Q&A with Dr Leon Watts

Ahead of our tech talks at Moving Brands and Wayne McGregor Studios, Dr Leon Watts of the University of Bath shares a few thoughts on ‘Macheba’ as one of the makers of the little robot and third performer in Solo for Two.

How did you guide the process and choices made in the physical design and realisation of robot ‘Macheba’?
As a scientist, it is important for me to combine ideas that have some demonstrated validity with new, open questions. I knew the physical design had to emphasis the behaviour and character of Macheba as an autonomous machine. As soon as people start to think about robots they automatically create mental images drawn from literature, film and television; typically images that are about symbols and societal change rather than functions and individual understanding and interaction. So I brought ideas for the design that stem from my research on basic emotional cues from behaviour, especially linked to interactive attention, to ask new questions about public interpretations of human and machine identity. Together with Michele Panegrossi (creative technologist) and Jean Abreu, we worked on this over a long period, learning from early sharings and experiences. We actually created three proto-types before getting to the construction of Macheba as you will see her now in the show.

What were the essential choices made?
In essence, there were three key decisions in the creation of the robot in Solo for Two.

Firstly, Macheba should not have a human appearance, as giving a robot a human form can be misleading. All my research has been on non-humanoid robots. So the first essential choice was to focus on Macheba’s embodied behaviour, everything stems from this: lights, sounds, motion are all connected to machine function, except for the main rotating light.

Secondly, the robot’s shell should be made of a transparent material to expose the circuitry. There is an ongoing debate in Artificial Intelligence and Human-Robot Interaction about ‘transparency’, which is to do with the potential for a person to understand what an intelligent computer is doing and why it is doing it. I wanted to apply this idea literally on Macheba.

Thirdly, the main light of the robot was very important to us. This light is part of the big question: how are robots designed to express what they ‘think’? Attention is vital to human interactions. By turning our heads, reaching out, glancing towards or looking away from something, people are acutely attuned to one another and I think robots should demonstrate attention in a similar way. Most people will interpret Macheba's light as an eye. But it is not her eye: it does not take information in from the world, it is the opposite. It puts information into the world for the benefit of the people who encounter Macheba. That information is about Macheba's interest or concern and, as with animal and human attention, varies by degree. The variable intensity of the light and the variable speed of the motors both correspond to degree of machine energy invested in any object or person Macheba encounters.

According to you, how does Macheba fit in the choreography? What is ‘her’ role?

This is a more difficult question for me to answer because the choreography is so much bigger than human-robot interaction or indeed AI transparency and beyond my field of research. The choreography is, in my opinion, a synthesis of creative genius, personal experience and theatrical expertise all speaking with one voice through dance. Macheba fulfils several roles: muse, echo, mirror, oracle, child. All these roles are intended to support the lead dancer, not to replace or upstage. Whether or not that intention is fulfilled really depends on the way the audience interprets the interactions between dancer and machine.

What do you consider innovative in the approach of using robotics in this new dance work Solo for Two?

The power of Macheba in this work owes to the thoughtful and coherent integration of machine behaviour with human identity and aspiration: to be recognised and respected in the eyes of others; to understand and to grow as a human being through embodied interactions and experiences with others; to see robots as reflections of our selves. I think this is a very different way of thinking about robots or indeed other technology in dance. Robots are positioned not as proxy humans or as opportunities for sophisticated technical befuddlement or gewgaws, but as simple autonomous entities that could encode and support human values through interactions.

Catch Leon Watts at one of the 'Embodying Technology' talks, two free events we've organised to discuss the use of technology in dance and the wider performance arts.

Join the discussion on Wednesday 1 May 

Find out more about Leon Watts' research