A combined sensory-instrumental tool for apple quality evaluation

• Sensory and instrumental data of 27 apple cultivars over 2 seasons were collected.
• A set of instrumental analyses to predict sensory properties of apples was proposed.
• An effective prediction was achieved for the main sensory attributes (Q2 ≥ 0.8).
• Instrumental analyses can support but cannot replace sensory evaluations.
• Multisensory interactions in the perception of flavour and texture were highlighted.

A combined approach for perceptible quality profiling of apples based on sensory and instrumental techniques was developed. This work studied the correlation between sensory and instrumental data, and defined proper models for predicting sensory properties through instrumental measurements. Descriptive sensory analysis performed by a trained panel was carried out during two consecutive years, on a total of 27 apple cultivars assessed after two months postharvest storage. The 11 attributes included in the sensory vocabulary discriminated among the different apple cultivars by describing their sensory properties. Simultaneous instrumental profiling including colorimeter, texture analyser (measuring mechanical and acoustic parameters) and basic chemical measurements, provided a description of the cultivars consistent with the sensory profiles. Regression analyses showed effective predictive models for all sensory attributes (Q2 ≥ 0.8), except for green flesh colour and astringency, that were less effective (Q2 = 0.5 for both). Interesting relationships were found between taste perception and flesh appearance, and the combination of chemical and colorimeter data led to the development of an effective prediction model for sweet taste. Thus, the innovative sensory-instrumental tool described here can be proposed for the reliable prediction of apple sensory properties.