A Robot with Personality: How to (responsibly) design a digital persona?
Social Robots and other artificial agents are often designed to have a digital persona. A digital persona can be viewed as a composite of elements of identity (such as demographics and background facts), behaviour, and interaction style.
These are deliberate design choices, often influenced by common perceptions, such as female voices being more `pleasant' when choosing a synthesised voice and gender.However, these design choices bear the risk of reinforcing social stereotypes, according to the UNESCO (West et al., 2019).
In this project, we will we will be looking to reflect personality not only by different synthesised voices, but also personality expressed in language and multimodal behaviour, conversational content and interaction style, building upon previous work on using neural generative models for personality-based linguistic style generation and generating responses which factually consistent with a persona profile.
We will also look to consider what role such persona profiles play in the context of human-robot interaction, whilst working closely with social psychologists to anticipate the social impacts these artificial personas might have. In particular, we will investigate:
(i) the (objective) impact/importance of robot persona (including gender) in HRI
(ii) if/how robots can/should effectively respond to abuse/inappropriate behaviour
Eventually, we aim to build a data-driven mapping between conversational behaviour (e.g. voice, linguistic style and content) and perceived personality traits, such as gender, age, trustworthiness etc.
Anastasia Kuzminykh, Jenny Sun, Nivetha Govindaraju, Jeff Avery, and Edward Lank. 2020. Genie in the bottle: Anthropomorphized perceptions ofconversational agents. In Proceedings of the 2020 CHI Conference on Human Factors in ComputingSystems, pages 1–13.
Clifford Nass and Scott Brave. 2007. Wired for Speech:How Voice Activates and Advances the Human-Computer Relationship. The MIT Press.
Shereen Oraby, Lena Reed, Shubhangi Tandon, Sharath T.S., Stephanie Lukin, and Marilyn Walker. 2018. Controlling personality-based stylistic variation with neural natural language generators. In Proceedings of the 19th Annual SIGdial Meeting on Dis-course and Dialogue.
Selina Jeanne Sutton. 2020. Gender ambiguous, notgenderless: Designing gender in voice user inter-faces (vuis) with sensitivity. InProceedings of the2nd Conference on Conversational User Interfaces,pages 1–8.
Mark West, Rebecca Kraut, and Han Ei Chew. 2019.I’d blush if i could: closing gender divides in dig-ital skills through education.Technical ReportGEN/2019/EQUALS/1 REV, UNESCO