Abstract

The acquisition of behaviours that enhance the acceptability of new feeds could improve the performance and animal welfare in the pig industry. Pigs may learn individually by trial and error and/or by social learning to consume and prefer new flavoured feeds. However, there is little information regarding the effects of learning on the palatability of those flavours. This study aimed to investigate possible changes in pigs? hedonic responses, estimated by the mean consumption time per approach (CT/A), for flavours previously learnt through individual experiences (Experiment 1) and socially by a brief interaction with an experienced conspecific (Experiment 2). A total of 96 nursery pigs were used to study their preference, acceptability and palatability for flavours previously learnt about. Flavours were associated with 4% sucrose (Experiment 1) or with a demonstrator that recently consumed the flavour (Experiment 2). Those flavours or positive conditioned stimulus (CS+) were compared with control flavours that were not positively conditioned (CS-). In experiment 1, pigs preferred (P = 0.010) and tended to accept more (P = 0.069) of the CS + than CS- in water solutions. Nevertheless, no palatability differences were found (P = 0.875). In experiment 2, observer animals had a higher intake of CS + following interaction with demonstrators that consumed the same flavoured feed (P = 0.041). Snout to snout interaction time between those demonstrators and observers tended to present a positive correlation with CT/A (r = 0.497, P = 0.1). Thus both individual trial and error learning, and social learning, increased pigs acceptance of, and preference for, food flavours. However, only social learning modified the pattern of consumption with the observation of a correlation between changes in the CT/A ratio and the degree of snout to snout contact time between demonstrators and observers. Thus the mechanisms and effects of social and individual learning appear to differ at least partially. Regardless of mechanism, the fact that associative learning can increase food flavour consumption suggests that interventions based on associative learning may be beneficial for addressing welfare problems in pig production linked to low consumption, in particular with respect to weaning where the low consumption may relate to the inexperience of the animals.