Electronic Nose Profiling of Pet Foods for Brand and Flavor Discrimination with Preference Analysis

Electronic nose (EN) technology, based on electrochemical sensor arrays, has emerged as a rapid and objective tool for aroma characterization in food systems. This study investigated factors influencing cat food preference using physicochemical analysis, texture measurement, preference testing, and EN technology. Nine (9) commercial cat foods from three brands (A: premium; B and C: medium price) with different flavors were evaluated. Proximate analysis revealed no significant differences (p > 0.05) in most nutrients, except crude fiber (CF) (p < 0.05), which was higher in lower-priced cat foods. Shear force showed a positive correlation with consumption (r = 0.72), while CF was negatively correlated (r = −0.52). Preference tests indicated that Brand A was most preferred, followed by Brand C, while Brand B was least preferred. Principal component analysis (PCA) identified variability in individual cat preferences, and one outlier cat was excluded. Discriminant analysis of EN data showed clear separation by brand rather than flavor, suggesting brand-related aroma profiles can be monitored rapidly through the digital odor fingerprint. Cat food acceptance is mainly driven by texture, CF content, and aroma rather than macronutrient composition. The results of the EN evaluation of cat food samples revealed the same group similarities and differences as the 8-month preference test performed with cats. Similar to this study, EN classification models can be developed using food preference data and the digital aroma fingerprints of foods. After validations, the EN models can be used as an effective tool to monitor the quality and assess new diets according to the known preferred odor fingerprint.