The Aroma of Canned Beef: Models for Correlation of Instrumental and Sensory Data

Abstract

Stevens' and Fechner's relations have been applied in a generalized form: they have been expressed as functions in several variables. These functions have been formulated in analogy with models used in other psychophysical contexts and are considered to handle simple interactions between chemical stimuli compounds. It is suggested that the different outcomes can be classified into three categories: ad hoc relations, predictive relations and causative relations. The meaning and interpretability of these are discussed. The models have been tested on data from samples of canned beef: different formulations and heating times. A large number of high correlations were obtained for the different odor qualities and the presence of possible causative relations among these are discussed.

Gas chromatographic and sensory analysis of volatiles from cling peaches

“…This may indicate a masking effect of the background odors by the floral and fruity odors in peaches. Persson et al (1973) defined "background" components as being important contributors to the aroma of a product, but they do not change much in concentration among similar products with differences in odor qualities and concentrations. Several of the 7-lactones were present in such low concentrations in the canned peaches that they were not included in the regression analysis.…”

The Aroma of Canned Beef: Processing and Formulation Aspects

“…HTST-sterilization It is evident (Table 2) that a higher temperature and shorter heating time at a constant Fe-value decreased the concentrations of those compounds that could be related to the off-flavors in canned beef (aldehydes and sulfur compounds; see Persson and von Sydow, 1973;Persson et al, 1973b). The differences in concentrations for the sulfur compounds were especially marked between 1 21°C and 13 1'C.…”

Aroma Characteristics of Corn Germ Protein Flours

“…CORRELATION between instrumental and sensory data has long been used to evaluate quality of food products. However, most studies were focused on predictive relations, and little was done to find causative relations (Persson et al, 1973). Predictive relations may be sufficient for many practical applications, such as quality control.…”