The effects of the odor training (odors paired with representative images) on desire to drink, as measured by the AUQ, were estimated in all 10 subjects using a 2(Session) × 4(Stimulus) within subject analysis of variance, with the Stimulus condition including the baseline and three odorant/image classes, and Session representing infusion type. There were no significant main effects for Session, or Session × Stimulus interactions. There was, however, a significant main effect for Stimulus (F = 5.5, p = 0.005). A planned comparison showed that the AUQ score after AO (and their representative images; 2.88 ± 1.67) was significantly greater than the average AUQ score over the remaining conditions (baseline, NApO and ApCO; 2.18 ± 1.07, F = 8.00, p < 0.05). Mood ratings were not significantly different between placebo and alcohol sessions, and were therefore averaged for analysis. As assessed using two repeated measures analyses of variance, neither positive (5.20 ± 1.25) nor negative (1.90 ± 0.84) mood at baseline differed from mood as measured after exposure to NApO (positive = 5.80 ± 1.14; negative = 1.75 ± 0.92),
