a, Bayesian decoding of position based on CA1 activity (dark-shaded bars) in one example animal was significantly greater than chance performance (light-shaded bars) in all five sessions of the three-chamber task. Example shown for one wild-type animal. (*p<0.05, **p<0.01, **p<0.001, two-sided Wilcoxon Rank Sum). b, Example Bayesian decoding of position from CA2 activity in two animals in the three-chamber task. For all 6 wild-type animals examined, the Bayesian decoder for position based on CA2 neuron activity never performed significantly better than chance. (p>0.05, two-sided Wilcoxon Rank Sum). P values relative to chance for all WT mice with CA2 recordings were equal to: WT7, p=0.48; WT2, p=0.36; WT3, p=0.06; WT5, p=0.17; WT10, p=0.19; WT9, p=0.49. Box plots display the center line as the mean; box limits are upper and lower quartiles; whiskers show min to max values in data sets. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.