Best lighting for visual appreciation of artistic paintings—experiments with real paintings and real illumination

Abstract

In a previous study based on monitor simulations of artistic paintings, it was found that the average correlated color temperature (CCT) of daylight preferred by a large set of observers to illuminate paintings was around 5100 K. The goal of the present study was to test if this result holds in real viewing conditions, i.e., with real paintings and real light sources. The same 11 paintings were tested in real conditions illuminated by a spectrally tunable light source and with accurate monitor simulations. To ensure uniform illumination across the paintings, only a central part of the paintings was visible to the observers. It was found that the average CCT preferred for real and monitor viewing conditions were very similar, 5500 and 5700 K, respectively. The somewhat larger CCT obtained with monitor viewing in relation to the former study was only observed in some paintings and was attributed to the smaller viewing area. These results confirm that CCT for best appreciation of paintings is higher than normally used in museums, and the viewing conditions, real or simulated, have only a minor effect.

Light dominates colour preference when correlated colour temperature differs

“…Such a design may unpremeditatedly help to reveal the influence of other contextual factors on colour preference, since it relatively weakens the impact of light when compared to the conditions where CCT differs. However, unlike colour fidelity, colour preference should not be restricted by a reference light source, since in many cases people actually want to choose a favorite light irrespective of the CCT 4,26,[33][34][35][36][37] . Secondly, the experimental object is also an important concern.…”

“…The quantity and type of experimental objects varied considerably between different reported studies. Some researchers invited the participants to perceive a single type of object under different lights, such as fruit and vegetables 2,9 , skin tones 38 , artworks 25,33 , printed images 36 , cosmetic products 10 , or consumer goods 4,11 . Meanwhile, in other contributions, groups of mixed objects were used 3,28,34 .…”

Interpreting Art with Light: Museum Lighting between Objectivity and Hyperrealism

“…The analyses partly reference the problem in that they analyze a single work of art or dissimilar artworks or sections of a painting or reference a specific painting technique or certain period, but a generalization with several artworks and differing painting techniques is usually not available in practice. The rise of light emitting diode (LED) technology has also enabled the spectrum to be modified and color temperature to be adapted to the pictorial content and matched to daylight (Csuti et al 2015;Nascimento and Masuda 2014).…”

Colour fidelity for picture gallery illumination, Part 1: Determining the optimum light-emitting diode spectrum

“…Several papers have dealt with the question of colour rendering and museum lighting, one of the most recent was that of Nascimento and Masuda, 14 who concluded from visual investigations that daylight with a CCT of about 5500 K was the most preferred illumination. This result comes near to the following postulate: For museum lighting colour fidelity lighting is of importance, as one certainly does not want to distort the perceived colours compared to how they were conceived by the painter, even if some observers would find the painting 'nicer' under a light source with preferred colour rendering.…”