Quibbler

We conducted a qualitative user study to find out how a user would interpret the controls we've designed.

Study 1: Perception of controls

We gave our physical prototype to test participants and asked them how they would have performed a series of robotic operations. We asked them to perform the following tasks:

a. Make the robot move forward
b. Bring a single robot's hand from his side above his head

Roughly halfthe participants successfully completed the task and the other half didn't. It's important to note that our controllers did not have labels on them because we wanted to test the intuition of the form since our target users may not necessarily recognize labels. The people who were successful acknowledged their success to similar types of controllers such as video game controllers and the Robosapien controller.

Study 2: Learnability of controller

Next we tried a more interactive study where we allowed two people to talk amongst themselves to figure out the controls to perform more complext tasks mostly with the arms. Here one of us was "the robot" and we rehearsed what the controls were. We had the test subjects perform the following tasks:

a. Put right arm over "heart"
b. Make both hands wave simultaneously

The participants completed their tasks though not without moments of difficulty within a ten minute frame. We found that once the participant figured out the first task, the second task in generally was shorter because they began to be aware of the range of the arm's functionality.