73G-12 |
Comparison of two instrumental tests for predicting texture profiles of frankfurters |
A. LOUP, W. K. Chung, and J. -. F. Meullenet. Food Science, University of Arkansas, 2650 N Young Avenue, Fayetteville, AR 72704 Texture is a very important parameter for the evaluation of food quality. Instrumental methods correlating highly with the sensory perception of food texture are needed for routine quality assessment. Numerous instrumental methods have been evaluated for predicting the texture characteristics of frankfurters. However, the use of multiple instrumental parameters in combination with multivariate analysis has yet to be reported. The objective of this study was to seek correlation between texture sensory attributes assessed by a trained descriptive panel and instrumental measurements using a texture analyzer on various types of commercial frankfurters. Twenty three commercial brands of frankfurters were assessed by a 9-member descriptive panel. In addition, instrumental measurements were performed using a Texture Analyzer (TA-XT2i, Texture Technologies, Scarsdale, NY) in combination with penetration and double compression tests. Instrumental parameters from the two tests were analyzed by Partial Least Squares Regression (PLSR) to predict sensory texture attributes. Predictive models with high correlation coefficients were reported for all sensory attributes. For example, springiness (Rcal=0.90), hardness (Rcal=0.90), cohesiveness (Rcal=0.87), cohesiveness of mass (Rcal=0.98), and chewiness (Rcal=0.92) were well predicted from a double compression test with 90% strain. The results from the penetration test were acceptable, but correlation tended to be lower than for the compression test. As a result, a double compression test with 90% deformation was found to be an appropriate method for predicting sensory texture attributes of frankfurters.
Session 73G, Sensory Evaluation
|