Figure 3

Increase of [Ca ²⁺ ]i using caffeine triggers long-term memory formation. Retention performances following an injection of caffeine (20 mM) or saline solution 20 min before one- or three-trial conditioning. A. Conditioned response (CR) to the learned odor at 72 h was significantly higher for the caffeine group (n = 60) than for the one-trial conditioning group (n = 78) (χ² = 10.3, P = 0.0013), but not different from the three-trial conditioning group (n = 68) (χ² = 1.64, P = 0.2). However, the response to the new odor was significantly different between the caffeine group and the one-trial conditioning (χ² = 4.7, P = 0.03). Nevertheless, the caffeine group responded in the same way to the new odor than the three-trial conditioning group (χ² = 3.2, P = 0.07). In addition, the control one-trial conditioning, the caffeine group and the control three-trial conditioning responded significantly more to the learned odor than to the new odor (respectively: McNemar χ² = 14.4, P < 0.001; McNemar χ² = 19.3, P < 0.001; McNemar χ² = 23.3, P < 0.001). Overall, specific response (SR) proportion of the caffeine group was significantly increased compared with those of the control one-trial conditioning (χ² = 3.9, P = 0.049) and was not different from the control three-trial conditioning (χ² = 0.9, P = 0.33). B. The percentage of specific responses (% SR) for caffeine and for the control one-trial conditioning at 3 h (Control: n = 95; Caffeine: n = 78) and at 24 h (Control: n = 77; Caffeine: n = 84) were not affected by caffeine treatment (respectively: χ² = 0.2, P = 0.62; χ² = 0.8, P = 0.37). The % SR presented at 72 h corresponds to the data of Figure 3A (*: P < 0.05, **: P < 0.01, ***: P < 0.001, NS: non-significant).