Two Late-Breaking Reports accepted for poster presentation at HRI2014 in Bielefeld, Germany.

Expectation Setting and Personality Attribution in HRI

People tend to treat robots as social actors and assign personality attributes to them. This study investigates the influence of expectation setting on the users’ attribution of personality traits to the robot and their impressions of that robot. Results show that personality attribution is depending on people’s prior expectations of the interaction. Moreover, people evaluate a robot better when they assigned it with a complementary personality and when they had high prior expectations of that robot.

 

Users’ Preferences of Robots for Domestic Use

This study identifies the design preferences of robots for domestic use. In an online survey, participants evaluated 16 robot pictures on several design and interaction aspects. The results show a strong overall preference for anthropomorphic robots, which were on almost all aspects more positively rated than other types of robots (e.g. zoomorphic, caricatured or functional robots), but were also associated with more risk. Moreover, negative feelings towards robot (e.g. negative attitudes and anxiety towards robots) negatively affected the design evaluations. Providing future users with positive information about domestic robots could improve their feelings towards robots, which, in turn, raises their ratings of robotic designs.

People say they are positive about robots, but their implicit associations tell us otherwise.

Abstract accepted for presentation at Etmaal van de Communicatiewetenschap [24 hours of communication science] 2014.

People’s Implicit and Explicit Associations with and Attitudes towards Robots

Robotics researchers predict that robots will enter our homes and become part of our everyday lives as assistants, servants, co-workers and even companions. But are people willing to accept these ‘creatures’ into their private spaces? The acceptability of robotic devices in home settings does not depend only on the practical benefits they can provide, but on complex relationships between cognitive, affective and emotional components of people’s associations with and attitudes towards robots. It is important to study what people assume when they hear the word ‘robot’, before discussing any differences in emotions and attitudes towards robots. However, this important area of research is rather unexplored. Moreover, previous research on people’s associations with and attitudes towards robots have only relied on questionnaires. However, it is not sufficient enough to rely solitarily on these explicit opinions. First because people are not always aware of the attitudes affecting their behavior. Secondly, people can try to conceal their attitudes and if this results in a desire to conform, it can lead to a self-presentational bias.

Studying both implicit and explicit associations with robots combined with attitudes towards robots makes it possible to provide the field of robotics useful insights into how people perceive robots which will enable designers to incorporate these insights into their creations. Research on the relation between associations and attitudes indicate that the types of robots people experience relate to their associations with robots, and these associations influence their attitudes towards robots.

To clarify this relation between attitudes towards and associations with robots, this study investigated people’s implicit and explicit associations with robots and their relation to people’s attitudes towards robots. A total of 207 respondents completed an Implicit Associations Test for the implicit associations and open-ended questions for explicit associations together with the Negative Attitudes towards Robots scale.

Although people explicitly told us they have positive associations of robots, their implicit associations indicate the opposite. Additionally, there was no relation found between the two measures. The conflicting associations and the non-significant correlation could indicate that people implicitly have different opinions about robots than they explicitly reveal to us. Moreover, the non-significant correlation between explicit associations of domestic robots and people’s attitudes towards robots as well as the negative correlation between the people’s implicit associations of robots and their attitude towards interaction with robots further confirms this line of though. Both people’s implicit and explicit associations of robots are related to anxiety towards robots. However, more negative implicit associations are connected to higher levels of anxiety, and more positive explicit associations are connected to lower levels of anxiety. Again, there is a friction between implicit and explicit association of robots. These findings also underlines the conclusion made earlier that people actually have negative associations of robots.

This study contributes valuable insights for domains in communication, psychology and robotic research by examining people’s opinions towards domestic robots. The contradiction between the implicit and explicit associations of domestic robots could indicate two different scenario’s. It could be that people in fact have a negative view on domestic robots but seem to think they should be positive about the topic, or it could be people that people are motivated to their opinion about domestic robots which results in their positive explicit associations to be their real attitude towards robots.