Research methods for spatial brightness
General requirements
Research methods for spatial brightness
This paper was a first attempt to investigate how variations in experimental design affect the conclusions drawn about lamp spectrum and spatial brightness. The topics included procedure, evaluation mode, visual task, and field size. The analyses are made by comparing results from past studies and these tended to be insufficient to draw robust conclusions. The conclusion stated “It is found that further work is needed to enable this comparison.” Further work has indeed been conducted and this is shown in other papers in these lists.
This report describes good practise for experimental research of spatial brightness. The general principals are, however, valid for a wider range of psychophysical research.
The main recommendations include:
to counterbalance all options (left-right positions in a side-by-side matching test)
to observe independent variables (e.g. light settings) in a randomised order
to include null condition trials to enable internal validation
to use multiple procedures to test the same set of stimuli
to describe an experiment in sufficient detail to enable others to repeat the work
to analyse the results using suitable statistical methods.
This paper exemplifies why research methods are important.
In 1941, Kruithof suggested a relationship between illuminance and CCT, specifically that high CCT is needed at high illuminance (and low CCT at low illuminance) if the visual environment is to be considered pleasing. This became known as the Kruithof curve. While Kruithof did not intend for it to be the case, the Kruithof curve was for many subsequent years taken as a design rule and assumed correct by many researchers.
Given that Kruithof did not describe the test method, nor the sample, nor the conditions evaluated, nor provide any results nor statistical analyses, the results should not be considered credible. Researchers who refer to Kruithof without recognising or raising these limitations are not conducting their work with sufficient rigour.
Fotios S. A revised Kruithof graph based on empirical data. Leukos 2017; 13(1); 3-17.
A literature review was conducted to identify studies investigating the effect of illuminance and CCT on pleasing (or preferred) lighting and brightness. Only those studies meeting the requirements of CIE212:2014 were retained.
The results of these studies demonstrated that:
Variation in CCT has negligible effect on ratings of brightness and pleasantness.
Low illuminances (less than approximately 300 lux) may be perceived as unpleasant; an illuminance of 500 lux is sufficient to provide a pleasant environment and further increase in illuminance above 500 lux is of little benefit.
Higher illuminances are perceived to be brighter and this effect appears to be stronger than for other relationships.
This suggests the one condition to avoid is low illuminance (these data refer to interior contexts, not exterior) and that variation in CCT (within the range of approximately 2500 to 6500K) does not affect pleasing conditions and can be chosen by other criteria. The Kruithof graph should therefore show instead a single curve, a straight line (y ≈ 300), with the region underneath suggesting likely unpleasant conditions and the region above suggesting likely acceptable conditions.
A discussion of research methods, whether planning new work or reviewing published work by others. This paper looks at research quality considerations, research design, internal and external validity, defining lighting conditions, statistical analysis and research ethics.
