Does energy intake influence diet selection of novel forages by horses?

While it has been shown that diet selection by equids can be influenced by nutritional factors, it is presently unclear how diet choices by horses are modulated when animals are faced with a familiar-novel dichotomy. Therefore, this study aimed to investigate the effect of energy intake on foraging behaviour in horses and the selection of familiar (FF) and nutritious novel (NF) forages. Twelve adult mares were maintained in yards during feeding and were housed in two groups in barren paddocks when not feeding. The experiment lasted for 8 weeks; in weeks 0, 4 and 8 horses received a maintenance diet and in weeks 1-3 and 5-7 horses were switched to a low energy (LE; 80% RDI) or high energy (HE; 120% RDI) diet. In weeks 3 and 7 a two-choice test was presented in a split-plot design, with 12 horses being the main-plot units. The LE and HE diets were applied at the main-plot level (n=6) while the subplots were the 10 min feeding intervals (combined over 3 consecutive days) where horses (one at a time) were introduced to a forage preference test (1 to 4). The tests were made up as pairs, based on the nutritional profile, with one being FF (oaten or lucerne chaff) and the other feed was NF (bamboo, tagasaste, willow or saltbush leaf chaff) and presented in a checkerboard design. In wk 8 the preference of all forages together was examined. Forage intake, the number of visits to each zone/bucket and time spent foraging or moving toward each zone/bucket were recorded. Horses demonstrated a greater preference for FF, but there were no differences between the LE and HE diet groups in the proportions of intake of NF. There was a higher acceptance of NF on Day 1, which declined on Day 2 and 3 for both diet groups (P<0.001). The LE group had a marginally higher proportion of zones visits to the NF compared to the HE group (P=0.009). In the final preference test, horses showed a greater acceptance of NF willow, bamboo and saltbush compared with tagasaste (P=0.01). Some possible explanations for the neophobic response seen in this study are discussed.