**Figure 3.** **Intermediate-term retention of everyday spatial memory in aging.** (**A**) Rats received a strong encoding trial (3 rewards, filled circle). Six hours later, they were tested in a probe trial with 5 non-rewarded sandwells (open circles). (**B**) The percentage of correct digging was not significantly above chance (dashed line) in older rats without prior training (one-sample *t*-test, t_{12} = 1.88, *p* = 0.085) but significantly above chance in older rats with prior training (one-sample *t*-test, t_{12} = 4.45, *p* < 0.001). The group difference was significant (unpaired *t*-test, t_{24} = 2.2, *p* = 0.04). (**C**) Similar to procedures in A except that exploration in a novel box (green box) was introduced or omitted at 30 min after a weak encoding trial (1 reward). (**D**) An example of a novel box. (**E**) In rats with no prior training, the percentage of correct digging was not significantly above chance (dashed line; one-sample *t*-test, t_{12} = 1.74, *p* = 0.11) after weak encoding and was significantly above chance after weak encoding with novelty (one-sample *t*-test, t_{11} = 2.89, *p* = 0.015). No difference was observed between the absence or presence of novelty (paired *t*-test, t_{11} = 0.87, *p* = 0.4). (**F**) In rats with prior training, the percentage of correct digging was significantly above chance (dashed line) in both conditions (absence of novelty: one-sample *t*-test, t_{12} = 2.8, *p* = 0.02; presence of novelty: one-sample *t*-test, t_{12} = 4.34, *p* = 0.00). No difference was observed between the absence or presence of novelty (paired *t*-test, t_{12} = 0.99, *p* = 0.34). Data are presented as mean ± SD. ^{*}*p* < 0.05, ^{***}*p* < 0.005.