Carbohydrate- and protein-conditioned flavor preferences: effects of nutrient preloads.

Food-deprived rats were trained to associate one flavor (CSProt) with intragastric (IG) infusions of protein (PROT; 10% calcium caseinate), a second flavor (CSCHO) with IG infusions of carbohydrate (CHO; 10% protein), and a third flavor (CS-) with IG water infusions during 30 min/day training sessions. (The CS flavors were cherry, grape, and orange saccharin solutions.) In subsequent two-bottle tests the rats reliably preferred both the CSProt 2nd CSCHO to the CS- and equally preferred the CSProt and CSCHO. The preference for the two nutrient-paired flavors was not altered by IG preloads of PROT or CHO delivered as three loads 120, 40, and 5 min prior to testing. However, single oral + gastric preloads of CSCHO + IG CHO and CSProt + IG PROT 45 min prior to test selectively increased the preference for the CSProt and CSCHO, respectively. In subsequent gastric-only and oral-only tests single IG preloads of PROT and CHO, but not CSProt and CSCHO preloads, selectively altered the rats' preference for CSCHO vs. CSProt. In a second experiment with new rats, oral + gastric preloads again selectively altered the preference for the CSCHO vs. CSProt, but gastric-only preloads failed to have this effect. These results demonstrate that rats can learn to associate different flavors with the postingestive effects of different nutrients, and modify their flavor preferences after nutrient preloads. Oral + gastric preloads were most effective in altering flavor preferences, whereas gastric-only preloads had inconsistent effects and oral-only preloads were ineffective.