Taste preferences and feeding behaviour in the facultative herbivorous fish, Nile tilapia Oreochromis niloticus

Taste preferences in fishes are known mainly for carnivorous species, whereas herbivorous consumers were rarely used in such studies. The main goal of the present study was to evaluate the taste preferences in the herbivorous African cichlid fish, Nile tilapia Oreochromis niloticus. In laboratory settings, the palatability of widely used taste substances (four taste substances that are considered to be sweet, sour, bitter and salty for humans - sucrose, citric acid, calcium chloride and sodium chloride; 21 free L-amino acids; 12 sugars and artificial sweetener Na-saccharin; 0.1-0.0001 M) was evaluated. In each trial, a standard agar pellet flavoured with a substance was offered for fish individually. The consumption of pellet, the number of grasps and the retention time before the pellet was finally ingested or rejected were registered. Overall, 21 of 38 substances were palatable, whereas other substances did not shift consumption of pellets in relation to blank pellets. Pellets containing citric acid, L-cysteine, L-norvaline, L-isoleucine, L-valine, Na-saccharin and D-sorbitol were consumed in >85% of trials. Taste attractiveness of amino acids was highly species-specific and was not associated with the trophic category of the 19 species compared. Moreover, it did not correlate with dietary quantitative requirements of Nile tilapia (r_s  = 0.27; P > 0.05). Palatability of sugars for O. niloticus and their sweetness for humans did not correlate as well (r_s  = 0.21; P > 0.05); nonetheless, Na-saccharin has the most attractive taste for both O. niloticus and humans. The most palatable amino acids lost their effect if the concentration was lowered to 0.01 M for L-cysteine and 0.001 M for L-norvaline (lower than 242.3 μg and 23.4 μg per a pellet, respectively). Single pellet grasp was characteristic of O. niloticus feeding behaviour (>95% of trials), and this pattern may be related to the social lifestyle of this species. Fish spent 4-8 s on average for orosensory evaluation of pellet edibility. The retention time correlated with the palatability of substances and was significantly longer in trials that ended up with pellet swallowing. It is suggested that prolonged orosensory evaluation of food before swallowing provides a reliable and accurate sensory evaluation, which, in turn, can reduce the probability that inadequate food will be consumed.