In my PhD research, I aim to broaden the knowledge about the uncanny valley effect. My project is mostly focused of understanding how uncanny feelings towards robots change over time and how this process can be influenced by the robot's interactions. As part of my work, I'm considering the influence of age, the robot's embodiment and different dimensions that shape the feeling of the uncanny.
The literature on the uncanny valley effect is very focused on understanding what exactly causes an agent or a robot to be perceived as uncanny. However, little is known about the durability of the uncanny valley effect. In my research, I want to find out if initial uncanny feelings towards a robot can be overcome over the course of repeated interactions and what in the robot's behavior can foster or hinder overcoming the uncanny feelings.
I am interested in understanding how robots can express a coherent and consistent personality. A crucial part in developing personalities is being able to communicate the robot's current mood and create a natural progress of the mood state over time. Eventually I want to understand how the robot's personalty influences peoples perception of the robot and how they interact with it.
Nellie is a continuation of Eve, the artificial agent we developed to play a rapid image matching game. Nellie engages in a cooperative game with a human in which they need to identify countries on the world map. This game is significantly more complex compared to the original image matching game and thus requires advanced natural language understanding and reference resolving skills.
In a collaboration with researchers at the Institut des Systèmes Intelligents et de Robotique, Université Pierre et Marie Curie, we investigate the link between mimicry and the uncanny. The aim of the project is to research if mimicry can be used as a potential future assessment criteria for uncanny feelings towards a robot. We use both the mixed-embodied Furhat robot and a 2D representation of the character and investigate differences between the embodiment and its influence on the task success.
We aim to broaden the knowledge of the uncanny valley effect by analyzing the interplay between perceptual mismatch in the category of gender. We use the Furhat mixed-embodied robot platform with a female and male appearance and voice. By altering different cues and thereby creating a perceptual mismatch, we aim to understand what triggers the feeling of uncanny in different age groups.
By analyzing the interaction strategies of a street performer with his dragon puppet, we develop design guidelines for what creates the feeling of uncanny and how to overcome this feeling using interaction techniques. In a second step, we will develop a semi-autonomous robot that replaces the dragon in the same street performance to be able to further investigate what in the robot's and moderator's interactions help children to overcome initial uncanny feelings towards the robot.
As part of the NSF founded project “Incremental Speech Processing for Rapid Dialogue“ we developed a dialogue agent that can engage in a two-player image matching game. In the game, a human acts as a director and describes one out of eight images they see on the screen. The automated matcher dialogue system processes the speech of the director and selects the most likely image candidate as soon as possible. The dialogue strategy of the agent is optimized to maximize the points the team scores per second.
Our first experiments have shown that the possibilities of altering Furhat's appearance are limited by the physical mask. The mask itself is perceived as male and dominant, which creates a perceptual mismatch when applying a female texture to it. In a collaboration with Furhat robotics, we are working on designing a new mask in a female shape that can be used in further experiments to create an unambiguous gender perception.