Biomedical subjects
Publications and source records attributed to M E Hasselmo.
Neural models assist in characterizing the processes carried out by cortical and hippocampal memory circuits. Recent models of memory have addressed issues including recognition and recall dynamics, sequences of activity as the unit of storage, and consolidation of intermediate-term episodic memory into long-term memory.
Modulation of inhibitory synaptic potentials in the piriform cortex. Intracellular recordings from pyramidal neurons in brain slice preparations of the piriform cortex were used to test results from a computational model about the effects of cholinergic agonists on inhibitory synaptic potentials induced by stimulation of afferent fibers in layer Ia and association/intrinsic fibers in layer Ib. A simple model of piriform cortex as an associative memory was used to analyze how suppression of inhibitory synaptic transmission influenced performance of the network. Levels of suppression of excitatory synaptic transmission were set at levels determined in previous experimental work. Levels of suppression of inhibitory synaptic transmission were then systematically varied within the model. This modeling work demonstrated that suppression of inhibitory synaptic transmission in layer Ib should be stronger than suppression of inhibitory synaptic transmission in layer Ia to keep activity levels high enough for effective storage. Experimental data showed that perfusion of the cholinergic agonist carbachol caused a significant suppression of inhibitory postsynaptic potentials (IPSPs) in the pyramidal neurons that were induced by stimulation of layer Ib, with a weaker effect on IPSPs induced by stimulation of layer Ia. As previously described, carbachol also selectively suppressed excitatory postsynaptic potentials (EPSPs) elicited by intrinsic but not afferent fiber stimulation. The decrease in amplitude of IPSPs induced by layer Ib stimulation did not appear to be directly related to the decrease in EPSP amplitude induced by layer Ib stimulation. The stimulation necessary to induce neuronal firing with layer Ia stimulation was reduced in the presence of carbachol, whereas that necessary to induce neuronal firing with layer Ib stimulation was increased, despite the depolarization of resting membrane potential. Thus physiological data on cholinergic modulation of inhibitory synaptic potentials in the piriform cortex is compatible with the functional requirements determined from computational models of piriform cortex associative memory function.
Electrical stimulation of the horizontal limb of the diagonal band of Broca (HDB) was coupled with recording of evoked potentials in the piriform cortex. Stimulation of the HDB caused an enhancement of the late, disynaptic component of the evoked potential elicited by stimulation of the lateral olfactory tract but caused a suppression of the synaptic potential elicited by stimulation of the posterior piriform cortex. The muscarinic antagonist scopolamine blocked both effects of HDB stimulation. The enhancement of disynaptic potentials could be due to cholinergic depolarization of pyramidal cells, whereas the suppression of potentials evoked by posterior piriform stimulation could be due to presynaptic inhibition of intrinsic fiber synaptic transmission by acetylcholine.
Changes in GABA modulation may underlie experimentally observed changes in the strength of synaptic transmission at different phases of the theta rhythm (Wyble, Linster, & Hasselmo, 1997). Analysis demonstrates that these changes improve sequence disambiguation by a neural network model of CA3. We show that in the framework of Hopfield and Tank (1985), changes in GABA suppression correspond to changes in the effective temperature and the relative energy of data terms and constraints of an analog network. These results suggest that phasic changes in the activity of inhibitory interneurons during a theta cycle may produce dynamics that resemble annealing. These dynamics may underlie a role for the theta cycle in improving sequence retrieval for spatial navigation.
Computational models of hippocampal region CA3 were used to study the role of theta rhythm in storage and retrieval of temporal sequences of neuronal activity patterns. Retrieval of multiple overlapping temporal sequences requires a mechanism for disambiguation, e.g., for choosing between two sequences with the same starting pattern but different final patterns (forked sequences). Modulatory input to the hippocampus from the medial septum may enhance the disambiguation of pattern sequences by causing phasic changes in the relative strength of afferent input and recurrent excitation. In the models, the strength of recurrent synaptic transmission is modulated by activation of GABA(B) receptors. Theta frequency inputs from the medial septum cause oscillations in the levels of GABA in the model, producing phasic changes in the strength of synaptic potentials during a theta cycle similar to those observed experimentally (Wyble et al., Soc Neurosci Abstr 1997;23: 197.7). These phasic changes in GABA(B) suppression improve sequence disambiguation in the simulations, as previously shown with analysis of a simpler model (Sohal and Hasselmo, Neural Comp 1998;10:889-902). In addition, tonic changes in levels of cholinergic modulation enhance the storage of forked sequences by preventing a strong influence of recurrent synapses during storage.
