I am not sure how much help I can provide but this is certainly an interesting observation. Do you have some literature on this observation? I have not seen the literature yet but my guess is that it might have to do with the following general reasons:
1. Relative grain-filling efficiencies; environmental factors are known to influence grain filling genes related to eating quality. For instance, ambient temperature is know to influence key eating quality traits like amylose content as well as amylopectin chain length distribution in rice grains. So your observation is a possibility.
2. Difference in fertility levels; lowland (submerged) soils are generally more fertile than upland (aerobic) soils, especially in the tropics. If you live in the tropics you might want to compare soil properties of the growth media or repeat your experiment under more controlled conditions.
3. Genetic background; If your observation was made with rice from different backgrounds then there is no basis for this. Otherwise you might need to compare the same variety (a lowland-upland or plastic variety wil be very useful in this case)