Extending computational models of abstract motion with movement qualities

The affectively rich expressive capacity of movement and motion is well established in art, performance, animation and visualization but research in perception, cognitive and social psychology provides only limited insight into the visual features that underpin this richness, and artistic principles are not amenable to computational modeling. Recent research has shown the communicative potential of simple abstract motions, absent of figure, to convey affect [23] based on a limited algorithmic model manipulating basic motion dimensions such as shape, speed and direction. Evidence suggests that descriptive frameworks of human movement expression, such as Laban Movement Analysis (LMA), are effective analytical tools with established principles and models; yet the benefits and challenges of incorporating these concepts into larger frameworks of motion and animation has not been rigorously explored. We present a computational model and prototype implementation that incorporates LMA core concepts and principles with established motion algorithms such that users can represent and explore LMA concepts using abstract motions. The model is the outcome of an indepth qualitative study with Certified Movement Analysts (CMAs) exploring, creating and analyzing the potential of low-level animation features to communicate expressive qualities of movement. A more comprehensive design space includes both new parameters for manipulation and a synthesis of lower-level dimensions into the more semantic concepts of Laban principles. In this paper, we discuss the evolution of the model to incorporate these principles of human movement, next steps, and relate the potential applicability of this research to applications in art, visualization and cognition.