In daily life, human beings see the world in the given visual environment. The interaction between human beings and the visual environment is the fundamental principle for visual communication and expression of forms. The interaction between light and human recognition and perception of colors in visual environment is of particular interest for many artists. The impressionist Claude Monet often painted the same motif under several natural conditions and recorded the dynamic effect of light on the renowned 33 canvases of La Cathedrale de Rouen.

But few people know that the impressionists' exploration of natural light by painting and outdoors was partly made possible by the advance in technology at that time of pressing zinc to manufacture tubes of paint. Today, the light source such as RGB (red, green, blue) monitors or full-color LED devices controlled by the additive-mixture color method is becoming common in daily life. But the RGB light emission is hard to adjust to human color perception in a straightforward way.

To solve this problem, I have developed an original algorithm based on color theory and colorengineering technique to control an RGB output device along with human psychological measurements of lightness, hue and chroma. And I build up a technique for people to interactively control the colors of the light-source by their position in space and a new color-composition technique to make object color pattern reflect the dynamic light-source effectively. The combined technique is my answer to the old question of "What are the colors we perceive?" or "What is the dynamics of color?" And it provides a method to interact with visual environment using light and color.

In the installation, I applied the method to manufacture an interface and to prepare a space for participants to play with the dynamic interaction of light and color, which they rarely experience in the daily life. The experience will provide a chance to reflect the way we perceive colors and to gain a new perspective on colors in the new media.

Supported by the International Media Research Foundation and the Japan Science and Technology Agency