An attractor network in the hippocampus: theory and neurophysiology

A quantitative computational theory of the operation of the CA3 system as an attractor or autoassociation network is described. Based on the proposal that CA3-CA3 autoassociative networks are important for episodic or event memory in which space is a component (place in rodents and spatial view in primates), it has been shown behaviorally that the CA3 supports spatial rapid one-trial learning and learning of arbitrary associations and pattern completion where space is a component. Consistent with the theory, single neurons in the primate CA3 respond to combinations of spatial view and object, and spatial view and reward. Furthermore, single CA3 neurons reflect the recall of a place from an object in a one-trial object-place event memory task. CA3 neurons also reflect in their firing a memory of spatial view that is retained and updated by idiothetic information to implement path integration when the spatial view is obscured. Based on the computational proposal that the dentate gyrus produces sparse representations by competitive learning and via the mossy fiber pathway forces new representations on the CA3 during learning (encoding), it has been shown behaviorally that the dentate gyrus supports spatial pattern separation during learning, and that the mossy fiber system to CA3 connections are involved in learning but not in recall. The perforant path input to CA3 is quantitatively appropriate to provide the cue for recall in CA3. The concept that the CA1 recodes information from CA3 and sets up associatively learned back-projections to neocortex to allow subsequent retrieval of information to neocortex provides a quantitative account of the large number of hippocampo-neocortical back-projections.     

Pattern separation, pattern completion, and new neuronal codes within a continuous CA3 map

The hippocampal CA3 subregion is critical for rapidly encoding new memories, which suggests that neuronal computations are implemented in its circuitry that cannot be performed elsewhere in the hippocampus or in the neocortex. Recording studies show that CA3 cells are bound to a large degree to a spatial coordinate system, while CA1 cells can become more independent of a map-based mechanism and allow for a larger degree of arbitrary associations, also in the temporal domain. The mapping of CA3 onto a spatial coordinate system intuitively points to its role in spatial navigation but does not directly suggest how such a mechanism may support memory processing. Although bound to spatial coordinates, the CA3 network can rapidly alter its firing rate in response to novel sensory inputs and is thus not as strictly tied to spatial mapping as grid cells in the medial entorhinal cortex. Such rate coding within an otherwise stable spatial map can immediately incorporate new sensory inputs into the two-dimensional matrix of CA3, where they can be integrated with already stored information about each place. CA3 cell ensembles may thus support the fast acquisition of detailed memories by providing a locally continuous, but globally orthogonal representation, which can rapidly provide a new neuronal index when information is encountered for the first time. This information can be interpreted in CA1 and other downstream cortical areas in the context of less spatially restricted information.     

Visual mental image generation does not overlap with visual short-term memory: A dual-task interference study

Visual mental imagery and working memory are often assumed to play similar roles in high-order functions, but little is known of their functional relationship. In this study, we investigated whether similar cognitive processes are involved in the generation of visual mental images, in short-term retention of those mental images, and in short-term retention of visual information. Participants encoded and recalled visually or aurally presented sequences of letters under two interference conditions: spatial tapping or irrelevant visual input (IVI). In Experiment 1, spatial tapping selectively interfered with the retention of sequences of letters when participants generated visual mental images from aural presentation of the letter names and when the letters were presented visually. In Experiment 2, encoding of the sequences was disrupted by both interference tasks. However, in Experiment 3, IVI interfered with the generation of the mental images, but not with their retention, whereas spatial tapping was more disruptive during retention than during encoding. Results suggest that the temporary retention of visual mental images and of visual information may be supported by the same visual short-term memory store but that this store is not involved in image generation.     

