Flavor release from a mint-flavored chewing gum model system was measured by atmospheric pressure chemical ionization mass spectroscopy (APCI-MS) and sensory time-intensity (TI). A data analysis method for handling the individual curves from both methods is presented. The APCI-MS data are ratio-scaled using the signal from acetone in the breath of subjects. Next, APCI-MS and sensory TI curves are smoothed by low-pass filtering. Principal component analysis of the individual curves is used to display graphically the product differentiation by APCI-MS or TI signals. It is shown that differences in gum composition can be measured by both instrumental and sensory techniques, providing comparable information. The peppermint oil level (0.5-2% w/w) in the gum influenced both the retronasal concentration and the perceived peppermint flavor. The sweeteners' (sorbitol or xylitol) effect is less apparent. Sensory adaptation and sensitivity differences of human perception versus APCI-MS detection might explain the divergence between the two dynamic measurement methods.
Journal of Agricultural and Food Chemistry, 2004, Vol 52, Issue 26, p. 8119-8126