How the amygdala affects emotional memory by altering brain network properties

Abstract

The amygdala has long been known to play a key role in supporting memory for emotionally arousing experiences. For example, classical fear conditioning depends on neural plasticity within this anterior medial temporal lobe region. Beneficial effects of emotional arousal on memory, however, are not restricted to simple associative learning. Our recollection of emotional experiences often includes rich representations of, e.g., spatiotemporal context, visceral states, and stimulus–response associations. Critically, such memory features are known to bear heavily on regions elsewhere in the brain. These observations led to the modulation account of amygdala function, which postulates that amygdala activation enhances memory consolidation by facilitating neural plasticity and information storage processes in its target regions. Rodent work in past decades has identified the most important brain regions and neurochemical processes involved in these modulatory actions, and neuropsychological and neuroimaging work in humans has produced a large body of convergent data. Importantly, recent methodological developments make it increasingly realistic to monitor neural interactions underlying such modulatory effects as they unfold. For instance, functional connectivity network modeling in humans has demonstrated how information exchanges between the amygdala and specific target regions occur within the context of large-scale neural network interactions. Furthermore, electrophysiological and optogenetic techniques in rodents are beginning to make it possible to quantify and even manipulate such interactions with millisecond precision. In this paper we will discuss that these developments will likely lead to an updated view of the amygdala as a critical nexus within large-scale networks supporting different aspects of memory processing for emotionally arousing experiences.

Introduction

Stressful and emotionally arousing experiences are preferentially retained in memory (Joëls et al., 2011, Schacter, 1999). It has long been known that the amygdala, an anterior medial temporal lobe structure, plays a pivotal role in this usually highly adaptive phenomenon. The notion that the amygdala is involved in affective processing dates back to the classic report by Klüver and Bucy (1937) on the effects of temporal lobectomy in rhesus monkeys. Bilateral lesions of this region were shown to result in striking behavioral changes, including visual agnosia, dietary changes, and hypersexuality, but also profound alterations in affective behaviors, including tameness and loss of fear. Further investigations into these effects by Weiskrantz in the 1950s demonstrated that these affective changes were largely due to anteromedial temporal lobe lesions, particularly to the amygdala (Weiskrantz, 1956). Weiskrantz’ work also demonstrated that amygdala lesions not only block the expression of conditioned fear, but also impair new fear learning. This observation set the stage for a line of research into the role of the amygdala in emotional memory that now spans multiple decades (McGaugh and Roozendaal, 2002, Phelps and LeDoux, 2005, Roozendaal et al., 2009). Initially, this research focused on the amygdala proper as a storage site for associations underlying fear memory (Davis and Whalen, 2000, LeDoux, 2000). For instance, it was shown that fear learning depends on the induction of neural plasticity within the basolateral complex of the amygdala (BLA) (Miserendino et al., 1990, Rogan et al., 1997). These findings led to the view that the BLA might be a crucial site where sensory input converges and synaptic plasticity can produce lasting changes in emotional responses to environmental stimuli.

However, the beneficial effects of emotional arousal on memory extend well beyond associative fear learning. Recollection of emotional experiences typically includes not only representations of sensory cues, but, for example, also of spatial and temporal context in which these have been encountered (Christianson, 1992, Diamond et al., 2007, Phelps, 2004, Sandi and Pinelo-Nava, 2007). Importantly, such types of declarative memory rely heavily on neural systems elsewhere in the brain (Bird and Burgess, 2008, Henke, 2010). This notion has been corroborated by an extensive body of evidence documenting beneficial effects of stress hormones and neurotransmitters released during emotionally arousing experiences on various types of memory, both in rodents (McGaugh, 1989) and in humans (Lupien et al., 2007, Schwabe et al., 2012). These developments led to the proposal that the amygdala contributes to enhancement of memory for emotional events primarily by integrating these neuromodulatory influences and modulating mnemonic activity and synaptic plasticity in other brain regions (McGaugh and Roozendaal, 2002, Roozendaal et al., 2009). Research into the role of the amygdala in influencing memory consolidation processes in other memory systems was pioneered by James L. McGaugh and colleagues. However, this modulation hypothesis of amygdala function can historically be traced back to Ralph W. Gerard, who hypothesized already many years before the first experiments were performed that amygdalar nuclei could “modify the ease and completeness of experience fixation even if the nuclei were not themselves the loci of engrams” (Gerard, 1961).

As we will address below, the modulation hypothesis has received wide empirical support over the years. Rodent studies have utilized pharmacological manipulations, selective lesions and immediate-early gene activation to delineate the relevant structures, pathways and neurochemical processes. This work has dovetailed tightly with behavioral, psychopharmacological, neuropsychological, and neuroimaging studies in humans. However, more recent methodological developments make it increasingly feasible to monitor neural activity in real-time, and thus to explore how information is processed and exchanged between brain regions. For instance, the rapid proliferation of techniques for functional connectivity network modeling in humans using functional magnetic resonance imaging (fMRI) has made it possible to study interactions within large-scale neural systems in the human brain (Raichle, 2009). Findings gathered using these techniques have generated novel insights into how brain regions implicated in different types of memory are part of distinct large-scale connectivity networks (Ranganath & Ritchey, 2012). As explained below, these findings yield important heuristics for translation back to basic neuroscience, which offers the technology to investigate these processes more mechanistically and in more spatial and temporal detail. For instance, in vivo electrophysiological techniques in rodents now make it possible to simultaneously record activity of neuronal ensembles distributed across different brain regions during different phases of memory processing (Buzsáki, 2004), while optogenetic techniques allow manipulations of specific neural connections with millisecond precision (Boyden et al., 2005, Tye et al., 2011). Future application of these techniques to the amygdala and its many efferent connections will lead to a sophisticated understanding of how the amygdala engages stress hormone and neurotransmitter systems to modulate large-scale network properties and influence distant neural processes underlying the formation and consolidation of memory for emotionally arousing experiences.