Long-lasting memories of adverse experiences are essential for individuals' survival but are also involved, in the form of recurrent recollections of the traumatic experience, in the aetiology of anxiety diseases (e.g., post-traumatic stress disorder [PTSD]). Extinction-based erasure of fear memories has long been pursued as a behavioral way to treat anxiety disorders; yet, such a procedure turns out to be transient, context-dependent, and ineffective unless it is applied immediately after trauma. Recent evidence indicates that, in both rats and humans, extinction training can prevent the return of fear if administered within the reconsolidation window, when memories become temporarily labile and susceptible of being updated. Here, we show that the reconsolidation–extinction procedure fails to prevent the spontaneous recovery of a remote contextual fear memory in a mouse model of PTSD, as well as the long-lasting behavioral abnormalities induced by traumatic experience on anxiety and in both social and cognitive domains (i.e., social withdrawal and spatial learning deficits). Such a failure appears to be related to the ineffectiveness of the reconsolidation–extinction procedure in targeting the pathogenic process of fear sensitization, a nonassociative component of traumatic memory that causes animals to react aberrantly to harmless stimuli. This indicates fear sensitization as a major target for treatments aimed at mitigating anxiety and the behavioral outcomes of traumatic experiences.

This review will consider studies concerning the effects of cannabinoid receptor agonists and antagonists on memory in laboratory animals. Two subtypes of cannabinoid receptors have been identified to date; the central CB1 subtype and the peripheral CB2 subtype. The receptor which specifically binds Δ9-tetrahydrocannabinol (Δ9-THC) and related compounds in rat and human brain has been discovered and cloned by a number of researchers. This cannabinoid receptor is localized with high concentrations in different brain areas, including hippocampus and amygdala, which play an important role in the modulation of memory. In recent years evidence has been obtained that cannabinoids influence memory processes. It has been shown, for example, that Δ9-THC impairs memory in rats, mice and monkeys tested in a variety of experimental conditions (radial maze, instrumental discrimination tasks, Morris water maze, etc.). In some of these researches the effect of Δ9-THC was antagonized by the CB1 receptor antagonist SR 141716A, showing the involvement of this subtype of cannabinoid receptor in its effect. Anandamide, arachidonylethanolamide, was recently discovered as the first endogenous ligand for the cannabinoid receptor. It has been reported to stimulate CB1 receptors and to mimic the pharmacological effects of cannabinoids. Experiments carried out by our group have shown that anandamide impairs memory consolidation in random bred mice (CD1), exerts genotype-dependent influences on memory in inbred strain of mice (C57 BL / 6 and DBA / 2), and that opioid and dopaminergic systems might be involved in its effects.