Cognitive performances and locomotor activity following dentate granule cell damage in rats: role of lesion extent and type of memory tested

Intradentate injection of colchicine is one of the techniques used to destroy granule cells. This study compared the behavioral effects of various amounts of colchicine (1.0, 3.0, and 6.0 microg; Col 1, Col 3, and Col 6, respectively) injected into the dentate gyrus of adult Long-Evans male rats. Starting 10 days after lesion surgery, behavioral testing assessed home-cage and open-field locomotion, alternation in a T-maze, water-maze, and radial-maze learning according to protocols placing emphasis on reference, and working memory. All of these tasks are sensitive to hippocampal disruption. Histological verifications showed that the extent of the lesions depends on the dose of colchicine (index of dentate gyrus shrinkage: -33% in Col 1, -54% in Col 3, and -67% in Col 6 rats). Colchicine dose-dependently increased nocturnal home cage activity (an effect found 10 days but not 5 months after surgery), but had no significant effect on open-field locomotion or T-maze alternation. A dose-dependent reference memory impairment was found during the acquisition of spatial navigation in the water maze; Col 3 and Col 6 rats were more impaired than Col 1 rats. During the probe trial (platform removed), control rats spent a longer distance swimming over the platform area than all rats with colchicine lesions. In the working memory version of the test, all rats with colchicine lesions showed significant deficits. The deficits were larger in Col 3 and Col 6 rats compared to Col 1 rats. The lesions had no effect on swimming speed. In the radial-maze test, there was also a dose-dependent working memory impairment. However, reference memory was disrupted in a manner that did not differ among the three groups of lesioned rats. Our data are in line with the view that the dentate gyrus plays an important role in the acquisition of new information and is an integral neural substrate for spatial reference and spatial working memory. They also suggest that damage to granule cells might have more pronounced effects on reference than on working memory in the radial maze. Finally, they demonstrate that part of the variability in the conclusions from previous experiments concerning the role of granule cells in cognitive processes, particularly in spatial learning and memory, may be due to the type of tests used and/or the extent of the damage produced.     

NMDA signaling in CA1 mediates selectively the spatial component of episodic memory

Recent studies focusing on the memory for temporal order have reported that CA1 plays a critical role in the memory for the sequences of events, in addition to its well-described role in spatial navigation. In contrast, CA3 was found to principally contribute to the memory for the association of items with spatial or contextual information in tasks focusing on spatial memory. Other studies have shown that NMDA signaling in the hippocampus is critical to memory performance in studies that have investigated spatial and temporal order memory independently. However, the role of NMDA signaling separately in CA1 and CA3 in memory that combines both spatial and temporal processing demands (episodic memory) has not been examined. Here we investigated the effect of the deletion of the NR1 subunit of the NMDA receptor in CA1 or CA3 on the spatial and the temporal aspects of episodic memory, using a behavioral task that allows for these two aspects of memory to be evaluated distinctly within the same task. Under these conditions, NMDA signaling in CA1 specifically contributes to the spatial aspect of memory function and is not required to support the memory for temporal order of events.     

Selective lesions of the dentate gyrus produce disruptions in place learning for adjacent spatial locations

The hippocampus (HPP) plays a known role in learning novel spatial information. More specifically, the dentate gyrus (DG) hippocampal subregion is thought to support pattern separation, a mechanism for encoding and separating spatially similar events into distinct representations. Several studies have shown that lesions of the dorsal DG (dDG) in rodents result in inefficient spatial pattern separation for working memory; however, it is unclear whether selective dDG lesions disrupt spatial pattern separation for reference memory. Therefore, the current study investigated the role of the dDG in pattern separation using a spatial reference memory paradigm to determine whether the dDG is necessary for acquiring spatial discriminations for adjacent locations. Male Long-Evans rats were randomly assigned to receive bilateral intracranial infusions of colchicine or saline (control) into the dDG. Following recovery from surgery, each rat was pseudo-randomly assigned to an adjacent arm or separate arm condition and subsequently tested on a place-learning task using an eight-arm radial maze. Rats were trained to discriminate between a rewarded arm and a nonrewarded arm that were either adjacent to one another or separated by a distance of two arm positions. Each rat received 10 trials per day and was tested until the animal reached a criterion of nine correct choices out of 10 consecutive trials across 2 consecutive days of testing. Both groups acquired spatial discriminations for the separate condition at similar rates. However, in the adjacent condition, dDG lesioned animals required significantly more trials to reach the learning criterion than controls. The results suggest that dDG lesions decrease efficiency in pattern separation resulting in impairments in the adjacent condition involving greater overlap among the distal cues. Conversely, in the separate condition, there was less overlap among distal cues during encoding and less need for pattern separation. These findings provide further support for a critical role for the dDG in spatial pattern separation by demonstrating the importance of a processing mechanism that is capable of reducing interference among overlapping spatial inputs across a variety of memory demands.     

