Showing papers in "Hippocampus in 1996"

Theta phase precession in hippocampal neuronal populations and the compression of temporal sequences.

Abstract: O'Keefe and Recce [1993] Hippocampus 3:317-330 described an interaction between the hippocampal theta rhythm and the spatial firing of pyramidal cells in the CA1 region of the rat hippocampus: they found that a cell's spike activity advances to earlier phases of the theta cycle as the rat passes through the cell's place field. The present study makes use of large-scale parallel recordings to clarify and extend this finding in several ways: 1) Most CA1 pyramidal cells show maximal activity at the same phase of the theta cycle. Although individual units exhibit deeper modulation, the depth of modulation of CA1 population activity is about 50%. The peak firing of inhibitory interneurons in CA1 occurs about 60 degrees in advance of the peak firing of pyramidal cells, but different interneurons vary widely in their peak phases. 2) The first spikes, as the rat enters a pyramidal cell's place field, come 90 degrees-120 degrees after the phase of maximal pyramidal cell population activity, near the phase where inhibition is least. 3) The phase advance is typically an accelerating, rather than linear, function of position within the place field. 4) These phenomena occur both on linear tracks and in two-dimensional environments where locomotion is not constrained to specific paths. 5) In two-dimensional environments, place-related firing is more spatially specific during the early part of the theta cycle than during the late part. This is also true, to a lesser extent, on a linear track. Thus, spatial selectivity waxes and wanes over the theta cycle. 6) Granule cells of the fascia dentata are also modulated by theta. The depth of modulation for the granule cell population approaches 100%, and the peak activity of the granule cell population comes about 90 degrees earlier in the theta cycle than the peak firing of CA1 pyramidal cells. 7) Granule cells, like pyramidal cells, show robust phase precession. 8) Cross-correlation analysis shows that portions of the temporal sequence of CA1 pyramidal cell place fields are replicated repeatedly within individual theta cycles, in highly compressed form. The compression ratio can be as much as 10:1. These findings indicate that phase precession is a very robust effect, distributed across the entire hippocampal population, and that it is likely to be inherited from the fascia dentata or an earlier stage in the hippocampal circuit, rather than generated intrinsically within CA1. It is hypothesized that the compression of temporal sequences of place fields within individual theta cycles permits the use of long-term potentiation for learning of sequential structure, thereby giving a temporal dimension to hippocampal memory traces.

A theory of hippocampal function in memory

Abstract: First, what is computed by the hippocampus is considered. Based on the effects of damage to the hippocampus and neuronal activity recorded in the primate hippocampus, it is suggested that it is involved in associating together information usually originating from different cortical regions, for example, about objects and their place in a spatial environment. The rapid formation of such context-dependent memories is prototypical of memories of particular events or episodes. Second, a computational theory of how it performs this function, based on neuroanatomical and neurophysiological information about the different neuronal systems contained within the hippocampus, is described. Key hypotheses are that the CA3 pyramidal cells operate as a single autoassociation network to store new episodic information as it arrives via a number of specialized preprocessing stages from many different association areas of the cerebral cortex, and that the dentate granule cell/mossy fiber system is important particularly during learning to help to produce a new pattern of firing in the CA3 cells for each episode. The computational analysis shows how many memories could be stored in the hippocampus, and how quickly the CA3 autoassociation system would operate during recall. The analysis is then extended to show how the CA3 system could be used to recall the whole of an episodic memory when only a fragment of it is presented. It is shown how this retrieval within the hippocampus could lead to recall of neuronal activity in association areas of the cerebral neocortex similar to that present during the original episode, via modified synapses in backprojection pathways from the hippocampus to the cerebral neocortex. The recalled information in the cerebral neocortex could then by used by the neocortex in the formation of long-term memories and/or in the selection of appropriate actions.

A sequence predicting CA3 is a flexible associator that learns and uses context to solve hippocampal‐like tasks

Abstract: The model discussed in this paper is, by hypothesis, a minimal, biologically plausible model of hippocampal region CA3. Because cognitive mapping can be viewed as a sequence prediction problem, we qualify this model as a successful sequence predictor. Since the model solves problems which require the use of context, the model is also able to learn and use context. The model also solves configural learning problems of which, at least one, requires a hippocampus. Thus, by solving sequence problems, by solving configural learning problems, and by creating codes for context, this model provides a computational unification of hippocampal functions which are often viewed as disparate.

Theory of rodent navigation based on interacting representations of space

Abstract: We present a computational theory of navigation in rodents based on interacting representations of place, head direction, and local view. An associated computer model is able to replicate a variety of behavioral and neurophysiological results from the rodent navigation literature. The theory and model generate predictions that are testable with current technologies.

Considerations arising from a complementary learning systems perspective on hippocampus and neocortex.

