Effect of concentration range on the accuracy of measuring sweetness potencies of sweeteners

Abstract

The main objective of the present study was to determine the effect of the concentration range of sweetener on the accuracy of sweetness potency, defined as the ratio of target sweetener and sucrose concentrations at equivalent sweetness intensity level (Wee, Tan, & Forde, 2018), measurement. Attempts were made to measure the sweetness potencies of two bulk sweeteners (powder allulose and liquefied allulose) and three intense sweeteners (sucralose, rebaudioside A, rebaudioside D) using a sucrose-sweetener combined (SSC) method. Three sets of sample combination were evaluated in three independent descriptive analysis testing sessions. One set of sweetener samples had a sucrose-equivalent concentration (SEC) range of 2-15%, wide range condition. The other two sample sets had narrower SEC ranges, of 1-9% for the narrow-range, low-concentration condition, and 8-16% for the narrow-range, high-concentration condition. All samples were evaluated by 10 trained panelists. Reference standards for sweetness intensities were provided during the testing sessions. Overall, the predicted sweetness potency measured over the wide concentration range was less accurate than that measured over the narrower range. Additionally, the results showed that the sweetness potencies were little affected by the concentration for bulk sweeteners, while the sweetness potencies changed dramatically for intense sweeteners. The SSC method was previously proposed for reducing the contextual bias caused by a range-frequency effect during sweetness intensity measurement. However, the present study has revealed that this method is not completely free from psychological biases and that the accuracy of measuring sweetness potency can be affected by the concentration range over which the panelist is evaluating. Thus, the concentration range of sample sets that are evaluated together should be determined carefully to ensure accurate measurement of sweetness.