Count balancing needed to avoid bias in side-by-side brightness matching tasks

Spectral power distribution (SPD) is one of the variables that can be manipulated in lighting design. This article examines twenty previous studies which have investigated the effect of SPD on brightness using the side-by-side matching technique. Three sources of experimental bias are identified and are shown to be present in the majority of these studies, and hence must be considered to provide an unreliable estimate of the magnitude of the SPD effect on brightness. The first bias is associated with test participants' use of dimming control and two forms are apparent – a response contraction bias when the test is repeated at multiple reference illuminances and a conservative bias when the test is carried out at a single reference illuminance. The second bias is associated with the primary direction in which dimming must take place and the third bias is associated with the relative position of the stimuli within the apparatus.