Petfood Industry - July 2014 - 16
SCIENTIFIC RELEASE // INTRODUCTION This database recorded 4 categories of variables with several modalities: > CAT OWNERS LIKE TO CATER TO THEIR CAT Individual cat variables: name, age, weight etc. PREFERENCES... Meal time is a privileged moment during which pet owners always wish to please their animals with tasty products they will enjoy. Cat owners also aspire to offer products best suited to their animal needs and expectations. Accordingly, pet food manufacturers develop a wide range segmentation meeting different age, lifestyle and physiological needs. However, dry cat food offers limited recipe choices and meal variety is restricted to the main flavors: poultry, beef, fish, ... Cat parents might feel they do not have enough choice to buy what they consider to be their animal favorite food. > ...BUT HOW CAN CATS PREFERENCES BE CHARACTERIZED? Several studies already demonstrate that cats have individual preferences that are both innate, such as the dislike of bitter taste1, and acquired. Indeed, antenatal, perinatal and postnatal exposures significantly influence olfactory preferences in kittens from birth till weaning2. Feeding history, influenced by behavioral and social elements2, and resulting from a balance between neophilia3 and neophobia4 will make cats taste evolve over time5. To further understand cat food palatability drivers, DIANA PET FOOD experts in palatability designed a complete study. It included palatability data from both expert and in-home cat panels to define whether it was possible to identify cat populations showing similar food preferences. 1ST STEP : IDENTIFYING CAT INDIVIDUAL PREFERENCES IN EXPERT PANELS > DESIGN OF THE STUDY In order to define whether cat populations with similar preferences exist, a first survey was conducted. It analyzed a database containing the preference data of over 380 cats collected during 3 years at Panelis (DIANA PET FOOD Division expert center in palatability measurement). Diet variables: kibble type, fat type, Palatability Enhancer (PE) form, nature or species etc. adding up to a total of 40 diet modalities. Exposure variables: total number of exposures per cat for each modality. Preference variables: A binary preference variable based on individual food consumption and consumption ratio was defined assigning a score of 1 to the most eaten product and a score of 0 to the other product. For each modality, several classes of individual cat exposure were defined ([0-10], [11-20], [21-50], [51-100] etc.). To limit neophilia and neophobia effects5 it was decided to consider the preference of cats that have been exposed at least 10 times to a modality. The database collecting this information finally contained over 350.000 rows (Figure 1). IDENTIFICATION VARIABLE CAT NAME GROUP EXPOSURE VARIABLE ... CLASS OF EXPOSURE PREFERENCE VARIABLE ... AVERAGE CONSUMPTION DIFFERENCE VARIABLE 1 AVERAGE RATIO VARIABLE 1 ... Abricot GR 7 ... [21-50] ... 21.36 74.25 ... Abscinthe GR 7 ... [101-200] ... -5.14 40.15 ... Abysse GR 5 ... [11-20] ... 3.14 53.17 ... ... ... ... ... Figure 1: Extract of Diana Pet Food expert panel database used for identification of individual preferences. The cat individual consumption and preference data were statistically processed using two forms of multivariate data analysis: PCA (principal components analysis) and HCPC (hierarchical clustering on principal components). > RESULTS This extensive data processing defined cat populations showing similar food preferences for several studied modalities. Focus is made here on 2 specific diets modalities "A" and "B", corresponding to 2 different PE aromatic compositions. Data processing for these 2 PE modalities clearly highlighted 4 distinct clusters of cat populations with 4 levels of preferences for A or B (Figure 2, Figure 3):
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.