This work consists in the study of apple consumer preferences which have been studied only partially until now. Indeed, the studied cultivars were not representative of all the sensory diversity and the aroma complexity was not totally taken into account. The first part of this work aims to describe almost all the sensory diversity of apples and to select a set of products representative of this diversity. The goal of the second part is to precisely study the preferences of apple consumers using the representative set of products. Preferences of all the consumer segments are explained by the sweet taste and the overall aroma intensity. Nevertheless, the segments differ on liking regarding texture and specific aroma notes. The third part deals with a methodological work on the impact of the chosen products in a preference study on the consumer segmentation and on the preference sensory keydrivers. This methodological question is crucial and has not been studied specifically to our knowledge. The resulting partitions of the three tested sets of products are not totally matching. Different preference patterns are underlined when different set of products are taken into account except a common preference direction explained by the overall aroma intensity and the sweet taste. A better understanding of perception phenomena and their interactions are key factors to improve consumer preferences. The study of some sensory interactions is the subject of the last part of this work. Colour is an important criterion during food consumption and particularly apple consumption. The influence of color on consumer preferences on the one hand and on in-mouth perceptions of trained panel on the other hand, is studied. The apple colour alters sensory responses in certain cases highlighting the existence of cognitive interactions between visual and in-mouth perceptions. Another side of the study of interactions between sensory systems is also considered. The influence of odour perception on aroma, taste and texture perceptions is investigated. The implied odour is particular in the sense as it is representative of an aroma perceived in the mouth. This study implies the use of an innovative system called "Mouth to Nose Merging System" which combines the simultaneous use of a mastication simulator and a trained panel. The system has the advantage to allow the use of complex real products to study sensory interactions. It allowed identifying cognitive interactions between odour and texture perceptions.