Abstract: We discuss a framework for the organization of learning systems in the mammalian brain, in which the hippocampus and related areas form a memory system complementary to learning mechanisms in neocortex and other areas. The hippocampal system stores new episodes and "replays" them to the neocortical system, interleaved with ongoing experience, allowing generalization as cortical memories form. The data to account for include: 1) neurophysiological findings concerning representations in hippocampal areas, 2) behavioral evidence demonstrating a spatial role for hippocampus, 3) and effects of surgical and pharmacological manipulations on neuronal firing in hippocampal regions in behaving animals. We hypothesize that the hippocampal memory system consists of three major modules: 1) an invertible encoder subsystem supported by the pathways between neocortex and entorhinal cortex, which provides a stable, compressed, invertible encoding in entorhinal cortex (EC) of cortical activity patterns, 2) a memory separation, storage, and retrieval subsystem, supported by pathways between EC, dentate gyrus and area CA3, including the CA3 recurrent collaterals, which facilitates encoding and storage in CA3 of individual EC patterns, and retrieval of those CA3 encodings, in a manner that minimizes interference, and 3) a memory decoding subsystem, supported by the Shaffer collaterals from area CA1 to area CA3 and the bi-directional pathways between EC and CA3, which provides the means by which a retrieved CA3 coding of an EC pattern can reinstate that pattern on EC. This model has shown that 1) there is a trade-off between the need for information-preserving, structure-extracting encoding of cortical traces and the need for effective storage and recall of arbitrary traces, 2) long-term depression of synaptic strength in the pathways subject to long-term potentiation is crucial in preserving information, 3) area CA1 must be able to exploit correlations in EC patterns in the direct perforant path synapses.

Neuronal computations underlying the firing of place cells and their role in navigation

Abstract: Our model of the spatial and temporal aspects of place cell firing and their role in rat navigation is reviewed. The model provides a candidate mechanism, at the level of individual cells, by which place cell information concerning self-localization could be used to guide navigation to previously visited reward sites. The model embodies specific predic- tions regarding the formation of placefields, the phase coding of place cell firing with respect to the hippocampal theta rhythm, and the formation of neuronal population vectors downstream from the place cells that code for the directions of goals during navigation. Recent experiments regard- ing the spatial distribution of place cell firing have confirmed our initial modeling hypothesis, that place fields are formed from Gaussian tuning curve inputs coding for the distances from environmental features, and enabled us to further specify the functional form of these inputs. Other recent experiments regarding the temporal distribution of place cell firing in two-dimensional environments have confirmed our predictions based on the temporal aspects of place cell firing on linear tracks. Directions for further experiments and refinements to the model are outlined for the future. r 1997 Wiley-Liss, Inc.

Neurophysiology of converging synaptic inputs from the rat prefrontal cortex, amygdala, midline thalamus, and hippocampal formation onto single neurons of the caudate/putamen and nucleus accumbens

Abstract: Neurophysiological responses mediated by projections from five telencephalic and diencephalic regions (the infra- and prelimbic portions of the prefrontal cortex, amygdala, midline and intralaminar thalamic nuclei, entorhinal cortex and subiculum/CA1) to the caudate/putamen (CPu) and nucleus accumbens (Acb) of the dorsal and ventral striatum were studied in chloral-hydrate-anesthetized rats. Both extra- and intracellular in vivo recording techniques were used. A retrograde tracer (wheatgerm agglutinin-apo-horseradish peroxidase-5 nm colloidal Gold) was deposited in some animals in the vicinity of recording sites to confirm that stimulating electrodes were located near cells that projected to the striatum. Electrical stimulation of these five regions, respectively, evoked excitatory responses in 60%, 22%, 51%, 25%, and 17% of striatal neurons. Some responses, particularly with thalamic stimulation, showed short-term frequency potentiation in which 5/s stimulation increased the probability of spike firing. About half of responsive cells showed convergent excitation to more than one stimulating site. It was possible with convergent excitatory responses to show synaptic interactions: simultaneous activation of more than one site produced spatial and temporal summation to increase the probability of spike firing. Up to 5-way convergence onto single striatal neurons and up to 3-way interactions could be shown. These results indicate that functional influences from the hippocampal formation can converge with other excitatory input onto single striatal neurons to effect synaptic integration.

Synaptic target selectivity and input of GABAergic basket and bistratified interneurons in the CA1 area of the rat hippocampus.

Abstract: To assess the position of interneurons in the hippocampal network, fast spiking cells were recorded intracellularly in vitro and filled with biocytin. Sixteen non-principal cells were selected on the basis of 1) cell bodies located in the pyramidal layer and in the middle of the slice, 2) extensive labeling of their axons, and 3) a branching pattern of the axon indicating that they were not axo-axonic cells. Examination of their efferent synapses (n = 400) demonstrated that the cells made synapses on cell bodies, dendritic shafts, spines, and axon initial segments (AIS). Statistical analysis of the distribution of different postsynaptic elements, together with published data (n = 288) for 12 similar cells, showed that the interneurons were heterogeneous with regard to the frequency of synapses given to different parts of pyramidal cells. When the cells were grouped according to whether they had less or more than 40% somatic synaptic targets, each population appeared homogeneous. The population (n = 19) innervating a high proportion of somata (53 ± 10%, SD) corresponds to basket cells. They also form synapses with proximal dendrites (44 ± 12%) and rarely with AISs and spines. One well-filled basket cell had 8,859 boutons within the slice, covering an area of 0.331 mm2 of pyramidal layer tangentially and containing 7,150 pyramidal cells, 933 (13%) of which were calculated to be innervated, assuming that each pyramidal cell received nine to ten synapses. It was extrapolated that the intact axon probably had about 10,800 boutons innervating 1,140 pyramids. The proportion of innervated pyramidal cells decreased from 28% in the middle to 4% at the edge of the axonal field. The other group of neurons, the bistratified cells (n = 9), showed a preference for dendritic shafts (79 ± 8%) and spines (17 ± 8%) as synaptic targets, rarely terminating on somata (4 ± 8%). Their axonal field was significantly larger (1,250 ± 180 μm) in the medio-lateral direction than that of basket cells (760 ± 130 μm). The axon terminals of bistratified cells were smaller than those of basket cells. Furthermore, in contrast to bistratified cells, basket cells had a significant proportion of dendrites in stratum lacunosum-moleculare suggesting a direct entorhinal input. The results define two distinct types of GABAergic neuron innervating pyramidal cells in a spatially segregated manner and predict different functions for the two inputs. The perisomatic termination of basket cells is suited for the synchronization of a subset of pyramidal cells that they select from the population within their axonal field, whereas the termination of bistratified cells in conjunction with Schaffer collateral/commissural terminals may govern the timing of CA3 input and/or voltage-dependent conductances in the dendrites. © 1996 Wiley-Liss, Inc.