NT-3 facilitates hippocampal plasticity and learning and memory by regulating neurogenesis

In the adult brain, the expression of NT-3 is largely confined to the hippocampal dentate gyrus (DG), an area exhibiting significant neurogenesis. Using a conditional mutant line in which the NT-3 gene is deleted in the brain, we investigated the role of NT-3 in adult neurogenesis, hippocampal plasticity, and memory. Bromodeoxyuridine (BrdU)-labeling experiments demonstrated that differentiation, rather than proliferation, of the neuronal precursor cells (NPCs) was significantly impaired in DG lacking NT-3. Triple labeling for BrdU, the neuronal marker NeuN, and the glial marker GFAP indicated that NT-3 affects the number of newly differentiated neurons, but not glia, in DG. Field recordings revealed a selective impairment in long-term potentiation (LTP) in the lateral, but not medial perforant path-granule neuron synapses. In parallel, the NT-3 mutant mice exhibited deficits in spatial memory tasks. In addition to identifying a novel role for NT-3 in adult NPC differentiation in vivo, our study provides a potential link between neurogenesis, dentate LTP, and spatial memory.     

Disconnection analysis of CA3 and DG in mediating encoding but not retrieval in a spatial maze learning task

The dentate gyrus (DG) subregion of the hippocampus has been shown to be involved in encoding but not retrieval in a spatial maze task (modified Hebb-Williams maze). The first experiment in this study examined whether a lesion to the CA3 would contribute to a similar encoding deficit. A DG group was included in order to replicate previous results. Relative to controls, animals receiving CA3 lesions were impaired in encoding, not retrieval, on the modified Hebb-Williams maze--similar to a group that received DG lesions. This suggests the possibility that CA3 and DG are working together to mediate encoding processes. The second experiment in this study was designed to test the interaction between CA3 and DG using a disconnection paradigm. Animals with contralateral lesions (CA3 lesioned in one hemisphere, DG lesioned in the other hemisphere) showed a significant disruption effect on encoding, but not retrieval, when compared with animals with ipsilateral lesions (CA3 and DG lesioned in the same hemisphere, leaving the other hemisphere intact). This suggests an interaction between CA3 and DG in supporting encoding but not retrieval processes in a spatial maze learning task.     

Imaging recollection and familiarity in the medial temporal lobe: a three-component model

The medial temporal lobe (MTL) plays a crucial role in supporting memory for events, but the functional organization of regions in the MTL remains controversial, especially regarding the extent to which different subregions support recognition based on familiarity or recollection. Here we review results from functional neuroimaging studies showing that, whereas activity in the hippocampus and posterior parahippocampal gyrus is disproportionately associated with recollection, activity in the anterior parahippocampal gyrus is disproportionately associated with familiarity. The results are consistent with the idea that the parahippocampal cortex (located in the posterior parahippocampal gyrus) supports recollection by encoding and retrieving contextual information, whereas the hippocampus supports recollection by associating item and context information. By contrast, perirhinal cortex (located in the anterior parahippocampal gyrus) supports familiarity by encoding and retrieving specific item information. We discuss the implications of a 'binding of item and context' (BIC) model for studies of recognition memory. This model argues that there is no simple mapping between MTL regions and recollection and familiarity, but rather that the involvement of MTL regions in these processes depends on the specific demands of the task and the type of information involved. We highlight several predictions for future imaging studies that follow from the BIC model.     

Double dissociations in visual and spatial short-term memory

A visual short-term memory task was more strongly disrupted by visual than spatial interference, and a spatial memory task was simultaneously more strongly disrupted by spatial than visual interference. This double dissociation supports a fractionation of visuospatial short-term memory into separate visual and spatial components. In 6 experiments, this interpretation could be defended against alternative explanations in terms of trade-offs in resource allocation between memory tasks and interference tasks, in terms of an involvement of short-term consolidation and long-term memory, in terms of differential phonological-loop and central-executive involvement, and in terms of similarity-based interference.     

