A Machiavellian robot in the wild, exploiting the culture of passersby to gain more help

dc.contributor.authorSanoubari, Elaheh
dc.contributor.examiningcommitteeBunt, Andrea (Computer Science)en_US
dc.contributor.examiningcommitteeLoureiro-Rodríguez, Veronica (Linguistics)en_US
dc.contributor.supervisorYoung, James (Computer Science)en_US
dc.date.accessioned2019-01-09T17:43:46Z
dc.date.available2019-01-09T17:43:46Z
dc.date.issued2018-12en_US
dc.date.submitted2018-12-28T23:44:07Zen
dc.degree.disciplineComputer Scienceen_US
dc.degree.levelMaster of Science (M.Sc.)en_US
dc.description.abstractRobots are entering public spaces where they use social techniques to interact with people. Robots can nowadays be found in public spaces such as airports, shopping malls, museums, or hospitals, where they interact with the general public. As these social entities are sharing people’s personal spaces and influencing their perceptions and actions, we must consider how they interact with people. The impact of robot’s interaction on a person is mediated by many factors, including personal difference and interaction context (Young et al., 2011). For both of these factors, the cultural background of the person is a particularly important component. Culture is deeply intertwined with all aspects of our social behaviors and impacts how we perceive our day-to-day interactions. As such, social robots can use culturally-appropriate language to improve how they are perceived by human users (Wang et al.,2010). In this work, we investigated if a robot can use social techniques to adapt to people to get more help from. More specifically, we investigated if a robot can do so by exploiting knowledge of a person’s culture. We conducted an in-the-wild experiment to investigate whether a robot adapting to a passerby culture would affect how much help it can get from them. The results of this study indicate that there is a significant increase in duration of their help in cases where the robot adapts to match their culture than when it mismatches. The results of this experiment contribute to the design techniques of adaptive social ro-bots by showing that it is possible for an agent to influence users’ actions by adapting to them. However, as this adaptation can happen without a person’s explicit knowledge, it is ethically questionable. By providing this proof of concept, our experiment sheds light on the discussions of the ethical aspects of robots interacting with humans in social contexts. Furthermore, we present the study design used for this experiment as a template for in-the-wild studies with cold-calling robots. We propose that researchers can use this template as a starting point and modify it for conducting their own similar robot in-the-wild research.en_US
dc.description.noteFebruary 2019en_US
dc.identifier.citationElaheh Sanoubari, James E. Young. “Hi human, can we talk? An in-the-wild study tem-plate for robots approaching unsuspecting participants”, In proceedings of the Work-shop on the Social Robots in the Wild of the 13th ACM/IEEE International Conference on Human-Robot Interaction. 2018 (pp. 237-238).en_US
dc.identifier.citationElaheh Sanoubari, James E. Young. “Explicit, Neutral, or Implicit: a cross-cultural ex-ploration of communication-style preferences in human robot interaction”, In adjunct Proceedings of the 13th ACM/IEEE International Conference on Human-Robot Interac-tion. 2018 (pp. 237-238). (extended abstract)en_US
dc.identifier.citationElaheh Sanoubari, Stela H. Seo, Diljot S. Garcha, James E. Young, Verónica Loureiro-Rodríguez. “Good Robot Design or Machiavellian? An in-the-wild robot leveraging minimal knowledge of passersby’s culture”. In Proceedings of alt.HRI track, ACM/IEEE International Conference on Human-Robot Interaction. 2019.en_US
dc.identifier.urihttp://hdl.handle.net/1993/33679
dc.language.isoengen_US
dc.rightsopen accessen_US
dc.subjectHuman-robot interactionen_US
dc.titleA Machiavellian robot in the wild, exploiting the culture of passersby to gain more helpen_US
dc.typemaster thesisen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Ellie's_MSc_Thesis_V10.pdf
Size:
2.53 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
2.2 KB
Format:
Item-specific license agreed to upon submission
Description: