Fig. 5. Extinction and conditioned taste aversion memory. (A) Naive snails (N=50) were first challenged with the sucrose taste. This elicited robust feeding. Snails were then conditioned taste aversion (CTA) trained. 10 min after training, snails were given the conditional stimulus (CS) alone (i.e. an extinction session) and the elicited feeding response noted. Snails were designated as either good (N=40; closed boxes) or poor (N=10; open boxes) performers. All snails received two more (at 10 min intervals) extinction sessions. Again in each extinction session the elicited feeding response was noted. All snails were then challenged with the sucrose stimulus 1 and 24 h later. Extinction did not occur. The three extinction sessions did not result in a behavioral phenotype resembling the naive state in either the good or the poor performers. (B) As in A, except carrot juice was used as the CS. Similar results were obtained. All data are means ± s.e.m. The differences between the feeding response of the good performers and that of the poor performers in the both cases were maintained for at least 24 h with *P<0.01 (Student's t-test), showing that the memory formed by CTA training cannot be extinguished.