How did the research in New York differ from the research in Munich?
In Munich, we tried our Mykie prototype in a shared flat with multiple users, with a person who travels a lot (an air steward), an older lady living in a tiny apartment and a young woman who has a baking blog.
In New York, on the other hand, we focused mainly on experts and designers, and those who spent a lot of time on their mobile phones and with other technology. We also took into account the needs and preferences of people living in different areas. For example, we interviewed people in both Brooklyn and Manhattan, we noted that in Brooklyn the shops are mainly small, specialist stores and grocery delivery services such as Hello Fresh are common, which gave us some ideas about what consumers wanted.
How emotional and ‘human’ did you want Mykie to be, what did you decide?
Apple’s Siri and Amazon Echo’s Alexa are like people in the way that they interact with the user, but they do not have a body or facial features. We wanted Mykie to go further and display his emotions. So, throughout the research and development processes we had to consider things like how the head would turn and how big it should be? And whether he should have a face? We did find, though, that eyes were more than enough to show basic emotions, and were, in fact, also easier to animate.
What did you learn about the experiences of users and their desired relationship with robots?
Our most important finding was that the testers wanted complete control over Mykie. In other words, they didn’t want him to be able to do things by himself, but to suggest things, such as ‘should I start the dishwasher’ so that the user could make their own decision. They also said that it would be creepy if Mykie was too human-like.
So, we played around with a number of different interaction modalities. Throughout our research in the USA, we tried out a range of different voices by fitting our plastic model with a Bluetooth loudspeaker, considered the tonality of his voice and played around with the idea of him giving orders or making jokes.
The methodology we used was called “Wizard of Oz” prototyping. As I said before, we used a rudimentary foam model of the completed Mykie, which was quite simple and asked users to apply some magic and role-play with the model, using their imaginations to see how they might interact with him.
What is Mykie’s current status?
The robotics team is currently working on the UI & ID design, twisting and turning function and interaction of Mykie, its voice and touch screen and has built a small micro-series, which is shown all over the world in selected locations and exhibitions. We also presented Mykie at this year’s IFA in Berlin, and took him to the CES in Las Vegas! The future is looking very exciting for Mykie and the BSH team behind him, we’re looking forward to seeing how he interacts with and affects our consumers’ lives.