The hippocampus has long been thought to be an important cortical region for associative learning and memory. After several decades of experimental and theoretical studies, a picture is emerging slowly of the generic types of learning tasks that this neural structure might be essential for solving. Recently, there have been attempts to unify electrophysiological and behavioral observations from rodents performing spatial learning tasks with data from primates performing various tests of conditional and discrimination learning. Most of these theoretical frameworks have rested primarily on behavioral observations. Complementing these perspectives,we ask the question: given certain physiological constraints at the neuronal and cortical level, what class of learning problems is the hippocampus, in particular, most suited to solve? From a computational point of view, we argue that this structure is involved most critically in learning and memory tasks in which discontiguous items must be associated, in terms of their temporal or spatial positioning, or both.
Pyramidal cells in piriform (olfactory) cortex receive afferent input from the olfactory bulb as well as intrinsic association input from piriform cortex and other cortical areas. These two functionally distinct inputs terminate on adjacent apical dendritic segments of the pyramidal cells located in layer Ia and layer Ib of piriform cortex. Studies with bath-applied cholinergic agonists have shown suppression of the fast component of the inhibitory postsynaptic potentials (IPSPs) evoked by stimulation of the association fibers. It was previously demonstrated that an associative form of LTP can be induced by coactivation of the two fiber systems after blockade of the fast, gamma-aminobutyric acid-A-mediated IPSP. In this report, we demonstrate that an associative form of long-term potentiation can be induced by coactivation of afferent and intrinsic fibers in the presence of the cholinergic agonist carbachol.
Free recall and recognition are simulated in a network model of the hippocampal formation, incorporating simplified simulations of neurons, synaptic connections, and the effects of acetylcholine. Simulations focus on modeling the effects of the acetylcholine receptor blocker scopolamine on human memory. Systemic administration of scopolamine is modeled by blockade of the cellular effects of acetylcholine in the model, resulting in memory impairments replicating data from studies on human subjects. This blockade of cholinergic effects impairs the encoding of new input patterns (as measured by delayed free recall), but does not impair the delayed free recall of input patterns learned before the blockade. The impairment is selective to the free recall but not the recognition of items encoded under the influence of scopolamine. In the model, scopolamine blocks strengthening of recurrent connections in region CA3 to form attractor states for new items (encoding impaired) but allows recurrent excitation to drive the network into previously stored attractor states (retrieval spared). Neuron populations representing items (individual words) have weaker recurrent connections than neuron populations representing experimental context. When scopolamine further weakens the strength of recurrent connections it selectively prevents the subsequent reactivation of item attractor states by context input (impaired free recall) without impairing the subsequent reactivation of context attractor states by item input (spared recognition). This asymmetry in the strength of attractor states also allows simulation of the list-strength effect for free recall but not recognition. Simulation of a paired associate learning paradigm predicts that scopolamine should greatly enhance proactive interference due to retrieval of previously encoded associations during storage of new associations.
Clinical and experimental observations indicate that the hippocampus is critical in the formation of declarative memories. Interestingly, electrophysiological studies have demonstrated that the region also has a particularly low seizure threshold, where globally synchronous synaptic activity can occur. By using a detailed biophysical model of area CA3, it is shown how septal cholinergic modulation, through three distinct mechanisms, can interact with intrinsic and synaptic conductances to influence population behavior. A dissection of each mechanism demonstrates a variety of population firing activity ranging from fully synchronized behavior to a mixture of repetitive bursting and oscillations in reduced subsets of neurons, ideal for forming accurate associations during a learning and recall task.
Norepinephrine has been proposed to influence signal-to-noise ratio within cortical structures, but the exact cellular mechanisms underlying this influence have not been described in detail. Here we present data on a cellular effect of norepinephrine that could contribute to the influence on signal-to-noise ratio. In brain slice preparations of the rat piriform (olfactory) cortex, perfusion of norepinephrine causes a dose-dependent suppression of excitatory synaptic potentials in the layer containing synapses among pyramidal cells in the cortex (layer Ib), while having a weaker effect on synaptic potentials in the afferent fiber layer (layer Ia). Effects of norepinephrine were similar in dose-response characteristics and laminar selectivity to the effects of the cholinergic agonist carbachol, and combined perfusion of both agonists caused effects similar to an equivalent concentration of a single agonist. In a computational model of the piriform cortex, we have analyzed the effect of noradrenergic suppression of synaptic transmission on signal-to-noise ratio. The selective suppression of excitatory intrinsic connectivity decreases the background activity of modeled neurons relative to the activity of neurons receiving direct afferent input. This can be interpreted as an increase in signal-to-noise ratio, but the term noise does not accurately characterize activity dependent on the intrinsic spread of excitation, which would more accurately be described as interpretation or retrieval. Increases in levels of norepinephrine mediated by locus coeruleus activity appear to enhance the influence of extrinsic input on cortical representations, allowing a pulse of norepinephrine in an arousing context to mediate formation of memories with a strong influence of environmental variables.