Preservation of long-term memory and synaptic plasticity despite short-term impairments in the Tc1 mouse model of Down syndrome

Down syndrome (DS) is a genetic disorder arising from the presence of a third copy of the human chromosome 21 (Hsa21). Recently, O’Doherty and colleagues in an earlier study generated a new genetic mouse model of DS (Tc1) that carries an almost complete Hsa21. Since DS is the most common genetic cause of mental retardation, we have undertaken a detailed analysis of cognitive function and synaptic plasticity in Tc1 mice. Here we show that Tc1 mice have impaired spatial working memory (WM) but spared long-term spatial reference memory (RM) in the Morris watermaze. Similarly, Tc1 mice are selectively impaired in short-term memory (STM) but have intact long-term memory (LTM) in the novel object recognition task. The pattern of impaired STM and normal LTM is paralleled by a corresponding phenotype in long-term potentiation (LTP). Freely-moving Tc1 mice exhibit reduced LTP 1 h after induction but normal maintenance over days in the dentate gyrus of the hippocampal formation. Biochemical analysis revealed a reduction in membrane surface expression of the AMPAR (α-amino-3-hydroxy-5-methyl-4-propionic acid receptor) subunit GluR1 in the hippocampus of Tc1 mice, suggesting a potential mechanism for the impairment in early LTP. Our observations also provide further evidence that STM and LTM for hippocampus-dependent tasks are subserved by parallel processing streams.     

Cannabinoid CB1 receptor deficiency increases contextual fear memory under highly aversive conditions and long-term potentiation in vivo

The cannabinoid receptor type 1 (CB1) is abundantly expressed in the central nervous system where it negatively controls the release of several neurotransmitters. CB1 activity plays a crucial role in learning and memory and in synaptic plasticity. In the present study, the role of CB1 was investigated in three different hippocampus-dependent memory tasks and in in vivo hippocampal synaptic plasticity in knockout (CB1-ko) and wildtype mice. There was no difference in short-term and long-term social and object recognition memory between CB1-ko and wildtype mice. In contrast, in background contextual fear conditioning CB1-ko mice showed enhanced freezing levels in the conditioning context and increased generalised contextual fear after a high-intensity conditioning foot shock of 1.5 mA, but not after 0.7 mA. In in vivo field potential recordings in the dentate gyrus, CB1-ko mice displayed a decreased paired-pulse facilitation of the populations spikes, suggesting an altered inhibitory synaptic drive onto hippocampal granule cells. Furthermore, CB1-ko mice displayed significantly higher levels of in vivo long-term potentiation (LTP) in the dentate gyrus. In conclusion, CB1 deficiency leads to enhanced contextual fear memory and altered synaptic plasticity in the hippocampus, supporting the key role of endocannabinoid signalling in learning and memory, in particular following highly aversive encounters.     

Neurotoxic dorsal CA1 lesions versus 4 VO ischaemic lesions: behavioural comparisons

Anterograde amnesia, a common consequence of transient cerebral ischaemia, has been attributed to cell loss in the hippocampal CA1 subfield. However, variable, widespread damage outside hippocampal CA1 can also occur following ischaemia. We compared the functional consequences of ischaemia and ibotenate acid CA1 lesions on 2 spatial memory tasks (water maze 'place' and 'matching-to-position') to address the possibility that extra-CA1 loss contributes to ischaemia-induced memory deficits in the rat. During place task acquisition, ischaemic rats showed deficits on more measures than ibotenic rats, and during a 1 min probe trial, only ischaemic rats were impaired. On the matching-to-position task, ibotenic rats showed greater impairment than ischaemic rats in terms of one-trial learning, whereas ischaemic rats were more impaired after Trial 2. Ischaemia and ibotenic acid lesions resulted in equivalent CA1 loss, but silver impregnation revealed additional extra-CA1 cell loss in ischaemic rats. Together with the greater behavioural deficits of ischaemic rats, these data indicate a role for extra-CA1 cell loss in ischaemia-induced memory impairments in both animals and humans.     