LTD, LTP, and the sliding threshold for long-term synaptic plasticity.

Abstract: LTD of synaptic transmission is a form of long-term synaptic plasticity with the potential to be as significant as LTP to both the activity-dependent development of neural circuitry and adult memory storage. In addition, interactions between LTP and LTD and the dynamic regulation of the gain of synaptic plasticity mechanisms are also very important. In particular, the computational ability of LTD to properly counterbalance LTP may be essential to maintaining synaptic strengths in the linear range, and to maximally sharpen the ability of synapses to compute and store frequency-based information about the phase relation between synapses. Experimental data confirm the presence of an activity-dependent "sliding threshold" with the expected properties. That is, when levels of neuronal activity are high, indicating circumstances increasing the likelihood of inducing LTP, compensatory changes cause the suppression of LTP and an enhanced likelihood of LTD. Conversely, we would predict that low levels of synaptic activity would shift the threshold in favor of greater LTP and less LTD, a hypothesis which has yet to be tested. The sliding threshold for LTP and LTD also has implications for underlying cellular mechanisms of both forms of long-term synaptic plasticity. If the thresholds for LTP and LTD are tightly and reciprocally co-regulated, that could imply that at least one component of LTD is a true depotentiation caused by reversal of a change mediating LTP. If so, the intuitively simplest hypothesis is that phosphorylation of AMPA glutamate receptors causes LTP of synaptic e.p.s.p.s, while dephosphorylation of the same site or sites causes depotentiation LTD. Of course, this hypothesis would refer only to a postsynaptic component of both LTP and LTD. There has been a recent report that, in neonatal rat hippocampus, a form of LTD that is expressed developmentally earlier than LTP appears to have a postsynaptic induction site, but is expressed as decreased presynaptic transmitter release (Bolshakov and Siegelbaum, 1994). Whether these properties will be retained as LTD matures is unknown, as is the likelihood that, if a component of LTP is expressed presynaptically, depotentiation of that presynaptic component can also occur. Equally unclear is the persistence of LTD relative to LTP. The few rigorous long-term anatomical studies available suggest that the latest phases of LTP may be expressed as changes in dendritic spine shapes and/or synaptic morphology. While heterosynaptic LTD has been reported to have a duration of weeks in vivo (Abraham et al., 1994), we do not know whether LTP-induced morphological changes that take many days to appear can be reversed in an activity-dependent manner. An important feature of the consolidation of memories may turn out to be the slow development of LTP that is resistant to reversal by LTD. While we still at an earlier stage in our understanding of the mechanisms underlying LTD compared to LTP, some things are becoming clear. LTD is induced by afferent neuronal activity that is relatively ineffective in exciting the postsynaptic cell--an "anti-hebbian" condition. This property, coupled with the hebbian properties of LTP and the dynamic nature of membrane conductances, necessarily confers upon synapses the ability to compute and store the results of a covariance function. However, the role of such a computation in processing and/or memory is unclear. In addition, LTD appears to require the activation of NMDA and metabotropic subtypes of glutamate receptors, release of Ca2+ from intracellular stores, and an increase in intracellular [Ca2+] that is lower than that necessary to induce LTP. The early evidence is consistent with some overlap of targets for modification by LTP and LTD, with some forms of LTD likely to be a reversal, or "depotentiation," of previous LTP, perhaps through dephosphorylation of AMPA receptors.

Evidence for extrahippocampal involvement in place learning and hippocampal involvement in path integration

Abstract: Although there is a good deal of evidence that animals require the hippocampus for learning place responses, animals with damage to the afferent and efferent fibers coursing through the fimbria-fornix have been shown to acquire a place response. This finding suggests either that the cells of the hippocampus proper (CA1–4 and dentate gyrus), via their connections to the temporal lobe, can mediate place learning or that some extrahippocampal structure is sufficient. We examined this question using rats with ibotenic acid lesions of the cells of the hippocampus. Rats were pretrained to swim to a visible platform and then given probe trials on which the visible platform was removed. Video and kinematic analyses showed that the hippocampal rats expected to find the platform at its previous location because they swam directly to that location and paused and turned at that location after the platform was removed. Additional tests confirmed that they had learned a place response. There were, however, abnormalities in their swimming patterns, and despite having acquired one place response, they did not then acquire new place responses when only the hidden platform training procedure was used. These results demonstrate that place learning can be acquired by rats in which the hippocampus proper is removed. Contrasts between conditions in which hippocampal rats acquire a place response and conditions in which they fail suggests that the hippocampus may serve as an on line system for monitoring movement and integrating movement paths. © 1996 Wiley-Liss, Inc.

