Instrumental classification of beer based on mouthfeel

Beer is one of the most consumed alcoholic beverages in the world. Classification of beer helps the consumer to find a preferred beer. Sensory assessments of taste are commonly done by sensory panels and therefore sus-ceptible to subjectivity. Mouthfeel is an important parameter for the total perception of beer flavor. Three di-mensions of mouthfeel are distinguished: contracting, coating, and drying. In this study 24 beer samples were evaluated chemically. The data were matched with sensorial data obtained from a trained panel. Different chemical analyses were performed; total acidity (TA), total flavonoids (TF), total polyphenols (TPC), total sugars (TS), color, pH, carbon dioxide content, ethanol, bitterness units (BU) and total iso-alpha-acids (TIA). The data were analyzed by performing several statistical techniques such as analysis of variance, principal component analysis, agglomerative hierarchical cluster analysis and multiple factor analysis. Sensory data obtained from trained panelists on the different mouthfeel attributes correlated with the data found instrumentally. The drying dimension could be expressed using the TPC, BU, TIA and pH. Contracting compounds correlated positively with TA and negatively with pH. As expected, ethanol was strongly associated with burning sensations and carbon dioxide with carbonation. The results of this experiment indicate that commercial beers can be classified into three mouthfeel attributes: drying, coating and contracting. The Principal Component Analysis (PCA) in this study confirmed the dimensions of the mouthfeel model and identified drying and coating as opposites, con-tracting forces interact on these dimensions. Moreover, these attributes were shown to be quantifiable by instrumental analysis which suggests that a data-driven approach based on mouthfeel could reduce subjectivity in the analysis of taste.