Extensive research has shown that the hippocampus is involved in learning and memory. Complementary studies have focused on long-term potentiation (LTP) as a physiological model of memory. It is perhaps paradoxical that fear-provoking experiences, which typically generate strong memories of the experience, can block the induction of hippocampal LTP (Kim & Diamond, 2002). Why should an experience that produces long-lasting memories impair LTP? My presentation will provide cognitive and physiological perspectives which may help to resolve this paradox.

First, I will discuss research from my group showing that psychological stress has no effect on the induction of LTP (100 pulses in 1 s in CA1 in vivo and in vitro, but does block a low threshold form of LTP, referred to as primed burst (PB) potentiation (induced by a total of 5 physiologically patterned pulses). This finding indicates that stress exerts an inhibitory bias on CA1 plasticity which is more readily observed in response to threshold (PB) than suprathreshold (LTP) stimulation (Diamond & Park, 2000). Second, we found that stress selectively impaired hippocampal-dependent, but not hippocampal-independent, spatial memory. Third, we have studied the relationship between corticosterone (CORT), an adrenal stress hormone, and stress effects on PB and spatial memory. There was an inverse relationship between stress-induced increases in serum cort levels and hippocampal processing, i.e. an increase in the level of CORT correlated with a reduced magnitude of PB and impaired spatial memory.

Although these findings indicate that CORT is involved in the stress-induced modulation of the hippocampus, we have also found that stress can impair spatial memory in adrenalectomized (ADX), i.e. CORT depleted, rats. Furthermore, exogenous CORT administration, alone, does not mimic the impairing effects of stress on memory. Finally, we and others have found other conditions in which there can be an elevation of CORT levels without there being an adverse effect on hippocampal-dependent memory or PB/LTP. For example, a lesion of the amygdala or administration of an antidepressant (tianeptine) can eliminate stress effects on the hippocampus without reducing stress-evoked increases in CORT levels (Kim & Diamond, 2002). Conversely, we have found that male rats exposed to an estrous female rat exhibited stress levels of CORT, but the male rats exposed to a female rat did not exhibit a spatial memory impairment. CORT, therefore, appears to be involved in the emotional modulation of hippocampal functioning, but the elevation of CORT needs to occur in conjunction with a fear-induced behavioural state to impair hippocampal processing.

These findings indicate that stress exerts a suppressive influence on hippocampal functioning. The challenge is to understand how these findings reconcile with the observation that emotional experiences can produce powerful and long-lasting memories. I will propose that the stress experience, itself, activates hippocampal storage mechanisms (i.e. stress generates endogenous LTP) via NMDA receptor activation. The stress-induced saturation of endogenous hippocampal plasticity then interferes with the induction of electrically induced PB. In summary, my presentation will offer a functional perspective to explain our findings showing that: (1) stress impairs hippocampal-dependent memory and PB, and (2) elevated serum CORT levels, in conjunction with a fear-induced behavioural state, impair hippocampal-dependent memory and plasticity.

This work was supported by grants from the U.S. Veterans Administration and Servier.