Automatic semantic encoding in verbal short-term memory: Evidence from the concreteness effect

The concreteness effect in verbal short-term memory (STM) tasks is assumed to be a consequence of semantic encoding in STM, with immediate recall of concrete words benefiting from richer semantic representations. We used the concreteness effect to test the hypothesis that semantic encoding in standard verbal STM tasks is a consequence of controlled, attention-demanding mechanisms of strategic semantic retrieval and encoding. Experiment 1 analysed the effect of presentation rate, with slow presentations being assumed to benefit strategic, time-dependent semantic encoding. Experiments 2 and 3 provided a more direct test of the strategic hypothesis by introducing three different concurrent attention-demanding tasks. Although Experiment 1 showed a larger concreteness effect with slow presentations, the following two experiments yielded strong evidence against the strategic hypothesis. Limiting available attention resources by concurrent tasks reduced global memory performance, but the concreteness effect was equivalent to that found in control conditions. We conclude that semantic effects in STM result from automatic semantic encoding and provide tentative explanations for the interaction between the concreteness effect and the presentation rate.     

Working memory task decreases the survival of newly born neurons in hippocampus

Throughout life new neurons are generated in dentate gyrus of hippocampus. Previous studies have found that spatial tasks can rescue newly born neurons from death. However, it is still unknown whether new neurons are similarly affected by all types of hippocampal-dependent tasks. Here we investigated the possible effects of working memory task (WMT) on immature neurons. Mice were trained in reference memory task and WMT respectively. The reference memory task used the classical hidden platform (HP) water maze task, while WMT used a delayed matching-to-place (DMTP) water maze task. Bromodeoxyuridine (BrdU) was administrated during the early or late phase of training, or 1week prior to training, in order to label dividing proliferating cells. After water maze training, the number of BrdU-labeled cells in dentate gyrus of hippocampus was compared. In addition, hippocampal brain-derived neurotrophic factor (BDNF) and Notch 1 receptor were characterized using Western blot. Serum corticosterone levels were also measured using enzyme immunoassay. Results showed that HP task and DMTP task did not change the number of BrdU-labeled cells produced during the early or late phase of training. As expected, the HP task increased the number of BrdU-labeled cells produced 1 week prior to training. However, DMTP task decreased the number of BrdU-labeled cells produced 1 week prior to training. Both tasks lead to a significant increase in serum corticosterone levels and did not change the expression of BDNF and Notch 1 receptor in hippocampus. Taken together, these results demonstrate that WMT has different effects on survival of immature neurons, and therefore suggests immature neurons may have more than one role depending on the demands of the tasks.     

催产素调控心理韧性:基于对海马的作用机制

心理韧性指个体面对逆境、挫折或重大威胁等应激情境下的有效且灵活适应的能力, 促进机体恢复正常的生理和心理功能。研究表明海马是调控心理韧性的重要脑区, 且催产素可能通过作用于海马增强心理韧性。海马内部环路内嗅皮层-齿状回-CA3可能调节恐惧记忆的泛化和消退以增强心理韧性; 海马外部环路齿状回-杏仁核-伏隔核及海马-伏隔核环路调节情绪, 可能分别通过促进奖赏和带来厌恶进而增强或降低心理韧性。催产素作用于海马增强心理韧性的可能途径有:催产素促进海马神经发生, 降低海马腹侧成熟神经元对应激的敏感性, 提高海马“模式分离”功能, 降低应激记忆泛化; 催产素恢复海马谢弗侧枝-CA1突触长时程增强, 促进机体适应应激; 催产素降低海马糖皮质激素受体水平, 重新建立机体稳态。     

Hippocampal gene expression profiling in spatial discrimination learning

Learning and long-term memory are thought to involve temporally defined changes in gene expression that lead to the strengthening of synaptic connections in selected brain regions. We used cDNA microarrays to study hippocampal gene expression in animals trained in a spatial discrimination-learning paradigm. Our analysis identified 19 genes that showed statistically significant changes in expression when comparing Nai;ve versus Trained animals. We confirmed the changes in expression for the genes encoding the nuclear protein prothymosin(alpha) and the delta-1 opioid receptor (DOR1) by Northern blotting or in situ hybridization. In additional studies, laser-capture microdissection (LCM) allowed us to obtain enriched neuronal populations from the dentate gyrus, CA1, and CA3 subregions of the hippocampus from Nai;ve, Pseudotrained, and spatially Trained animals. Real-time PCR examined the spatial learning specificity of hippocampal modulation of the genes encoding protein kinase B (PKB, also known as Akt), protein kinase C(delta) (PKC(delta)), cell adhesion kinase(beta) (CAK(beta), also known as Pyk2), and receptor protein tyrosine phosphatase(zeta/beta) (RPTP(zeta/beta)). These studies showed subregion specificity of spatial learning-induced changes in gene expression within the hippocampus, a feature that was particular to each gene studied. We suggest that statistically valid gene expression profiles generated with cDNA microarrays may provide important insights as to the cellular and molecular events subserving learning and memory processes in the brain.     

