Ruminants generally succeed in selecting an appropriate diet relative to their nutrient requirements, and macronutrients such as energy and protein are considered as playing a primary role in their food preferences. Diet learning through post-ingestive consequences, evidenced by many studies that involved situations with simple discrimination tasks, may explain this ability to select a suitable diet However, in more complex feeding environments where animals are faced with many plants varying both in their energy and protein contents, the extent to which such a mechanism operates is still uncertain. Our objective was to investigate a more complex feeding situation than previously tested and to determine whether lambs were able to associate three forages with both energy-based and protein-based post-ingestive consequences of various intensities. The doses of starch and casein used to elicit these post-ingestive consequences were chosen so that the apparent metabolisable energy and crude protein values of hays were within the range found in plants normally encountered under field conditions. We applied five treatments: EmPm, EhPm, EmPh, EhPh, where medium (m) or high (h) levels of nutrients (E = energy and P = protein) were associated with different hays, and E0N0 where no stimuli were associated. Each animal experienced only three treatments: E0N0-EmPm-EhPm or E0N0-EmPh-EhPh, each one being associated with one hay during conditioning, on the basis of one treatment per day over three consecutive days. Animals then had a choice between the hay associated with E0N0 and one of the two others, successively on the two following days. This procedure was repeated through five 1-week-long periods, and applied to two groups of lambs (n = 12) maintained at feeding levels sufficient for maintenance or for a 150 g/day growth rate. This experiment was complemented by a methodological trial that aimed to estimate the doses of starch and casein eliciting preferences and a post-trial that aimed to evaluate the effect of our treatments on sheep ruminal environment. The methodological trial showed that lambs perceived the stimuli and the doses, and the post-trial showed that treatments affected the ruminal environment. In the main experiment, the lambs preferred the rewarded hays but did not discriminate in their choices the variation of post-ingestive consequences due to energy and protein. This suggests that in situations closer to reality where animals experience many foods varying in nutrient contents, the learning process may be impaired. The lambs do not seem to have strictly relied on it for shaping their diet choices.