Prediction of Long-Grain Rice Texture from Pasting Properties1

Textural attributes of cooked rice (100 samples) were predicted using pasting properties. Two textural attributes (hardness and stickiness) were evaluated by a texture analyzer and seven pasting-property parameters as well as profile data measured by a rapid visco-analyzer (RVA) were used to evaluate predictive models of the textural attributes using Partial Least Squares Regression. The textural attributes evaluated by a texture analyzer were relatively well predicted using data from RVA profiles (optimized Rcal>0.77). However, the RVA parameters did not seem to be good predictors of the textural attributes of cooked rice.