A breakage index for characterizing in vitro nut fragmentation and predicting human oral fragmentation

Breakage of food influences eating experience and sensory perception. The aims of the study were to identify an appropriate breakage index and to develop an in vitro method for predicting the ease of oral breakage of nuts. Kernels of five types of nuts were fragmented in vitro using a texture analyzer and 12 subjects therefore performed molar bites. In addition, peanuts were differently roasted (over 0, 15, 25, and 35 min) to vary texture within the same nut type. Projected particle areas were determined using imaging. Two Breakage Indices were compared (1) BI-I, the difference, after and before fragmentation, in square root values of ratios between total projected area and volume [Agrawal et al., 1997, Archives of Oral Biology, 42(1), 1-9], and (2) BI-II, the ratio of the total projected area after and before fragmentation. BI-II gives a stronger linear regression than BI-I between in vivo and in vitro index values for different types of nuts; Pearson's r = 0.834 versus 0.499 (12 subjects with all data pooled). Using BI-II, a subject's regression result in fragmentation tests with differently roasted peanuts was as strong as when testing different nut types: Pearson's r = 0.984 versus 0.964. Since the range of the in vitro BI-II values was 5.5 times smaller in the peanut tests, the finding of a similarly strong regression indicates a high sensitivity of BI-II to detect differences in food texture. BI-II is useful for food industry to determine how easily solid foods break down and thereby compare the potential of flavor release between foods during chewing. Kernels of five types of nuts were fragmented in vitro using a texture analyzer and 12 subjects therefore performed molar bites. In addition, peanuts were differently roasted (over 0, 15, 25, and 35 min) to vary texture within the same nut type. Projected particle areas were determined using imaging. Two Breakage Indices were compared (1) BI-I, the difference, after and before fragmentation, in square root values of ratios between total projected area and volume, and (2) BI-II, the ratio of the total projected area after and before fragmentation. BI-II gives a stronger linear regression than BI-I between in vivo and in vitro index values for different types of nuts; Pearson's r = 0.834 versus 0.499, 12 subjects. Using BI-II, a subject's regression result in fragmentation tests with differently roasted peanuts was as strong as when testing different nut types: Pearson's r = 0.977 versus 0.964. The finding of similar regression strength suggests that BI-II is sufficiently sensitive to detect differences in food texture. BI-II is useful for food industry to determine how easily solid foods break down and to compare the potential of flavor release between foods during chewing.image