Dietary experience modifies horses' feeding behavior and selection patterns of three macronutrient rich diets.

Choice feeding is often used to investigate an animal's nutritional requirements and dietary preferences. A problem with this approach is that animals with long gut transit times, such as the horse, may find it difficult to associate a chosen food with its nutritional consequence when alternative foods are presented simultaneously. One solution is to present foods singly for a period of time before a simultaneous choice session to allow the development of learned associations. This method was used to determine if horse's voluntary intake and feeding behavior was influenced by the macronutrient composition of the diet. Seven stabled horses, maintained on a low intensity exercise regimen, were allowed, on an ad libitum basis, haylage and 3 isocaloric forage based diets that were rich in 1 of 3 macronutrients (protein, lipid, and hydrolyzable carbohydrate). Initially, diets were presented as a 3-way choice for 5 d (self-selection a [SSa]), then singly (monadic phase) with exposure to each diet for 2 separate periods of 3 d each, and finally again as a choice for 5 d (self-selection b [SSb]). The total amount of trial diet offered differed with trial phase, with 2 to 2.5% of BW during SSa and the monadic phase, increasing to ad libitum access during SSb. To control differences in the total amount of trial diet offered, 2 measurements of voluntary intake were taken at 4 and 22 h postpresentation. Daily macronutrient and energy intakes were estimated from proximate analysis of the trial diets and batches of haylage fed. Feeding behavior was observed over a single 4-h period during both self-selection phases. Horses showed no initial preference after 4 h for any 1 diet during SSa. Following the monadic phase, horses demonstrated a preference for the protein and hydrolyzable carbohydrate rich diets over the lipid rich diet (P < 0.001). Dietary experience modified foraging behavior as the total number of visits to the diets decreased during SSb (P < 0.005). Analysis of 24 -h macronutrient consumption showed that protein and hydrolyzable carbohydrate intake increased during SSb, whereas lipid intake remained constant over both self-selection phases (P < 0.001). These data indicate for perhaps the first time that horses can respond to dietary macronutrient content and that single presentations during choice studies facilitates expression of dietary preferences.