Background: Food palatability increases food intake and may lead to overeating. The mechanisms behind this observation are still largely unknown. Objectives: The aims of this study were the following: 1) to elucidate the plasma responses of endocannabinoids, N-acylethanolamines, and gastrointestinal peptides to a palatable (sweet), unpalatable (bitter), and sensory-acceptable (tasteless control) food, and 2) to verify whether some of these bioactive compounds can serve as plasma biomarkers of food liking in humans. Methods: Three puddings providing 60 kcal (35% from proteins, 62% from carbohydrates, and 3% from fats) but with different taste were developed. Twenty healthy subjects (11 women and 9 men; mean age 28 y and BMI 22.7 kg/m2), selected because they liked the puddings in the order sweet > control > bitter, participated in a randomized crossover study based on a modified sham feeding (MSF) protocol. Blood samples at baseline and every 5 min up to 20 min after the MSF were analyzed for gastrointestinal peptides, endocannabinoids, and N-acylethanolamines. Thirty minutes after the MSF, energy intake at an ad libitum breakfast was measured. Results: After the MSF, no response was observed in 7 of 9 gastrointestinal peptides measured. The plasma ghrelin concentration at 20 min after the sweet and bitter puddings was 25% lower than after the control pudding (P = 0.04), and the pancreatic polypeptide response after the sweet pudding was 23% greater than after the bitter pudding (P = 0.02). The plasma response of 2-arachidonoylglycerol after the sweet pudding was 37% and 15% higher than after the bitter (P < 0.001) and control (P = 0.03) puddings, respectively. Trends for greater responses of anandamide (P = 0.06), linoleoylethanolamide (P = 0.07), palmitoylethanolamide (P = 0.06), and oleoylethanolamide (P = 0.09) were found after the sweet pudding than after the bitter pudding. No differences in subsequent energy intake were recorded. Conclusions: The data demonstrated that food palatability influenced some plasma endocannabinoid and N-acylethanolamine concentrations during the cephalic phase response and indicated that 2-arachidonoylglycerol and pancreatic polypeptide can be used as biomarkers of food liking in humans.