Physiological properties of anatomically identified basket and bistratified cells in the CA1 area of the rat hippocampus in vitro

Abstract: Basket and bistratified cells form two anatomically distinct classes of GABAergic local-circuit neurons in the CA1 region of the rat hippocampus. A physiological comparison was made of intracellularly recorded basket (n = 13) and bistratified neurons (n = 6), all of which had been anatomically defined by their efferent target profile (Halasy et al., 1996). Basket cells had an average resting membrane potential of -64.2 +/- 7.2 vs. -69.2 +/- 4.6 mV in bistratified cells. The latter had considerably higher mean input resistances (60.2 +/- 42.1 vs. 31.3 +/- 10.9 M Ohms) and longer membrane time constants (18.6 +/- 8.1 vs. 9.8 +/- 4.5 ms) than basket cells. Differences were also apparent in the duration of action potentials, those of basket cells being 364 +/- 77 and those of bistratified cells being 527 +/- 138 microseconds at half-amplitude. Action potentials were generally followed by prominent, fast after-hyperpolarizing potentials which in basket cells were 13.5 +/- 6.7 mV in amplitude vs. 10.5 +/- 5.1 in bistratified cells. The differences in membrane time constant, resting membrane potential, and action potential duration reached statistical significance (P < 0.05). Extracellular stimulation of Schaffer collateral/commissural afferents elicited short-latency excitatory postsynaptic potentials (EPSPs) in both cell types. The average 10-90% rise time and duration (at half-amplitude) of subthreshold EPSPs in basket cells were 1.9 +/- 0.5 and 10.7 +/- 5.6 ms, compared to 3.3 +/- 1.3 and 20.1 +/- 9.7 ms in bistratified cells, the difference in EPSP rise times being statistically significant. Basket and bistratified EPSPs were highly sensitive to a bath applied antagonist of non-N-methyl-D-aspartate (NMDA) receptors, whereas the remaining slow-rise EPSP could be abolished by an NMDA receptor antagonist. Increasing stimulation intensity elicited biphasic inhibitory postsynaptic potentials (IPSPs) in both basket and bistratified cells. In conclusion, basket and bistratified cells in the CA1 area show prominent differences in several of their membrane and firing properties. Both cell classes are activated by Schaffer collateral/commissural axons in a feedforward manner and receive inhibitory input from other, as yet unidentified, local-circuit neurons.

TraceLink: A model of amnesia and consolidation of memory

Abstract: A model of amnesia is introduced, called TraceLink, that consists of three systems: 1) a trace system (neocortex), 2) a link system (hippocampus), and 3) a modulatory system (hippocampus/fornix/basal forebrain). It aims to explain salient aspects of the neuropsychology of amnesia, such as Ribot gradients in retrograde amnesia, patterns of dissociation between anterograde and retrograde amnesia, recovery from amnesia, and a newly discovered form of amnesia (semantic dementia) that results from certain temporal lobe lesions that do not affect the hippocampus. The model, furthermore, offers a new explanation for the global neuroanatomy of the hippocampus and neocortex based on the assumption that the brain aims to minimize connectivity volume. It also offers various strategies for the consolidation of memory, the effects of which are explored in computer simulations. The paper concludes with ten, largely untested; predictions derived from the TraceLink model.

Ischemic brain damage and memory impairment: A commentary

Abstract: Studies in humans and monkeys have identified structures in the medial temporal lobe essential for memory (the hippocampal region, i.e., the dentate gyrus, the hippocampus, and the subicular complex, and the adjacent perirhinal, entorhinal, and parahippocampal cortices). Additional work has revealed that for both species, damage limited to the hippocampal region produces less severe memory impairment than damage that includes additional structures within the medial temporal lobe. This work has been based on both neurosurgical lesions and on lesions produced by global ischemia or anoxia. An important issue about ischemic damage is whether the damage identifiable in histopathological examination provides an accurate estimate of direct neural damage or whether additional direct damage might be present that is sufficient to disrupt neuronal function in areas important for memory and sufficient to impair behavioral performance, but not sufficient to progress to cell death and to be detectable in conventional histopathology. This commentary explores the issue of ischemic damage and memory impairment. Although few studies have addressed this issue directly, the currently available data from global ischemia in rats, monkeys, and humans are consistent with the hypothesis that the detectable neuronal damage is responsible for the severity of the observed behavioral impairment. Yet it is also true that this hypothesis has not been the target of very much systematic work. We encourage additional experimental work, especially in rats, that could further illuminate how to evaluate the behavioral effects of ischemic lesions.

Cerebral ischemia: Are the memory deficits associated with hippocampal cell loss?

