The Influence of Speed and Amplitude on Visibility and Perceived Subtlety of Dynamic Light

LED lighting can provide pleasant temporal dynamics, though the understanding of what constitutes pleasant has so far remained largely in the experiential domain of lighting designers. Only recently has research begun to uncover what affects preferences for dynamic lighting. Based on the insight that visible, yet subtle, dynamic lighting is desirable, an experiment was conducted which studied the influence of the speed and amplitude of periodic temporal color transitions on the thresholds for visibility and perceived subtlety. Using RGB LED lights to illuminate a wall in a living room setting, stimuli with periodic color transitions were presented to 40 observers, half of whom evaluated whether the dynamic transitions were visible, and half of whom evaluated whether the dynamics were subtle. The observers’ data were fit using psychometric surfaces over the independent variables speed and amplitude of transitions. The 2D contours at the middle of the height of these surfaces were taken as thresholds. Experimental confidence was estimated via parametric bootstrapping, showing significant effects of the location and direction in color space in which the transitions were made, further strengthening the need for a temporal difference based color space. Interestingly, a distinct difference between the threshold for visibility and the threshold for subtlety can be seen, allowing for the creation of transitions that are both visibly and yet subtly dynamic. The results indicated that observers are more sensitive to hue changes than to chroma changes, and observers are more sensitive to hue changes in the orange region than to those in the blue region. Additionally, questionnaire responses provided insight into interpretations of the word subtle, understood to be pleasant, gentle, and smooth, which are also terms used to describe desired characteristics of dynamic lighting. Introduction New lighting technologies, notably LED, enable inexpensive ways to create lighting effects that are colorful and temporally dynamic. Solid state lighting systems can enable full control of both intensity and color, whereas traditional lighting systems typically change only the intensity of the illumination produced (for most of them only at two levels – on and off). Furthermore, the speed at which these changes can be controlled in solid state lighting is very fast, enabling the creation of fully dynamic atmospheres. A Dutch study [1] showed that people who were familiar with the Philips Living Colors, a LED based decorative lamp, liked the dynamic demo mode of the device and considered the addition of controlled dynamic light effects a valuable feature. The appeal of dynamic lighting is not surprising as light in nature is typically dynamic: sunlight changes in color temperature and intensity over the day, clouds attenuate it periodically, and windblown leaves provide ever-changing patterns of shadow and light. The subtle dynamics of natural light are familiar and pleasant. In some applications, professional lighting designers tailor dynamic solutions for specific purposes or customers, for example a retail store or hotel lobby. In other applications, such as for home use, a lighting designer is unlikely to be involved, so a dynamic lighting solution must be pre-packaged. Thus, for the maker of consumer dynamic lighting solutions, much must be known about the perception of dynamic color effects and the preferences of the end user. However, there is little research in technical literature on the topic. Recent work by Hartog et al. [2, 3] studied preferences for the speed, direction, and amplitude of periodic color changes. The quantative findings show that in general small amplitudes of change as well as slow light effects were seen as more attractive, showing preference for subtle dynamics. In an interview afterwards, 72% of the participants indicated that they want dynamic lighting in their living room. One interesting finding in her work was that people overwhelmingly like the idea of dynamic lighting, but the dynamics in most of the presented stimuli were seen as too obtrusive or too fast. Inspired by this, the main question of the experiment presented in this work was: what makes a temporal light transition appear subtle? Experiment The specific goal of the experiment was to understand the influence of speed and amplitude of periodic temporal transitions on the thresholds for visibility and perceived subtlety. Of special interest was whether these thresholds are distinct, which would mean that it is possible to generate visibly dynamic transitions that are subtle. Recognizing that existing color spaces (i.e. CIELAB) are not perceptually uniform for temporal color differences, two additional factors, the location and the direction of the transition in color space, were added to the experiment. Laboratory Setup The experiment was conducted in a laboratory arranged in a living room setting, with a sofa, chair, coffee table, and television. The dynamic lighting effects were provided by a pair of LED luminaires, one on each side of the television. The luminaires were controlled via a network interface by a Java program. Each luminaire was a tall vertical cylinder with two sets of RGB LEDs mounted on the ends of a solid cylindrical PMMA light mixing chamber. The chamber had a patterned diffuse coating on one side to couple the light out toward the wall. The result of this was a bright, oblong, soft-edged patch of light on the wall. Test participants sat on the sofa, from which vantage point they could see the two light patches on the wall, separated by 40 degrees of visual angle and each at a distance of 3m. They were asked to Page 1 of 5