The temporal dynamics model of emotional memory processing: a synthesis on the neurobiological basis of stress-induced amnesia, flashbulb and traumatic memories, and the Yerkes-Dodson law

Abstract

We have reviewed research on the effects of stress on LTP in the hippocampus, amygdala and prefrontal cortex (PFC) and present new findings which provide insight into how the attention and memory-related functions of these structures are influenced by strong emotionality. We have incorporated the stress-LTP findings into our "temporal dynamics" model, which provides a framework for understanding the neurobiological basis of flashbulb and traumatic memories, as well as stress-induced amnesia. An important feature of the model is the idea that endogenous mechanisms of plasticity in the hippocampus and amygdala are rapidly activated for a relatively short period of time by a strong emotional learning experience. Following this activational period, both structures undergo a state in which the induction of new plasticity is suppressed, which facilitates the memory consolidation process. We further propose that with the onset of strong emotionality, the hippocampus rapidly shifts from a "configural/cognitive map" mode to a "flashbulb memory" mode, which underlies the long-lasting, but fragmented, nature of traumatic memories. Finally, we have speculated on the significance of stress-LTP interactions in the context of the Yerkes-Dodson Law, a well-cited, but misunderstood, century-old principle which states that the relationship between arousal and behavioral performance can be linear or curvilinear, depending on the difficulty of the task.

Figure 1

A subset of data from Yerkes and Dodson [38]. Mice were trained to avoid shock in a simple versus difficult visual discrimination task. The simple task involved a dark versus bright discrimination and the more difficult task involved a discrimination in which the two sides of the escape box were at similar levels of illumination. Behavioral performance increased linearly with increased levels of shock in the simple task, but performance was maximal at an intermediate level of shock for mice trained in the difficult discrimination.

Figure 2

A comparison of the Hebbian version of the Yerkes-Dodson law, as it has been commonly represented for the past 50 years (a), and the original version, based on the actual findings and theorizing of Yerkes and Dodson ([38]; (b)). The Hebbian version incorrectly states that high levels of stress, anxiety, or motivation produce a monolithic impairment of performance. The original version based on the actual [38] Yerkes-Dodson findings takes into account the finding that strong emotionality can enhance performance under “simple” learning conditions, such as when learning involves focused attention on a restricted range of cues, and impairs performance under more complex or challenging learning situations, such as in divided attention, multitasking, and working memory tasks. Graph (a) is adapted from 5 decades of publications and books, for example, Hebb [53], Loftus [54], and Radvansky [55].

Figure 3

Temporal dynamics model of how stress affects memory-related processing in the hippocampus, amygdale, and prefrontal cortex. The initiation of a strong emotional experience activates memory-related neuroplasticity in the hippocampus and amygdala, and suppresses PFC functioning (phase 1). The most rapid actions would involve increases in ACTH, CRF, NE, acetylcholine, dopamine, and changes in GABA receptor binding (phase 1A), followed within minutes by elevated levels of glucocorticoids (phase 1B). The combination of the activation of the hippocampus by these neuromodulators with coincident tetanizing stimulation produces a great enhancement of LTP. Within minutes of the initiation of phase 1, the hippocampus undergoes a reversal of its plasticity state, based, in part, on the reduction in the sensitivity of NMDA receptors (phase 2). Tetanizing stimulation delivered to the hippocampus during phase 2 will thereby result in an impairment of the induction of LTP. The amygdala continues in its form of phase 1 longer than the hippocampus, but eventually, the amygdala, as well, exhibits an inhibitory phase, perhaps as it is involved in the consolidation of the emotional memory. The PFC is only inhibited by stress; the recovery from its suppression of functioning would depend on the nature and intensity of the stressor, interacting with the ability of the individual to cope with the experience. In the case of trauma-induced PTSD, the PFC may not recover to its original state of efficiency in suppressing the activity of lower brain areas, such as the amygdala and brain stem nuclei.

Figure 4

Brief stress immediately before training enhances, and prolonged stress impairs, hippocampus-dependent memory. (a) illustrates the influence of 2 minutes of predator exposure on spatial memory. Rats were exposed to a cat for 2 minutes and then they were given minimal radial arm water maze training (4 trials to find the hidden platform) either immediately or 30 minutes later. Rats trained under nonstress conditions or with cat exposure 30 minutes before training showed no evidence of memory for the platform location 24 hours later (open circle and open triangle). In contrast, rats trained immediately after brief exposure to a cat showed strong 24-hour memory (filled square). The dashed line at 2.5 errors indicates chance level of performance. (b) illustrates the effects of brief versus prolonged water immersion on contextual and cued fear conditioning. Rats given brief water stress either immediately (open bar) or 8 minutes (diagonal lines) before fear conditioning exhibited intact contextual and cued fear memory which was equivalent to that found in the no-stress group (gray bar). Rats given repeated pretraining water immersions (solid bar), by contrast, exhibited intact cued fear memory, but had a complete absence of contextual fear memory. “Precue” indicates baseline freezing in the nonshock context (3-minute duration) prior to the delivery of the tone (3-minute duration). Prolonged pretraining stress, therefore, completely suppressed contextual (hippocampus-dependent) fear conditioning without having any effect on cued (amygdala-dependent) fear conditioning. In both graphs, ∗ = P < .05 (ANOVA and Holm-Sidak post-hoc test) compared to the no-stress group.

Figure 4

Brief stress immediately before training enhances, and prolonged stress impairs, hippocampus-dependent memory. (a) illustrates the influence of 2 minutes of predator exposure on spatial memory. Rats were exposed to a cat for 2 minutes and then they were given minimal radial arm water maze training (4 trials to find the hidden platform) either immediately or 30 minutes later. Rats trained under nonstress conditions or with cat exposure 30 minutes before training showed no evidence of memory for the platform location 24 hours later (open circle and open triangle). In contrast, rats trained immediately after brief exposure to a cat showed strong 24-hour memory (filled square). The dashed line at 2.5 errors indicates chance level of performance. (b) illustrates the effects of brief versus prolonged water immersion on contextual and cued fear conditioning. Rats given brief water stress either immediately (open bar) or 8 minutes (diagonal lines) before fear conditioning exhibited intact contextual and cued fear memory which was equivalent to that found in the no-stress group (gray bar). Rats given repeated pretraining water immersions (solid bar), by contrast, exhibited intact cued fear memory, but had a complete absence of contextual fear memory. “Precue” indicates baseline freezing in the nonshock context (3-minute duration) prior to the delivery of the tone (3-minute duration). Prolonged pretraining stress, therefore, completely suppressed contextual (hippocampus-dependent) fear conditioning without having any effect on cued (amygdala-dependent) fear conditioning. In both graphs, ∗ = P < .05 (ANOVA and Holm-Sidak post-hoc test) compared to the no-stress group.