Abstract: The long-standing notion that damage restricted to the hippocampal formation is sufficient to produce a significant global memory deficit derives from clinical data. Specifically, it is based on the observation that transient global ischemia, which leads to partial cell loss within the hippocampal formation but not in other brain areas important for memory, can produce global amnesia in humans. This view is, however, challenged by a number of experimental findings. First, in both monkeys and rats, there is evidence that ischemia disrupts delayed object recognition, a memory process found to be largely intact following selective hippocampal lesions. These findings indicate that damage confined to the hippocampal formation cannot account for all aspects of the ischemia-induced memory impairments. Second, although some groups of hippocampal neurons are the most prone to degeneration following ischemia, a wide array of extra-hippocampal damage has been observed in all species, for which the precise extent and distribution may well be underestimated by conventional histological evaluations of ischemic brains. Partial neuronal degeneration reported in regions such as the rhinal areas, medial dorsal thalamic nucleus, or cingulate cortex may contribute to varying degrees to ischemia-induced memory deficits. Third, experimental studies have failed to generate a general consensus on the correlation between extent of hippocampal cell loss and memory performance. In sum, the experimental studies do not, as yet, support the view that hippocampal damage is solely responsible for ischemia-induced memory deficits. Rather, they suggest that both the intra- and extra-hippocampal damage contribute to the pattern of memory impairments observed following ischemia. Consequently, although animals with global and focal ischemia represent valuable models for neuropathological and therapeutic studies, they may not be so useful in assessing the role of the hippocampal formation and its sub-components in memory processes. © 1996 Wiley-Liss, Inc.

Dynamic Synapse: A New Concept of Neural Representation and Computation

Abstract: Presynaptic mechanisms influencing the probability of neurotransmitter release from an axon terminal, such as facilitation, augmentation, and presynaptic feedback inhibition, are fundamental features of biological neurons and are cardinal physiological properties of synaptic connections in the hippocampus. The consequence of these presynaptic mechanisms is that the probability of release becomes a function of the temporal pattern of action potential occurrence, and hence, the strength of a given synapse varies upon the arrival of each action potential invading the terminal region. From the perspective of neural information processing, the capability of dynamically tuning the synaptic strength as a function of the level of neuronal activation gives rise to a significant representational and processing power of temporal spike patterns at the synaptic level. Furthermore, there is an exponential growth in such computational power when the specific dynamics of presynaptic mechanisms varies quantitatively across axon terminals of a single neuron, a recently established characteristic of hippocampal synapses. During learning, alterations in the presynaptic mechanisms lead to different pattern transformation functions, whereas changes in the postsynaptic mechanisms determine how the synaptic signals are to be combined. We demonstrate the computational capability of dynamic synapses by performing speech recognition from unprocessed, noisy raw waveforms of words spoken by multiple speakers with a simple neural network consisting of a small number of neurons connected with synapses incorporating dynamically determined probability of release. The dynamics included in the model are consistent with available experimental data on hippocampal neurons in that parameter values were chosen so as to be consistent with time constants of facilitative and inhibitory processes governing the dynamics of hippocampal synaptic transmission studied using nonlinear systems analytic procedures.

Pyramidal cell dendrites are the primary targets of calbindin D28k-immunoreactive interneurons in the hippocampus

Abstract: The axonal arborization and postsynaptic targets of calbindin D28k (CB)-immunoreactive nonprincipal neurons have been studied in the rat dorsal hippocampus. Two types of neurons were distinguished on the basis of soma location, the characteristics of the dendritic tree, and the axon arborisation pattern. Type I cells were located in stratum radiatum of the CA1 and CA3 regions and occasionally in strata pyramidale and oriens. These cells had multipolar or bitufted dendritic trees primarily located in stratum radiatum. Their axons could be followed for a considerable distance, arborised within stratum radiatum, and were covered with regularly spaced small boutons. As demonstrated with postembedding immunogold staining, their axon terminals were γ-aminobutyric acid (GABA) immunoreactive, and formed symmetrical synapses pre-dominantly on proximal and distal dendrites of pyramidal cells (28% and 58%, respectively), and occasionally on spines (9%) or on GABA-positive dendrites (5%). Type II cells were found exclusively in stratum oriens of the CA1 and CA3 regions and possessed large, fusiform cell bodies and long, horizontally oriented dendrites. Their axon initial segments turned towards the alveus and disappeared in a myelin sheet, which was often possible to follow into the white matter. We conclude that type I CB-immunoreactive cells are likely to represent a major source of inhibitory synapses in the dendritic region of pyramidal cells, which are responsible for the control of dendritic electrogenesis. The distribution of local collaterals of type II cells—if they have any—remains unknown, but their main axon is likely to project to the medial septum. © 1996 Wiley-Liss, Inc.

Memory for places: a navigational model in support of Marr's theory of hippocampal function.

Abstract: In this report we describe a model that applies Marr's theory of hippocampal function to the problem of map-based navigation. Like many others we attribute a spatial memory function to the hippocampus, but we suggest that the additional functional components required for map-based navigation are located elsewhere in the brain. One of the key functional components in this model is an egocentric map of space, located in the neocortex, that is continuously updated using ideothetic (self-motion) information. The hippocampus stores snapshots of this egocentric map. The modeled activity pattern of head direction cells is used to set the best egocentric map rotation to match the snapshots stored in the hippocampus, resulting in place cells with a nondirectional firing pattern. We describe an evaluation of this model using a mobile robot and demonstrate that with this model the robot can recognize an environment and find a hidden goal. This model is discussed in the context of prior experiments that were designed to discover the map-based spatial processing of animals. We also predict the results of further experiments.

NMDA receptor-dependent LTD in different subfields of hippocampus in vivo and in vitro.

