Conjunctive representations,the hippocampus,and contextual fear conditioning

The context in which events occur can be represented as both (1) a set of independent features, the feature representation view, and (2) a set of features bound into a unitary representation, the conjunction representation view. It is assumed that extrahippocampal (e.g., neocortical) areas provide a basis for feature representations, but the hippocampal formation makes an essential contribution to the automatic storage of conjunctive representations. We develop this dual-representation view and explore its implications for hippocampal contributions to contextual fear conditioning processes. To this end, we discuss how our framework can resolve some of the conflicts in the recent literature relating the hippocampus to contextual fear conditioning. We also present new data supporting the role of a key mechanism afforded by conjunctive representations—pattern completion (the ability of a subset of a memory pattern to activate the complete memory)—in contextual fear conditioning. As is implied by this mechanism, we report that fear can be conditioned to the memory representation of a context that is not actually present at the time of shock. Moreover, this result is predicted by our computational model of cortical and hippocampal function. We suggest that pattern completion demonstrated in animals and by our model provides a mechanistic bridge to human declarative memory.     

Integrating incremental learning and episodic memory models of the hippocampal region

By integrating previous computational models of corticohippocampal function, the authors develop and test a unified theory of the neural substrates of familiarity, recollection, and classical conditioning. This approach integrates models from 2 traditions of hippocampal modeling, those of episodic memory and incremental learning, by drawing on an earlier mathematical model of conditioning, SOP (A. Wagner, 1981). The model describes how a familiarity signal may arise from parahippocampal cortices, giving a novel explanation for the finding that the neural response to a stimulus in these regions decreases with increasing stimulus familiarity. Recollection is ascribed to the hippocampus proper. It is shown how the properties of episodic representations in the neocortex, parahippocampal gyrus, and hippocampus proper may explain phenomena in classical conditioning. The model reproduces the effects of hippocampal, septal, and broad hippocampal region lesions on contextual modulation of classical conditioning, blocking, learned irrelevance, and latent inhibition.     

Aversive experiences are associated with a rapid and transient activation of ERKs in the rat hippocampus.

The extracellular signal-regulated kinases (ERKs) family of mitogen-activated protein kinases (MAPKs) has been shown to participate in memory formation. We recently found that a hippocampal ERK/MAPK cascade is required for memory formation of an inhibitory avoidance training in rats. Here we reported that this learning task is accompanied by a rapid increase in the phosphorylation of hippocampal p44 MAPK. A single mild foot shock produced a similar effect and three consecutive foot shocks provoked the activation of both p44 and p42 MAPKs. In contrast, a brief exposure to the training box or the habituation to an open field did not alter hippocampal ERK/MAPK levels. Together, these findings indicate that aversive behavioral experiences induced a rapid and transient activation of ERK/MAPKs in the hippocampus.     

Models in search of a brain

Mental localization efforts tend to stress the where more than the what. We argue that the proper targets for localization are well-specified cognitive models. We make this case by relating an existing cognitive model of category learning to a learning circuit involving the hippocampus, perirhinal, and prefrontal cortices. Results from groups varying in function along this circuit (e.g., infants, amnesics, and older adults) are successfully simulated by reducing the model’s ability to form new clusters in response to surprising events, such as an error in supervised learning or an unfamiliar stimulus in unsupervised learning. Clusters in the model are akin to conjunctive codes that are rooted in an episodic experience (the surprising event) yet can develop to resemble abstract codes as they are updated by subsequent experiences. Thus, the model holds that the line separating episodic and semantic information can become blurred. Dissociations (categorization vs. recognition) are explained in terms of cluster recruitment demands.     

Ontogeny and neural substrates of the context preexposure facilitation effect

Contextual fear conditioning emerges around post-natal day (PD) 23 in the rat. This is thought to reflect hippocampus-dependent conjunctive learning, which binds the individual features of the context into a unified representation (Rudy, 1993). However, context conditioning can also be supported by hippocampus-independent, feature-based simple associations (Rudy, 2009) and these may operate at PD 23-24 (Pugh & Rudy, 1996). To address this issue, we studied the ontogeny of a variant of contextual fear conditioning, termed the context-preexposure-facilitation-effect (CPFE), in which exposure to context and (immediate) foot shock occur on successive occasions. This variant requires conjunctive as opposed to feature-based simple associations (Rudy, 2009). We tested PD 17, 24, and 31 rats on the CPFE vs. conventional context conditioning (Exp. 1) and on the CPFE with stronger reinforcement (Exp. 2). The CPFE emerged on PD 24 regardless of reinforcer strength and in parallel with context conditioning. Infusions of the NMDA-receptor antagonist, MK-801, into the dorsal hippocampus just before pre-exposure on PD 24 eliminated the CPFE, whereas infusions occurring after pre-exposure had no effect (Exp. 3). These findings demonstrate a role of hippocampal NMDA receptors in the CPFE as early as PD 24 and implicate conjunctive learning mechanisms in the ontogeny of contextual fear conditioning.     

