Fig. 5. Delay of crowding and LTM formation. (A) A cohort of naïve snails (N=23 operantly conditioned and N=23 yoked controls) was trained using the memory augmentation procedure and placed in an uncrowded aquarium for 1h immediately after training. Following this period, all snails were subjected to crowded conditions for 23h. All snails were then tested for memory (MT). Data were subjected to a repeated-measures ANOVA (F_45,3=10.982; P<0.01) followed by a Tukey–Kramer comparison test. Snails that were subjected to the operant conditioning procedure exhibited LTM (i.e. MT was significantly less than TS; ^*P<0.01). When these same snails were subjected to a change of context challenge (CT, carrot context) 2h later they did not exhibit LTM. That is, CT is not significantly different from TS (P>0.05). Snails subjected to the yoked control procedure also did not exhibit LTM (i.e. yoked is not significantly different from TS, P>0.05). In addition, the response of the yoked control snails was not significantly different from the response to CT. (B) Another cohort of naïve snails (N=40; 20 operantly conditioned and 20 yoked control snails) was subjected to operant conditioning and the yoked control procedure, respectively. Immediately after their respective training procedures they were placed into a crowded aquarium for 1h. Following this period, all snails were placed into an uncrowded aquarium for 23h. All snails were then tested for LTM. Data were subjected to a repeated-measures ANOVA (F_39,2=0.5398; P>0.05) followed by a Tukey–Kramer comparison test. The operantly trained snails do not exhibit LTM (MT is not significantly different from TS, P>0.05). In addition, the yoked control snails subjected to the same crowded and uncrowded conditions do not exhibit LTM (Yoked is not significantly different from TS). Values are means ± s.e.m.