Prior studies have obtained strong conditioned preferences (> 90%) for flavors paired with intragastric (IG) infusions of Polycose (hydrolyzed starch) over different flavors paired with IG water infusions. These experiments used concentrated Polycose solutions (16-32%; 0.6-1.2 kcal/ml) that contributed 50% or more of the animals' daily caloric intake. The present experiment sought to determine if calorically dilute solutions would also support preference conditioning. Adult female rats with gastric cannulas were trained with a cue solution (the CS+) paired with Polycose infusions. On alternate days a different cue solution (the CS-) was paired with IG water infusions. The CSs were unsweetened or saccharin-sweetened citric acid and sucrose octaacetate solutions that were available 22 h/day; chow was freely available. Separate groups were infused with Polycose at concentrations of 0.5%, 1%, 2%, and 4% (0.019-0.152 kcal/ml). The three higher concentrations conditioned flavor preferences of up to 89%. The infused Polycose represented only 1-14% of the animals' daily caloric intake and did not reliably reduce chow consumption. These findings demonstrate the sensitivity of the viscerosensory feedback system mediating flavor conditioning.