A detailed biophysical model of hippocampal region CA3 was constructed to study how GABAergic modulation influences place field development and the learning and recall of sequence information. Simulations included 1,000 multicompartmental pyramidal cells, each consisting of seven intrinsic and four synaptic currents, and 200 multicompartmental interneurons, consisting of two intrinsic and four synaptic currents. Excitatory rhythmic septal input to the apical dendrites of pyramidal cells and both excitatory and inhibitory input to interneurons at theta frequencies provided a cellular basis for the development of theta and gamma frequency oscillations in population activity. The fundamental frequency of theta oscillations was dictated by the driving rhythm from the septum. Gamma oscillation frequency, however, was determined by both the decay time of the gamma-aminobutyric acid-A (GABA(A))-receptor-mediated synaptic current and the overall level of excitability in interneurons due to alpha-amino-3-hydroxy-5-methyl-4-isoxazole proprionic acid and N-methyl-D-aspartate (NMDA)-receptor-gated channel activation. During theta population activity, total GABA(B)-receptor-mediated conductance levels were found to gradually rise and fall in rhythmic fashion with the predominant population frequency (theta rhythm). This resulted in periodic GABA(B)-receptor-mediated suppression of excitatory synaptic transmission at recurrent collaterals (intrinsic fibers) of pyramidal cells and suppression of inhibitory synaptic transmission to both pyramidal cells and interneurons. To test the ability of the model to learn and recall temporal sequence information, a completion task was employed. During learning, the network was presented a sequence of nonorthogonal spatial patterns. Each input pattern represented a spatial "location" of a simulated rat running a specific navigational path. Hebbian-type learning was expressed as an increase in postsynaptic NMDA-receptor-mediated conductances. Because of several factors including the sparse, asymmetric excitatory synaptic connections among pyramidal cells in the model and a sufficient degree of random "background" firing unrelated to the input patterns, repeated simulated runs resulted in the gradual emergence of place fields where a given cell began to respond to a contiguous segment of locations on the path. During recall, the simulated rat was placed at a random location on the previously learned path and tested to see whether the sequence of locations could be completed on the basis of this initial position. Periodic GABA(B)-receptor-mediated suppression of excitatory and inhibitory transmission at intrinsic but not afferent fibers resulted in sensory information about location being dominant during early portions of each theta cycle when GABA(B)-receptor-related effects were highest. This suppression declined with levels of GABA(B) receptor activation toward the end of a theta cycle, resulting in an increase in synaptic transmission at intrinsic fibers and the subsequent recall of a segment of the entire location sequence. This scenario typically continued across theta cycles until the full sequence was recalled. When the GABA(B)-receptor-mediated suppression of excitatory and inhibitory transmission at intrinsic fibers was not included in the model, place field development was curtailed and the network consequently exhibited poor learning and recall performance. This was, in part, due to increased competition of information from intrinsic and afferent fibers during early portions of each theta cycle. Because afferent sensory information did not dominate early in each cycle, the current location of the rat was obscured by ongoing activity from intrinsic sources. (ABSTRACT TRUNCATED)
Whole-cell and extracellular recording techniques were used to examine local circuit inhibition in the CA1 region of the rat hippocampus in vitro. Activation, primarily of the recurrent inhibitory circuit by alvear stimulation, elicited an IPSP in pyramidal neurons that was dependent, in part, on NMDA receptor activation. Application of a tetanizing stimulus to the alveus evoked long-term potentiation (LTP) of the intracellularly recorded recurrent IPSPs. This LTP also was NMDA-dependent and was more sensitive to blockade by the NMDA antagonists 2-amino-5-phosphonovalerate (APV) and N-acetyl-aspartyl-glutamate, than the excitatory LTP produced by Schaffer collateral stimulation. With regard to APV, the sensitivity of inhibitory LTP was an order of magnitude greater. A biophysical simulation of hippocampal CA1 circuitry was used in a model of learned pattern recognition that included LTP in both excitatory and inhibitory recurrent circuits. In this model, selective blockade of inhibitory LTP produced aberrant spread of lateral excitation, resulting in confusion of normally distinguishable patterns of neuronal activity. Consideration is given to the possibility that selective disruption of NMDA-dependent modulation of local circuit inhibition may serve as a model for some aspects of dysfunction associated with NMDA-antagonist exposure and schizophrenia.