Unpacking the cognitive map: the parallel map theory of hippocampal function

In the parallel map theory, the hippocampus encodes space with 2 mapping systems. The bearing map is constructed primarily in the dentate gyrus from directional cues such as stimulus gradients. The sketch map is constructed within the hippocampus proper from positional cues. The integrated map emerges when data from the bearing and sketch maps are combined. Because the component maps work in parallel, the impairment of one can reveal residual learning by the other. Such parallel function may explain paradoxes of spatial learning, such as learning after partial hippocampal lesions, taxonomic and sex differences in spatial learning, and the function of hippocampal neurogenesis. By integrating evidence from physiology to phylogeny, the parallel map theory offers a unified explanation for hippocampal function.     

Cellular and molecular mechanisms underlying learning and memory impairments produced by cannabinoids

Why does smoking marijuana impair learning and memory? Behavioral studies suggest that a disruption of normal hippocampal function contributes to these deficits. In vitro experiments find that cannabinoid receptor activation reduces neurotransmitter release below the levels required to trigger long-term changes in synaptic strength in the hippocampus. Cannabinoids reduce glutamate release through a G-protein-mediated inhibition of the calcium channels responsible for neurotransmitter release from hippocampal neurons. These mechanisms likely play a role in the learning and memory impairments produced by cannabinoids and by endogenous cannabinoid receptor ligands.     

Inactivation of the anterior cingulate cortex impairs extinction of rabbit jaw movement conditioning and prevents extinction-related inhibition of hippocampal activity

Although past research has highlighted the involvement of limbic structures such as the anterior cingulate cortex (ACC) and hippocampus in learning, few have addressed the nature of their interaction. The current study of rabbit jaw movement conditioning used a combination of reversible lesions and electrophysiology to examine the involvement of the hippocampus and the ACC during acquisition, performance, and extinction. We found that microinfusions of procaine into the ACC did not significantly alter the rate of behavioral learning or the amplitude of hippocampal conditioned unit responses, but that they disrupted the rhythmic periodicity of conditioned jaw movements. During extinction, whereas controls showed a rapid decline in behavioral CRs and active inhibition of hippocampal unit responses, ACC lesioned rabbits showed a persistence of conditioning-related hippocampal activity and behavioral responding. The results show that the ACC can be important for adaptive suppression of conditioned behavior and suggest a crucial physiological modulation of hippocampus by ACC during extinction.     

Neural representations of location outside the hippocampus

Place cells of the rat hippocampus are a dominant model system for understanding the role of the hippocampus in learning and memory at the level of single-unit and neural ensemble responses. A complete understanding of the information processing and computations performed by the hippocampus requires detailed knowledge about the properties of the representations that are present in hippocampal afferents and efferents in order to decipher the transformations that occur to these representations in the hippocampal circuitry. Neural recordings in behaving rats have revealed a number of brain areas that contain place-related firing properties in the parahippocampal regions and in other brain regions that are thought to interact with the hippocampus in certain behavioral tasks. Although investigators have just begun to scratch the surface in terms of understanding these properties, differences in the precise nature of the spatial firing between the hippocampus and these other regions promise to reveal important clues regarding the exact role of the hippocampus in learning and memory and the nature of its interactions with other brain systems to support adaptive behavior.     

Generalization through the recurrent interaction of episodic memories: a model of the hippocampal system

In this article, we present a perspective on the role of the hippocampal system in generalization, instantiated in a computational model called REMERGE (recurrency and episodic memory results in generalization). We expose a fundamental, but neglected, tension between prevailing computational theories that emphasize the function of the hippocampus in pattern separation (Marr, 1971; McClelland, McNaughton, & O'Reilly, 1995), and empirical support for its role in generalization and flexible relational memory (Cohen & Eichenbaum, 1993; Eichenbaum, 1999). Our account provides a means by which to resolve this conflict, by demonstrating that the basic representational scheme envisioned by complementary learning systems theory (McClelland et al., 1995), which relies upon orthogonalized codes in the hippocampus, is compatible with efficient generalization-as long as there is recurrence rather than unidirectional flow within the hippocampal circuit or, more widely, between the hippocampus and neocortex. We propose that recurrent similarity computation, a process that facilitates the discovery of higher-order relationships between a set of related experiences, expands the scope of classical exemplar-based models of memory (e.g., Nosofsky, 1984) and allows the hippocampus to support generalization through interactions that unfold within a dynamically created memory space.     

The perceptual-mnemonic/feature conjunction model of perirhinal cortex function.

