Strawberry sweetness and consumer preference are enhanced by specific volatile compounds

Summary

This seven-year multi-site study of 148 strawberry cultivars and breeding selections combined sensory evaluation (>15,000 consumer data points from panels of ≥100 consumers) with chemical analysis to identify volatile compounds driving sweetness perception and consumer preference independently of sugar content. Using partial least squares regression and machine learning, the authors identified 20 volatiles enhancing sweetness perception and 18 enhancing overall liking; predictive models incorporating volatiles explained substantially more variation than those based on sugars and acids alone. The findings identify specific volatile targets (γ-dodecalactone, 5-hepten-2-one 6-methyl, and medium-chain fatty acid esters) and two genetic loci on linkage group 6A controlling ester production, supporting a breeding paradigm where consumer preferences drive identification of chemical and ultimately genetic targets.

UK applicability

The identified volatile compounds and genetic markers could inform strawberry breeding programmes in the United Kingdom, though UK growing conditions and consumer preferences may differ from the United States study context. The methodology—linking sensory evaluation to chemical composition through large consumer panels—is transferable and could support development of more flavourful strawberry varieties suited to UK production and consumer markets.

Key measures

Consumer sensory ratings (sweetness intensity, texture liking, flavour intensity, sourness intensity, overall liking); quantification of three sugars, two acids, and 113 volatile compounds; partial least squares regression analysis; machine learning predictive models; genetic association loci

Outcomes reported

The study identified 20 volatile compounds that increase sweetness perception independently of sugar content, and 18 volatiles that increase consumer liking independently of sugars, through sensory evaluation of 148 strawberry samples across seven years with over 15,000 consumer data points. Machine learning models incorporating volatile compounds explained at least 25% more variation in sweetness and liking than models based on sugars and acids alone.

Topic tags