Taste is expected to represent a food's nutrient content. The objective was to investigate whether taste acts as nutrient-sensor, within the context of the current diet, which is high in processed foods. Intensities of the five basic tastes of fifty commonly consumed foods were rated by nineteen subjects (aged 21·0 (sd 1·7) years, BMI 21·5 (sd 2·0) kg/m2). Linear regression was used to test associations between taste and nutrient contents. Food groups based on taste were identified using cluster analysis; nutrient content was compared between food groups, using ANOVA. Sweetness was associated with mono- and disaccharides (R2 0·45, P < 0·01). Saltiness and savouriness were correlated, with r 0·92 (P < 0·01) and both were associated with Na (both: R2 0·33, P < 0·01) and protein (R2 0·27, P < 0·01 and R2 0·33, P < 0·01, respectively). Cluster analysis indicated four food groups: neutral, salty and savoury, sweet–sour and sweet foods. Mono- and disaccharide content was highest in sweet foods (P < 0·01). In salty and savoury foods, protein content (P = 0·01 with sweet–sour foods, not significant with neutral or sweet foods) and Na content (P < 0·05) were the highest. Associations were more pronounced in raw and moderately processed foods, than in highly processed foods. The findings suggest that sweetness, saltiness and savouriness signal nutrient content, particularly for simple sugars, protein and Na. In highly processed foods, however, the ability to sense nutrient content based on taste seems limited.