The perirhinal cortex was once thought to be "silent cortex", virtually ignored by researchers interested in the neurobiology of learning and memory. Following studies of brain damage associated with cases of amnesia, perirhinal cortex is now widely regarded as part of a "medial temporal lobe (MTL) memory system". This system is thought to be more or less functionally homogeneous, having a special role in declarative memory, and making little or no contribution to other functions such as perception. In the present article, we summarize an alternative view. First, we propose that components of the putative MTL system such as the hippocampus and perirhinal cortex have distinct and dissociable functions. Second, we provide evidence that the perirhinal cortex has a role in visual discrimination. In addition, we propose a specific role for perirhinal cortex in visual discrimination: the contribution of complex conjunctive representations to the solution of visual discrimination problems with a high degree of "feature ambiguity". These proposals constitute a new view of perirhinal cortex function, one that does not assume strict modularity of function in the occipito-temporal visual stream, but replaces this idea with the notion of a hierarchical representational continuum.     

Short-term estrogen treatment in ovariectomized rats augments hippocampal acetylcholine release during place learning

Estrogen modulates learning and memory in ovariectomized and naturally cycling female rats, especially in tasks using spatial learning and navigation. Estrogen also modulates cholinergic function in various forebrain structures. Past studies have shown positive correlations between hippocampal ACh output and performance on hippocampus-dependent tasks. The present study examined whether estradiol replacement would potentiate hippocampal ACh release during place learning. In vivo microdialysis and HPLC were used to measure extracellular ACh levels in the hippocampus of ovariectomized female rats that had received s.c. injections of 17beta-estradiol (10 microg) or sesame oil (vehicle treatment) 48 and 24h prior to training on a place task. Estrogen did not alter baseline levels of extracellular ACh in the hippocampus. During training, hippocampal ACh increased in ovariectomized rats regardless of estrogen status. However, while estradiol did not enhance learning in this experiment, estradiol significantly potentiated the increase in hippocampal ACh release seen during place training. This represents the first demonstration of on-line assessment of ACh output in hippocampus during learning in female rats and suggests that estrogen-dependent modulation of ACh release during training might control activation of different neural systems used during learning.     

Testing a neurocomputational model of recollection, familiarity, and source recognition

The authors assess whether the complementary learning systems model of the medial temporal lobes (Norman & O'Reilly, 2003) is able to account for source recognition receiver operating characteristics (ROCs). The model assumes that recognition reflects the contribution of a hippocampally mediated recollection process and a cortically mediated familiarity process. The hippocampal process is found to produce threshold output functions that lead to U-shaped zROCs, whereas the cortical process produces Gaussian signal detection functions and linear zROCs. The model is consistent with several dual process theories of recognition and is capable of producing the types of zROCs observed in studies of item and source recognition. In addition, the model makes the novel prediction that as the level of feature similarity across items increases, the ability of the hippocampus to encode distinct representations for each stimulus will diminish, and the threshold nature of recollection will break down, leading source zROCs to become more linear. The authors conducted 3 new behavioral source experiments that confirmed the model's prediction. The results demonstrate that the model provides a viable account of item and source recognition performance.     

Towards learning agents with personality traits: Modeling Openness to Experience

Recent advances in neurosciences and cognitive sciences show us that the human neocortex is not a slave to the experiences from our perception and that the memories stored in hippocampus are goal weighted during the replay of the experiences for the purpose of re-learning from them. Temporal difference reinforcement learning systems that use neural networks as function approximators rely on an experience replay memory structure similar to the hippocampus. We bring forward this similarity and present a novel way of using a goal weighted prioritization of the memory that is biologically inspired. Furthermore, we introduce a novel prioritization criteria called Variety of Experience Index, or VEI, for weighting the selection of the experiences that are stored in the replay memory. Weighting the experiences based on two different extremes of VEI can behaviourally modify the agent’s learning process, generating different types of learning agents that exhibit different personality traits along the dimension of Openness to Experience.     

The perceptual-mnemonic/feature conjunction model of perirhinal cortex function

The perirhinal cortex was once thought to be “silent cortex”, virtually ignored by researchers interested in the neurobiology of learning and memory. Following studies of brain damage associated with cases of amnesia, perirhinal cortex is now widely regarded as part of a “medial temporal lobe (MTL) memory system”. This system is thought to be more or less functionally homogeneous, having a special role in declarative memory, and making little or no contribution to other functions such as perception. In the present article, we summarize an alternative view. First, we propose that components of the putative MTL system such as the hippocampus and perirhinal cortex have distinct and dissociable functions. Second, we provide evidence that the perirhinal cortex has a role in visual discrimination. In addition, we propose a specific role for perirhinal cortex in visual discrimination: the contribution of complex conjunctive representations to the solution of visual discrimination problems with a high degree of “feature ambiguity”. These proposals constitute a new view of perirhinal cortex function, one that does not assume strict modularity of function in the occipito-temporal visual stream, but replaces this idea with the notion of a hierarchical representational continuum.     