Abstract: In simulations with artificial neural networks, efficient information processing and storage has been shown to require that the strength of connections between network elements has the capacity to both increase and decrease in a use-dependent manner. In contrast to long-term potentiation (LTP) of excitatory synaptic transmission, activity-dependent long-term depression (LTD) has been difficult to demonstrate in forebrain in vivo. Theoretical arguments indicate that coincidence of presynaptic excitation and low-magnitude postsynaptic activation are the necessary prerequisites for LTD induction. Here we report that stimulation paradigms which cause 1) sufficient excitation to result in NMDA receptor activation and simultaneously 2) attenuate the level of postsynaptic activation by recruitment of GABAA receptor-mediated inhibition consistently produce LTD of commissural input to area CA1 in the hippocampus of anesthetized adult rats, and of the perforant path input to the dentate gyrus in the hippocampus of anesthetized and unanesthetized adult rabbits. A functionally similar pre- and postsynaptic activation pattern applied to the hippocampal slice preparation by injecting hyperpolarizing current into the postsynaptic cell during NMDA receptor-mediated excitation also was effective in consistently inducing LTD. Results of studies in vitro show that Ca2+ influx through the NMDA channel is necessary for the induction of LTD, and moreover, that NMDA receptors also participate in the expression of LTD. Our findings demonstrate a general mechanism for the implementation of a theoretically derived learning rule in adult forebrain in vivo and in vitro and provide justification for the inclusion of use-dependent decreases of connection weights in formal models of cognitive processing.

Low‐frequency trains of paired stimuli induce long‐term depression in area CA1 but not in dentate gyrus of the intact rat

Abstract: We have examined the efficacy of a recently introduced protocol for inducing homosynaptic long-term depression (LTD) in area CA1 of the anesthetized rat (Thiels et al. [1994] J Neurophysiol 72:3009-3116.). In area CA1 of the awake animal, this protocol, consisting of 200 pairs of pulses delivered at 0.5 Hz, with an interpulse interval of 25 ms, consistently produced LTD, provided the initial pulse was sufficiently strong to produce significant paired-pulse depression of the evoked response. We extended these experiments to the dentate gyrus, using either paired pulses given to the perforant path in the awake adult rat, or, in the anesthetized adult, a two-pathway pairing procedure, in which the first pulse was delivered to the commissural input to the dentate gyrus and the second to the perforant path. In both cases, the first pulse led to substantial suppression of the response evoked by the second pulse. With neither protocol, however, was there any evidence for LTD or depotentiation. Paired-pulse stimulation of the perforant path of young rats (10-11 days) also failed to induce LTD or depotentiation of the population excitatory postsynaptic potential (EPSP). Thus, the dentate gyrus in the intact animal appears to be less susceptible to LTD and depotentiation than area CA1, a conclusion consistent with previous experiments in which we found that stimulation at 1-5 Hz produced LTD/depotentiation in area CA1 of young (but not adult) rats in vivo but was ineffective at any age in the dentate gyrus. Our results do not rule out the possibility that other, untested protocols may produce homosynaptic LTD and/or depotentiation in the dentate gyrus in vivo.

Modeling the spontaneous reactivation of experience-specific hippocampal cell assembles during sleep.

Abstract: During slow-wave sleep (SWS) following periods of spatial activity, hippocampal place cells that were temporally correlated, by virtue of the overlap of their place fields, exhibit enhanced temporal correlations, even though the animal sleeps in a different location (Wilson and McNaughton [1994] Science 267:676-679). The discharge of cells with overlapped place fields is more correlated in subsequent sleep, particularly during sharp waves, than in sleep episodes prior to the behavior, or than cell pairs with non-overlapped place fields. The reactivated correlated states appear during hippocampal sharp waves (SPWs), and are weak or absent in the inter-SPW interval. A simple conceptual hypothesis for this phenomenon is developed, based on the idea that hippocampal place fields reflect a two-dimensional distribution of continuously overlapping dynamic attractors in which each location is represented by the self-sustaining activity of a small subset of neurons with overlapping place fields. A numerical simulation of this hypothesis, based on a simplified representation of the CA8 recurrent network, accounts qualitatively for the main observations, including SPW-like dynamics. It is shown that, under conditions in which the connection patterns have been previously established, either associative or nonassociative mechanisms might underlie the reactivation of recently experienced states. These two alternatives appear, under at least some conditions (e.g., sparse coding), to be indistinguishable.

Homosynaptic LTD and depotentiation: do they differ in name only?

Abstract: Long-term depression (LTD) now occupies a major place in theories of the cellular basis of learning and memory and other nervous system phenomena involving persistent changes in synaptic responsiveness. LTD can be induced using a variety of stimulation paradigms. Homosynaptic LTD in this review refers to a depression of basal responses that is restricted to the pathway that has been stimulated by a low-frequency (1 Hz) stimulus train. Despite the intensive interest in LTD, there has been controversy about the ease with which LTD can be induced and reports range from no success to routine success. There has been much less controversy about a related form of response depression now called "depotentiation" which shares many similarities with LTD. Depotentiation is the response reduction that affects, not the basal responses affected by LTD, but responses that have been increased by the process of long-term potentiation (LTP). LTD and depotentiation can be induced by similar stimulation and have many biochemical properties in common, but it has not been clear whether or not they represent the same phenomenon, in part because it often occurs that the same preparation that does not undergo LTD readily expresses depotentiation. We review work that indicates that the major differences between LTD and depotentiation involve age-dependence, the need for priming stimulation and sensitivity to GABA receptor antagonists. We present a hypothetical model that can reconcile the apparent disparities between LTD and depotentiation.

