People vary greatly in their sensitivities to bitter compounds. However, these compounds are all lumped under the umbrella term “bitter.” To further characterize bitter subattributes which may define these compounds, a protocol was developed by which thirteen bitter stimuli could be assessed using sorting, a semantic-free perceptual mapping technique. The clusters made by fifteen PROP nontasters and sixteen PROP tasters were compared, and the combined nontaster and taster data was analyzed. Similarity data were assessed using multidimensional scaling (MDS) to produce perceptual maps for the nontaster, taster, and combined nontaster and taster data. Overall, groupings were largely inconsistent and inconclusive, which pointed to shortcomings in the protocol or perhaps the lack of bitter subattributes. Limitations are discussed, and suggestions are provided for the optimization and future direction of characterizing bitter subqualities.
