No evidence of flavour-nutrient learning in a two-week 'home exposure' study in humans

Flavour-nutrient learning is robust in animals but remains elusive in humans. Recent evidence suggests flavour-nutrient learning may be more likely to occur with beverages that contain relatively few calories (compared to no calories), while others show that learned associations can influence satiation, without an effect on preference. The objective of this research was to determine whether acquired liking for a caloric drink could be observed in a 'home learning' context over 2 weeks, and whether it is impacted by viscosity. In combination, we also explored changes in learning relating to fullness and expected satiety. In a double-blind study, participants (N = 83; BMI = 23.3 kg/m²) were randomly allocated to one of four groups differing in either calories (0 kcal vs. 112.5 kcal) or viscosity (low vs. high) and consumed a novel-flavoured drink over 15 days. Measures of flavour (10 ml sample) and beverage liking, grip force (a measure of beverage reward value), fullness, and expected satiety were taken at the start and the end of the study. While the high-viscous beverages were less liked (M = 40.3 mm, SD = 24.7) than the low viscous beverages (M = 64.4 mm, SD = 15.3; p = .022), there was no evidence that repeated exposure to a calorie-containing beverage impacted subsequent liking for the flavour (p = .115) or for the beverage (p = .448), grip force (ps > .26), fullness, and expected satiety (ps > .12). Accordingly, we conclude that we found no evidence of flavour-nutrient learning and flavour-satiety learning. This null finding accords with previous observations indicating that humans do not acquire flavour-nutrient associations as readily as some non-human animals.