Consumer and instrumental evaluation of Maine wild blueberries for the fresh pack market
D. W. Donahue, P. W. Benoit, B. J. Lagasse and W. R. Buss
Postharvest Biology and Technology Vol: 19 Issue: 3 Pages: 221-228
2000
บทคัดย่อ
Maine wild blueberries prepared for the fresh pack market compete with cultivated blueberries for market share. In an effort to increase market share, Maine producers are seeking an understanding of consumer preference and how it relates to objective blueberry quality measures. Consumer and instrumental measurements were evaluated over a 3-year period from 1996 to 1998. Freshly harvested blueberries were sized according to diameter; 6¯8, 9¯10 and 11¯12 mm. A consumer panel, recruited from the university and surrounding communities, ranged in age from 18 to 62 and was geographically and gender diverse. The panel was presented the various size classes of blueberries monadically in balanced random order and asked to rate the samples on a nine-point hedonic scale. Quasi-static compression measurements of firmness were paired with these individual sensory measurements. Statistical regression models and back propagation neural networks were used to determine and predict consumer response from the instrumental response. There seemed to be a compression of the consumer data to the upper end of the hedonic scale. Year and time of season were significant for the overall consumer response and the force and firmness instrumental measurements. Berry size was significant with respect to all sensory and instrumental measurements. The results of the overall sensory and instrumental evaluation revealed that size class 9¯10 mm was the ideal size for fresh market selection, however, panelists preferred the large (11¯12 mm) berries in visual acceptance evaluations. The statistical regression analysis revealed no linear relationships except for the size class 9¯10 mm, where positive linear relationships were found between sensory and instrumental measurements. The neural networks were poor predictors of the sensory data based on the instrumental data. These findings highlight the difficulties in characterizing consumer preference with instrumental and hedonic measurements.