Differential Effects of Global Ischemia on Delayed Matching- and Non-Matching-to-Position Tasks in the Water Maze and Skinner Box

In order to assess effects of global ischemia in tasks of spatial learning and working memory, male Wistar rats were subjected to four vessel occlusion (4 VO) for periods of 5, 10, and 20 min and compared with sham-operated controls over four test phases, from 6 to 54 weeks after surgery. Rats were assessed on acquisition in the water maze, a task that is sensitive to ischemic impairments, before testing in Skinner box and water maze working memory tasks, which both require the short-term storage of information, but make different demands on spatial information processing. Phases 1 and 3 assessed spatial learning in a standard water maze procedure (12 and 10 training days, 2 trials/day with a 10-min intertrial interval: ITI). Phase 2 involved training and testing in delayed non-matching-to-position task in the Skinner box, with delays of 2–10 s between the information and choice stages. Phase 4 examined working memory in a water maze delayed matching-to-position task with 4 trials/day, an ITI of 30 s, and a novel platform position on each day. Ischemic rats showed duration-related impairments in water maze acquisition and working memory, but not in the less spatially demanding Skinner box task. Since water maze acquisition deficits were seen both before and after testing in the Skinner box the lack of effect cannot be attributed to time or to prior training. Ischemic deficits were more marked in Phase 3 than in Phase 1 of acquisition, suggesting that impairment may be progressive. Histological assessment showed that cell loss was largely confined to the hippocampal CA1 field and was linearly related to duration of occlusion. At the maximal level of loss (5.7 mm before the interaural line) the 20-min group showed 90% loss, the 10-min group 60% loss, and the 5-min group, which did not differ from controls, less than 10% loss. Only the 20-min group showed significant damage beyond the CA1 field, ranging from 30–40% loss in the CA3 field to 5% loss in one striatal area. No cortical damage was seen. The extent of CA1 cell loss correlated modestly with water maze acquisition (Phase 3) and working memory scores, but not with trials to criterion in the Skinner box task. There were significant correlations between different measures both within and between water maze tasks, but not Skinner box tasks, suggesting that the two types of procedure engaged different cognitive processes. The results indicate that the intrahippocampal damage induced by 4 VO impaired tasks which required processing of allocentric spatial information, but did not impair the storage of limited spatial information in working memory.     

Remembering the past and imagining the future: a neural model of spatial memory and imagery

The authors model the neural mechanisms underlying spatial cognition, integrating neuronal systems and behavioral data, and address the relationships between long-term memory, short-term memory, and imagery, and between egocentric and allocentric and visual and ideothetic representations. Long-term spatial memory is modeled as attractor dynamics within medial-temporal allocentric representations, and short-term memory is modeled as egocentric parietal representations driven by perception, retrieval, and imagery and modulated by directed attention. Both encoding and retrieval/imagery require translation between egocentric and allocentric representations, which are mediated by posterior parietal and retrosplenial areas and the use of head direction representations in Papez's circuit. Thus, the hippocampus effectively indexes information by real or imagined location, whereas Papez's circuit translates to imagery or from perception according to the direction of view. Modulation of this translation by motor efference allows spatial updating of representations, whereas prefrontal simulated motor efference allows mental exploration. The alternating temporal-parietal flows of information are organized by the theta rhythm. Simulations demonstrate the retrieval and updating of familiar spatial scenes, hemispatial neglect in memory, and the effects on hippocampal place cell firing of lesioned head direction representations and of conflicting visual and ideothetic inputs.