The Irresistible Animacy of Lively Artefacts
This thesis explores the perception of ‘liveliness’, or ‘animacy’, in robotically driven artefacts. This perception is irresistible, pervasive, aesthetically potent and poorly understood. I argue that the Cartesian rationalist tendencies of robotic and artificial intelligence research cultures, and associated cognitivist theories of mind, fail to acknowledge the perceptual and instinctual emotional affects that lively artefacts elicit. The thesis examines how we see artefacts with particular qualities of motion to be alive, and asks what notions of cognition can explain these perceptions. ‘Irresistible Animacy’ is our human tendency to be drawn to the primitive and strangely thrilling nature of experiencing lively artefacts. I have two research methodologies; one is interdisciplinary scholarship and the other is my artistic practice of building lively artefacts. I have developed an approach that draws on first-order cybernetics’ central animating principle of feedback-control, and second-order cybernetics’ concerns with cognition. The foundations of this approach are based upon practices of machine making to embody and perform animate behaviour, both as scientific and artistic pursuits. These have inspired embodied, embedded, enactive, and extended notions of cognition. I have developed an understanding using a theoretical framework, drawing upon literature on visual perception, behavioural and social psychology, puppetry, animation, cybernetics, robotics, interaction and aesthetics. I take as a starting point, the understanding that the visual cortex of the vertebrate eye includes active feature-detection for animate agents in our environment, and actively constructs the causal and social structure of this environment. I suggest perceptual ambiguity is at the centre of all animated art forms. Ambiguity encourages natural curiosity and interactive participation. It also elicits complex visceral qualities of presence and the uncanny. In the making of my own Lively Artefacts, I demonstrate a series of different approaches including the use of abstraction, artificial life algorithms, and reactive techniques.