Neural network modeling of the hippocampal formation spatial signals and their possible role in navigation: A modular approach

Abstract: Cells throughout the hippocampal formation show striking spatial firing correlates as a rat navigates through space. These cells are thought to play a critical role in orchestrating the navigational abilities of the animals, since damage to the hippocampal formation causes spatial learning deficits. Here, we present a theoretical framework aimed at explaining how the different spatial signals are generated, as well as how they may help guide navigational behavior. Earlier work from our laboratory has presented a simple model for how the location-related signals exhibited by hippocampal place cells could be generated, based on convergent sensory information. Here, the results of this work are combined with two more recent models, to provide a more comprehensive theoretical framework. Specifically, we present 1) A neural network model of head direction cells, based on the idea that the directional signals are generated using a path integration mechanism. Cells which combine directional and angular head velocity information project onto the head direction cells, to "update" the current directional signal. This model reproduces the basic phenomenon of direction-specific firing, as well as the anticipatory nature of this firing, reported for some head direction cells. 2) A network simulation of how the hippocampal spatial signals could be used to orchestrate instrumental learning. Here, place and directional signals converge onto motor cells, each of which are thus driven to fire to specific combinations of location and directional heading. Each active motor cell generates a small leftward or rightward "step" of the simulated animal. When the simulated goal is encountered, recently active synapses are strengthened, so that goal-directed trajectories are "stamped in". We have found these models useful in helping to clarify our thinking about the proposed theoretical principles, as well as in generating testable predictions.

Attention, configuration, and hippocampal function.

Abstract: We present a neural network that characterizes a remarkably large number of classical conditioning paradigms and describes the effects of many neurophysiological manipulations. First, the network 1) describes behavior in real time 2) comprises simple and configural stimulus representations, and 3) includes attentional control of storage and retrieval. Second, mapping of the network onto the brain can be summarized by several information processing loops: 1) a hippocampal-cortical configural loop, 2) a hippocampal-cerebellar conditioned-response loop, 3) a hippocampal-accumbens-thalamic attentional loop, and 4) a hippocampal-medial raphe-medial septum error loop. Third, within this global view of brain function, it is assumed that the hippocampal formation computes 1) the aggregate prediction of environmental events and 2) the error signals for cortical learning. These assumptions are supported by rigorous computer simulations consistent with a large body of data on hippocampal and septal neural activity, induction and blockade of hippocampal long-term potentiation, administration of cholinergic agonists and antagonists, administration of haloperidol, and selective and nonselective hippocampal and cortical lesions. © 1997 Wiley-Liss, Inc.

Distinct memory circuits composing the hippocampal region

Abstract: The very different anatomical designs of the adjacent circuitries of the cortico-hippocampal pathway, along with their somewhat different synaptic plasticity mechanisms, suggest a nearly serial pathway of distinct memory circuits each contributing its own specialized processing operation to overall hippocampal function. Modeling and formal theoretical analysis of the prominent anatomical design features of particular circuits (piriform/entorhinal cortex; hippocampal field CA3; hippocampal field CA1) are found to identify potential emergent function not readily arrived at in the absence of these formal models, and yet which once derived can be seen potentially to confer unique capabilities to an integrated hippocampal mechanism for processing memories during behavior.

Hippocampal place cells connected by Hebbian synapses can solve spatial problems.

Abstract: We propose that a cognitive map can be stored in the synapses between the pyramidal cells of CA3 in the form of the pattern of synaptic strengths connecting them. The model requires only that there are place cells in CA3 and that the connections between them are modifiable in a Hebbian manner. Given these suppositions, the synaptic strengths must evolve to represent the distance between firing centers of synaptically connected place cells. We argue that this arrangement of synaptic weights embodies all the formal properties of a map. We demonstrate that the information stored in such a structure is sufficient to solve several classic spatial problems including finding shortest paths, and negotiating detours. It is clear that much of the physiology and anatomy necessary to more precisely characterize the model is not known at this time. Nevertheless the model is robust under a variety of cell and connection densities. It also performs well under several different functions relating distance to synaptic strength. What is most remarkable in the model is that it is a logical consequence of the several key anatomical and physiological properties of the CA3 region of rats. Whether this information is used by the rat is difficult to assess at this time. Regardless of the outcome of this question, the model has promising applications to the field of robot navigation.

Task‐relevant late positive component in rats: Is it related to hippocampal theta rhythm?

Abstract: Long-latency components of event-related potentials (like the P300 or P3) correlate with the ability of subjects to detect and process unexpected, novel or task-relevant events. Task-relevant late positive components were recorded in the neocortex and hippocampus of rats performing an auditory discrimination task, similar to the "odd-ball" paradigm used in human experiments. Surface and depth electrodes were implanted in anaesthetized rats at frontal, temporal and anterior occipital neocortical regions and the hippocampus. After recovery from surgery rats were trained to discriminate two auditory signals, a frequent irrelevant tone and a rare tone related to water reward. In response to the task-relevant tone but not the irrelevant tone, P300-like late positive components (mean latency of 274 ms) were recorded throughout the surface of the neocortex. The largest amplitudes were found at the anterior occipital cortex situated above the hippocampal CA1 region. The amplitude of the task-relevant positive component increased further with cortical depth without reversing its polarity. An amplitude maximum was found in the CA1 region with a polarity reversal at the pyramidal cell layer and the largest negative amplitude in stratum radiatum. Power spectra of differences between responses evoked by task-relevant tones and those evoked by irrelevant tones revealed peaks in the theta range (4-12 Hz). It is suggested that the P300-like component in rats corresponds to a theta wave out of a burst of hippocampal theta cycles.