Parallel Neural Systems for Classical Conditioning: Support From Computational Modeling

Classical conditioning has been explained by two main types of theories that postulate different learning mechanisms. Rescorla and Wagner (1972) put forth a theory in which conditioning is based on the ability of the US to drive learning through error correction. Alternatively, Mackintosh (1973) put forth a theory in which the ability of the CS to be associated with the unconditioned stimulus is modulated. We have proposed a reconciliation of these two mechanisms as working in parallel within different neural systems: a cerebellar system for US modulation and a hippocampal system for CS modulation. We developed a computational model of cerebellar function in eyeblink conditioning based on the error correction mechanism of the Rescorla-Wagner rule in which learning-related activity from the cerebellum inhibits the inferior olive, which is the US input pathway to the cerebellum (Gluck et al., 1994). We developed a computational model of the hippocampal region that forms altered representations of conditioned stimuli based on their behavioral outcomes (Gluck & Myers, 1993; Myers et al., 1995). Overall, computational modeling and empirical findings support the idea that, at least in the case of eyeblink conditioning, there may be two different neural systems: the cerebellum which mediates US-based error correction and hippocampus which alters representations of CSs.     

Parallel neural systems for classical conditioning: Support from computational modeling

Classical conditioning has been explained by two main types of theories that postulate different learning mechanisms. Rescorla and Wagner (1972) put forth a theory in which conditioning is based on the ability of the US to drive learning through error correction. Alternatively, Mackintosh (1973) put forth a theory in which the ability of the CS to be associated with the unconditioned stimulus is modulated. We have proposed a reconciliation of these two mechanisms as working in parallel within different neural systems: a cerebellar system for US modulation and a hippocampal system for CS modulation. We developed a computational model of cerebellar function in eyeblink conditioning based on the error correction mechanism of the Rescorla-Wagner rule in which learningrelated activity from the cerebellum inhibits the inferior olive, which is the US input pathway to the cerebellum (Gluck et al., 1994). We developed a computational model of the hippocampal region that forms altered representations of conditioned stimuli based on their behavioral outcomes (Gluck & Myers, 1993; Myers et al., 1995). Overall, computational modeling and empirical findings support the idea that, at least in the case of eyeblink conditioning, there may be two different neural systems: the cerebellum which mediates US-based error correction and hippocampus which alters representations of CSs.     

Neurogenesis in adulthood: a possible role in learning

The role of the hippocampal formation in learning and memory has long been recognized. However, despite decades of intensive research, the neurobiological basis of this process in the hippocampus remains enigmatic. Over 30 years ago, the production of new neurons was found to occur in the brains of adult rodents. More recently, the documentation of adult neurogenesis in the hippocampal formation of a variety of mammals, including humans, has suggested a novel approach towards understanding the biological bases of hippocampal function. Contemporary theories of hippocampal function include an important role for this brain region in associative learning. The addition of new neurons and consequently, their novel contribution to hippocampal circuitry could conceivably be a mechanism for relating spatially or temporally disparate events. In this review, we examine several lines of evidence suggesting that adult-generated neurons are involved in hippocampal-dependent learning. In particular, we examine the variables that modulate hippocampal neurogenesis in adulthood and their relation to learning and memory.     

Encoding multielement scenes: statistical learning of visual feature hierarchies

The authors investigated how human adults encode and remember parts of multielement scenes composed of recursively embedded visual shape combinations. The authors found that shape combinations that are parts of larger configurations are less well remembered than shape combinations of the same kind that are not embedded. Combined with basic mechanisms of statistical learning, this embeddedness constraint enables the development of complex new features for acquiring internal representations efficiently without being computationally intractable. The resulting representations also encode parts and wholes by chunking the visual input into components according to the statistical coherence of their constituents. These results suggest that a bootstrapping approach of constrained statistical learning offers a unified framework for investigating the formation of different internal representations in pattern and scene perception.     

Computational models of the hippocampal region: linking incremental learning and episodic memory

The hippocampal region, a group of brain structures important for learning and memory, has been the focus of a large number of computational models. These tend to fall into two groups: (1) models of the role of the hippocampal region in incremental learning, which focus on the development of new representations that are sensitive to stimulus regularities and environmental context; (2) models that focus on the role of the hippocampal region in the rapid storage and retrieval of episodic memories. Rather than being in conflict, it is becoming apparent that both approaches are partially correct and might reflect the different functions of substructures of the hippocampal region. Future computational models will help to elaborate how these different substructures interact.