%X The ‘Hue-Heat Hypothesis’ states that light with wavelengths predominantly at the red end of the spectrum (or of a low colour temperature) are felt as warmer, whilst light with wavelengths mainly in the blue end (or of a high colour temperature) are felt as cooler. If confirmed, the Hue-Heat-Hypothesis could be a powerful tool for energy savings:  Temperatures could be lowered under a reddish light in the heating season. Conversely, less air-conditioning might be needed during the cooling season if higher temperatures were accepted under a bluish light. Even a transitory effect would be beneficial in managing power demand by allowing building temperatures to drift over a wider range of temperatures before heating or cooling was required. 
We used an experimental design to study the Hue-Heat-Hypothesis. Testing took place in a climate chamber, in which ambient temperature, relative humidity, and air speed can be controlled.  We installed a LED-lighting system in the chamber covering the range of correlated colour temperatures from 2700K, a warm, reddish light, to 6500K which appears bluish-cold. Participants (age range 18 to 35 years) were exposed to combinations of colour temperature and ambient temperature and completed standard thermal comfort surveys at specific time-points. Prior to testing, participants filled in a survey that asked about other factors potentially impacting on thermal comfort. 
In Study 1, temperature in the climate chamber was cooled continuously from 24°C to 20°C over a 60-minute period. Comfort ratings were obtained every 10 minutes. Participants (N = 32) were either exposed to a colour temperature of 2700K or 6500K (between-subjects design). Thermal comfort was higher under the warm colour temperature (2700K) than under the cold one (6500K). This difference was particularly pronounced for temperatures around 21 and 22°C. For the same subjective thermal comfort rating, ambient temperatures differed by around half a degree under the different lighting conditions. The magnitude of the effect varied with temperature (greatest effects observed around 22-23˚C) and by comfort question asked. 
In Study 2, a within-subject design was used. Subjects were exposed to three different lights (2700K, 4440K,6500 K) with temperature decreasing from 23°C to 19°C. Preliminary results were similar to those of Study 1 with higher comfort under the warm-appearing light. Comfort ratings under the medium colour temperature of 4400K were positioned between the higher ratings obtained for 2700K and lower ratings for 6500K, again, only in a limited corridor of ambient temperature.
In Study 3, light was changed gradually from a cold to a warm colour temperature whilst ambient temperature decreased from 23°C to 19°C. Testing is on-going, but initial results indicate that the changes in self-reported comfort under decreasing temperatures are slower in the condition of changing light dynamically than under a stable 
Our studies support the Hue-Heat-Hypothesis. Varying the ambient light has an effect on thermal comfort and hence may be a suitable tool for energy savings and reducing of peak power demand.
%C Oxford, UK
%D 2014
%A Gesche M Huebner
%A David Shipworth
%A Stephanie Gauthier
%A Wing-San Chan
%A Christoph Witzel
%T Saving energy through changing light: The impact of illumination on thermal comfort
%L discovery10159119
%K thermal comfort, light, colour temperature