Genetic variation in the morphology of the septo-hippocampal cholinergic and GABAergic system in mice. I. Cholinergic and GABAergic markers

Abstract: In the present study, variations of cholinergic and GABAergic markers in the medial septum/vertical limb of the diagonal band of Broca (MS/vDB) and the hippocampus of eight different inbred mouse strains were investigated. By means of immunocytochemistry against the acetylcholine-synthesizing enzyme choline acetyltransferase (ChAT), the cholinergic neurons were visualized and the number of ChAT-positive neuronal profiles in the MS/vDB was counted. Cholinergic and GABAergic septo-hippocampal projection neurons were detected with a combined retrograde tracing and immunocytochemical approach. In order to quantify the cholinergic innervation of various hippocampal sub-regions, we estimated the density of acetylcholinesterase (AChE)-containing fibers as visualized by AChE histochemistry. Additionally, the densities of muscarinic receptors (mainly the subtypes M1 and M2) in different hippocampal areas of seven inbred strains were measured by means of quantitative receptor autoradiography. We found significant strain differences for the number of ChAT-positive neurons in the MS/vDB; in the numbers of cholinergic septo-hippocampal projection neurons; in the density of cholinergic fibers in hippocampal subfields CA3c, CA1, and in the dentate gyrus; and in the density of muscarinic receptors in the hippocampus. In contrast the GABAergic component of the septo-hippocampal projection did not differ between the strains investigated. The number of ChAT-reactive neurons in the MS/vDB was not correlated with either hippocampal cholinergic markers. This might be attributed to different collateralization of cholinergic neurons or to different projections of these neurons to other brain regions. These results show a strong hereditary variability within the septo-hippocampal cholinergic system in mice. In view of the role of the cholinergic system in learning and memory processes, strain differences in cholinergic markers might be helpful in explaining behavioral variation.

Genetic variation in the morphology of the septo-hippocampal cholinergic and GABAergic systems in mice: II. Morpho-behavioral correlations

Abstract: We investigated the contribution of the septo-hippocampal cholinergic and GABAergic system to spatial and nonspatial aspects of learning and memory that had previously been found to correlate with the extent of the hippocampal intra- and infrapyramidal mossy fiber projection in different inbred mouse strains. The following cholinergic and GABAergic markers were measured in the septi and hippocampi of male mice: the number of cholinergic and parvalbumin-containing neurons in the medial septum/vertical limb of the diagonal band of Broca (MS/vDB), the number of septo-hippocampal cholinergic and GABAergic projection neurons, the density of cholinergic fibers in different hippocampal subfields, and the density of muscarinic receptors (predominantly M1 and M2) in the hippocampus. In addition, animals were behaviorally tested for spatially dependent and activity-dependent variables in a water maze and spatial and nonspatial working and reference memory in different experimental set-ups in an eight-arm radial maze. Using only those variables for which significant strain differences were obtained, we looked for covariations between behavior and neuroanatomy. The density of cholinergic fibers in the dentate gyrus was significantly correlated with activity-dependent learning in the water maze, whereas the number of septo-hippocampal cholinergic projection neurons correlated with spatial and, to a lesser extent, also with nonspatial aspects of radial maze learning. Only weak correlations were found between receptor densities and behavioral traits. From these data we conclude that variations in the septo-hippocampal cholinergic system, like variations in the mossy fiber projection, entail functional consequences for different types of maze learning in mice.

Dendritic electrogenesis in rat hippocampal CA1 pyramidal neurons: functional aspects of Na+ and Ca2+ currents in apical dendrites

Abstract: The regenerative properties of CA1 pyramidal neurons were studied through differential polarization with external electrical fields. Recordings were obtained from somata and apical dendrites in the presence of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), DL-2-amino-5-phosphonovaleric acid (APV), and bicuculline. S+ fields hyperpolarized the distal apical dendrites and depolarized the rest of the cell, whereas S divided by fields reversed the polarization. During intradendritic recordings, S+ fields evoked either fast spikes or compound spiking. The threshold response consisted of a low-amplitude fast spike and a slow depolarizing potential. At higher field intensities the slow depolarizing potential increased in amplitude, and additional spikes of high amplitude appeared. During intrasomatic recordings, S+ field evoked repetitive firing of fast spikes, whereas S divided by fields evoked a slow depolarizing potential on top of which high- and low-amplitude spikes were evoked. Tetrodotoxin (TTX) blocked all types of responses in both dendrites and somata. Perfusion with Ca(2+)-free, Co(2+)-containing medium increased the frequency and amplitude of fast spikes evoked by S+ field and substantially reduced the slow depolarizing potential evoked by S+ field and substantially reduced the slow depolarizing potential evoked by S divided by fields. Antidromic stimulation revealed that an all-or-none dendritic component was activated in the distal apical dendrites by back-propagating somatic spikes. The dendritic component had an absolute refractory period of about 4 ms and a relative refractory period of 10-12 ms. Ca(2+)-dependent spikes in the dendrites were followed by a long-lasting afterhyperpolarization (AHP) and a decrease in membrane input resistance, during which dendritic excitability was selectively reduced. The data suggest that generation of fast Na+ currents and slow Ca2+ currents in the distal part of apical dendrites is highly sensitive to the dynamic state of the dendritic membrane. Depending on the mode and frequency of activation these currents can exert a substantial influence on the input-output behavior of the pyramidal neurons.