthermore, there is a significant, but weak linear relationship between the recognition accuracy and perceived anthropomorphism. The small variance explained by the inversion effect renders this test inferior to the questionnaire based Godspeed Anthropomorphism Scale.

Understanding Anthropomorphism: Anthropomorphism is not a Reverse Process of Dehumanization

Anthropomorphism plays an important role in human interaction with robots. However, our understanding of this phenomenon is still limited. In the previous research, we proposed to look at the work on dehumanization in order to understand what factors can affect a robot’s anthropomorphism. Moreover, considering that there are two distinct dimensions of humanness, a two-dimensional model of anthropomorphism was proposed. In this paper we present a study in which we manipulated the perceived intentionality of a robot and appearance (Robovie R2 vs Geminoid HI-2), and measured how they affected the anthropomorphization of the robots on two dimensions of humanness. We did not find statistically significant differences in attribution of human traits and mind along two dimensions of humanness. However, after dividing the traits based on their valence, we found that Geminoid HI-2 was attributed significantly more negative human traits than Robovie R2. These results do not support the proposed two-dimensional model of anthropomorphism.

Stop! That is close enough. How body postures influence human-robot proximity

In this paper we present a study that investigates human-robot interpersonal distances and the influence of posture, either sitting or standing on the interpersonal distances. The study is based on a human approaching a robot and a robot approaching a human, in which the human/robot maintain either a sitting or standing posture while being approached. We collected and analysed data from twenty-two participants and the results revealed that robot posture has a significant impact on the interpersonal distances in human-robot interactions. Previous interactions with a robot, and lower negative attitudes towards robots also impacted interpersonal distances. Although the effects of gender, height and age did not yield significant results, we discuss their influence on the interpersonal distances between humans and robots and how they are of interest for future research. We present design implications for human-robot interaction research and humanoid robot design.

Appearance of a Robot Affects the Impact of its Behaviour on Perceived Trustworthiness and Empathy

Abstract An increasing number of companion robots have started reaching the public in the recent years. These robots vary in their appearance and behavior. Since these two factors can have an impact on lasting human–robot relationships, it is important to understand their effect for companion robots.We have conducted an experiment that evaluated the impact of a robot’s appearance and its behaviour in repeated interactions on its perceived empathy, trustworthiness and anxiety experienced by a human. The results indicate that a highly humanlike robot is perceived as less trustworthy and empathic than a more machinelike robot. Moreover, negative behaviour of a machinelike robot reduces its trustworthiness and perceived empathy stronger than for highly humanlike robot. In addition, we found that a robot which disapproves of what a human says can induce anxiety felt towards its communication capabilities. Our findings suggest that more machinelike robots can be more suitable as companions than highly humanlike robots. Moreover, a robot disagreeing with a human interaction partner should be able to provide feedback on its understanding of the partner’s message in order to reduce her anxiety.

I Sing the Body Electric An experimental theatre play with robots [video abstract]

Robots seem to be natural candidates to be used as actors in theatre plays. The interest in robotic actors in performance led to several publications [1-3] and was also included in HRI 2012 conference video session [4], [5].

Persistence of the Uncanny Valley

In recent years, the uncanny valley theory has been heavily investigated by researchers from various fields. However, the videos and images used in these studies do not permit any human interaction with the uncanny objects. Therefore, in the field of human–robot interaction, it is still unclear what impact, if any, an uncanny-looking robot will have in the context of an interaction. In this paper, we describe an exploratory empirical study using a live interaction paradigm that involves repeated interactions with robots that differ in embodiment and their attitude toward humans. We find that both components of uncanniness investigated here (likeability and eeriness) can be affected by an interaction with a robot. The likeability of a robot is mainly affected by its attitude, and this effect is especially prominent for a machinelike robot. Merely repeating interactions is sufficient to reduce the degree of eeriness, irrespective of a robot’s embodiment. As a result, we urge other researchers to investigate the uncanny valley theory in studies that involve actual human–robot interactions in order to fully understand the changing nature of this phenomenon.