In recent years, consumer preferences have shifted towards better-quality rice, particularly towards varieties with good eating quality. Texture is an extremely important attribute for cooked rice and has been used as an indicator for consumer acceptance. Cooked rice texture is affected by a wide range of factors, such as the amylose content, postharvest processing, the milling ratio, the cooking method, etc., but the actual molecular reasons for the texture of cooked rice grains are still unclear. Since texture has been defined as a multidimensional characteristic that only humans can perceive, define, and measure, sensory descriptive analysis is a useful tool for characterizing texture properties of cooked rice. However, the cost associated with training and maintaining a descriptive panel has prompted many researchers to evaluate less costly and less time-consuming approaches. The overall objectives of this thesis are to explore the molecular mechanisms for the hardness and stickiness of cooked rice grains, increase understanding of the human textural perception of cooked rice, and develop an improved instrumental method to evaluate and/or predict the texture of cooked rice.The first chapter of this thesis reviews current understanding of the texture of cooked rice, which involves the factors affecting rice texture, the evaluation methods for cooked rice texture, and the scientific questions generating from the literature review and associating to the overall objectives of this thesis.In chapter 2, statistically and causally meaningful relationships are established between starch molecular structure (the molecular size distribution of whole (branched) starch and the chain length distribution of debranched starch) and texture (hardness and stickiness) of cooked rice grains. The amounts of amylose chains with degree of polymerization (DP) 100-20000, and of long amylopectin chains, positively correlate with hardness, while amylopectin chains with DP<70 and amylose molecular size both show negative correlations with hardness (p<0.05).There is also a significant negative correlation between stickiness and the amounts of long amylopectin chains (p<0.01). For rices with similar amylose content, the amount of amylose chains with DP 1000-2000 positively correlates with hardness while size negatively correlates with hardness (p<0.05). This indicates for the first time that, regardless of amylose content, rice varieties with smaller amylose molecular sizes and with higher proportions of