Colorful insights supporting the modeling of creative processes across language, music and emotion

Abstract

In an attempt to unite advances in color science, the creative arts, cognitive psychology and engineering to foster an improved understanding of creative processes, we elaborate on and discuss some recent findings across creative art experiments, cognitive psychology and perceptual color compression. Motivated by a fundamental phenomenon of efficiently managing functional complexity, which has been observed, for example, in design tool development as well as product development in microelectronics, we propose associating functional structures of language and music with perceptual sensitivity to hue, simultaneous color contrast, and color saturation across high dynamic range color processing while simultaneously considering elementary models used in quantum mechanics. A multi-dimensional functional hierarchy not only reveals similarities but also offers a platform towards open questions about creative thinking processes which can be more easily elaborated in a distributive differential context provided by a dynamic spatio-temporal (space-time) model. A suitable test scenario emerges from forward-looking creative experiments utilizing the arts in combination with a highly generic perceptual color compression model. A well engineered verification toolset based on nonlinear control theory appears useful in addressing fundamental challenges in highly complex neuronal tasks across language, music and emotion. We also focus on interfacing between terminologies of cognitive, mathematical and engineering domains together with their underlying functionalities. We present preliminary thoughts and discuss our unconventionally structured approach in multi-modal and multi-sensory contexts via comparison of similarities as well as differences between creativity tasks and perceptual color compression. Ongoing work shall focus specifically on elaborating effective strategies that combine the proposed functional models with promising Deep Learning architectures.