Comparison of consumer-based methodologies for sensory characterization: Case study with four sample sets of powdered drinks

•CATA questions, projective mapping (PM) and PSP with consumers were compared.•Sample configurations did not largely differ among the three methodologies.•Results were similar to those obtained using descriptive analysis.•The methodologies differed in the relative importance attached to sensory characteristics.•CATA showed the lowest ability to discriminate among samples.

Interest in consumer-based methodologies for sensory characterization has largely increased in the last decade. However, research on their applicability and reliability is still necessary. In this context, the aim of the present work was to compare three consumer-based methodologies and descriptive analysis for sensory characterization, considering as case study four sets of orange-flavoured powdered drinks, differing in the number of samples and the extent to which they represented the whole product category. A trained assessor panel evaluated sample sets using descriptive analysis, while groups of 100 consumers evaluated samples using one of three methodologies: CATA questions, projective mapping (PM) and polarized sensory positioning (PSP). Across the four sample sets, the three-consumer based methodologies provided similar information regarding the main similarities and differences among samples, which did not largely differ from that obtained using descriptive analysis. The main difference among the methodologies was related to the relative importance given to the sensory characteristics when evaluating differences among samples, which led to differences in the dimensionality of the sensory spaces. Sample configurations from PM and PSP required more dimensions than those from descriptive analysis and CATA questions to fully explain differences among samples. None of the consumer-based methodologies outperformed descriptive analysis in terms of ability to discriminate samples, being CATA questions the methodology that showed the lowest discrimination. Implications for the selection of consumer-based methodologies for specific applications are discussed.