Consumer acceptability versus trained sensory panel scores of powdered milk shelf-life defects.

The objective of the present work was to correlate consumer panel acceptability versus trained sensory panel scores for appearance and flavor defects likely to appear during storage of whole milk powder. Descriptors selected for the study were: acid, caramel, cooked, dark color, lipolysis, and oxidized. For each descriptor a set of nine samples with different intensities were measured for acceptability and likelihood to consume by a 50-member consumer panel and for sensory intensity by a trained panel. Linear correlations between sensory acceptability and trained sensory panel scores were used to determine the sensory failure cut-off point for each descriptor, except caramel and cooked, which were not critical from the consumer's point of view. Differences in acceptability were found between Argentine and Uruguayan consumers for oxidized samples, while for lipolysis flavor, Argentine and Costa Rican consumers behaved similarly. For the color descriptor, significant changes in acceptability measured on a hedonic scale did not mean that consumers refused to consume the product. In contrast, for flavor descriptors, as soon as a significant decrease in acceptability occurred approximately 30% of the consumers said they would not consume the product. The sensory failure cut-off points presented in this paper can be used as a guide in future studies on the shelf life of MP and can also be of value in establishing sensory specifications for quality control programs. The methodology of correlating consumer acceptability to sensory panel scores and, thus, defining sensory failure is an improvement over more arbitrary criteria presented in most shelf-life studies.