The multisensory perception of sweetness in apple

Sweetness, together with crispness and juiciness, is a primary target for consumer-focused apple breeding programs. Due to the complexity of sweetness evaluation by a trained panel, soluble solid content (SSC) is commonly used to approximate the sugar content and estimate this parameter. However, it has been demonstrated that SSC and sweet taste are poorly correlated. Even poorer correlation was found between total sugars and sweetness. With this study we investigate the influence of single sugars (sucrose, glucose, fructose, xylose), and sorbitol content on perceived sweetness in apples, in association with malic acid (the main organic acid in apple) and SSC evaluation. Furthermore, we studied whether other important features as volatile compound release and texture may influence apple taste. The aim is to better understand the multisensory nature of sweet taste and set an optimum model for the instrumental prediction of sweetness. A combined sensory/instrumental approach was used on 40 apple batches covering a wide range of sensory variability. Descriptive sensory analysis was carried out by a trained panel based on a consensus-developed lexicon. Simultaneously, on the same apple samples, single sugars, sorbitol, organic acids, volatile compounds, texture parameters (mechanical and acoustic) and SSC were determined. Multivariate correlation models (O-PLS) were developed in order to highlight relationships between sweetness and instrumental data. A model based on soluble components explains less than 60 % of perceived sweetness. Models considering also volatile compounds and texture parameters explain more than 90 % of perceived sweetness. These results confirm that sensory perception is regulated by multisensory interactions between taste, flavor and texture and that a proper model to estimate sweetness on the basis of instrumental parameters can not relay only on SSC. These results should help breeders in the selection of new cultivars with better perceived quality.