This research focuses on linking episodic memory function to the cellular physiology of hippocampal neurons, with a particular emphasis on modulatory effects at cholinergic and gamma-aminobutyric acid B receptors. Drugs which block acetylcholine receptors (e.g., scopolamine) have been shown to impair encoding of new information in humans, nonhuman primates, and rodents. Extensive data have been gathered about the cellular effects of acetylcholine in the hippocampus. In this research, models of individual hippocampal subregions have been utilized to understand the significance of particular features of modulation, and these hippocampal subregions have been combined in a network simulation which can replicate the selective encoding impairment produced by scopolamine in human subjects.
We studied the effect of long term baclofen treatment on recognition memory and novelty detection in rats using a habituation paradigm in an open field setting. Rats pretreated with 3 weeks' daily baclofen injection (0, 2 and 5 mg/kg) were tested in four 10 min sessions (familiarization session and three testing sessions: S1, S2 and S3) with 10-min intersession intervals. During S1, S2 and S3, rats were repeatedly exposed to the same two odor stimuli. During S3, for half of the rats in each treatment group, the spatial locations of the two stimuli were switched (Change) and for the other half the stimuli were replaced in the same locations (No Change). Two habituation scores were measured for each subject: H1 = N1 - N2; H2 = N2 - N3 (Ni the number of contacts made during Si). Baclofen at the highest dose (5 mg/kg) reduced the amount of habituation between S1 and S2 (H1) and increased responses to novel spatial arrangement, measured as the difference between H2 for the No-Change and Change groups. These results suggest a simultaneous impairment of recognition memory and enhancement of spatial novelty detection.
Selective suppression of synaptic transmission during learning is proposed as a physiological mechanism for combining associative memory function at feedback synapses with self-organization of feedforward synapses in neocortical structures. A computational model demonstrates how selective suppression of feedback transmission allows this combination of synaptic function. During learning, sensory stimuli and the desired response are simultaneously presented as input to the network. Feedforward connections form self-organized representations of input, while suppressed feedback connections learn the transpose of the feedforward connectivity. During recall, suppression of transmission is removed, input activates the self-organized representation, and activity settles into a learned solution to the problem. This computational model can be used for learning of problems which are not linearly separable, including the negative patterning task (the XOR problem). Experiments in brain slice preparations of the rat somatosensory cortex tested whether the combination of self-organization and associative memory function could be provided by cholinergic suppression selective for feedback versus feedforward synapses. The cholinergic agonist carbachol selectively suppressed synaptic potentials elicited by stimulation of layer I (which contains a high percentage of feedback synapses), while having no effect on synaptic potentials elicited by stimulation of layer IV (with a high percentage of afferent and feedforward synapses).
Neuromodulators including acetylcholine, norepinephrine, serotonin, dopamine and a range of peptides alter the processing characteristics of cortical networks through effects on excitatory and inhibitory synaptic transmission, on the adaptation of cortical pyramidal cells, on membrane potential, on the rate of synaptic modification, and on other cortical parameters. Computational models of self-organization and associative memory function in cortical structures such as the hippocampus, piriform cortex and neocortex provide a theoretical framework in which the role of these neuromodulatory effects can be analyzed. Neuromodulators such as acetylcholine and norepinephrine appear to enhance the influence of synapses from afferent fibers arising outside the cortex relative to the synapses of intrinsic and association fibers arising from other cortical pyramidal cells. This provides a continuum between a predominant influence of external stimulation to a predominant influence of internal recall (extrinsic vs. intrinsic). Modulatory influence along this continuum may underlie effects described in terms of learning and memory, signal to noise ratio, and attention.
1. The effect of cholinergic modulation on cortical oscillatory dynamics was studied in a computational model of the piriform (olfactory) cortex. The model included the cholinergic suppression of neuronal adaptation, the cholinergic suppression of intrinsic fiber synaptic transmission, the cholinergic enhancement of interneuron activity, and the cholinergic suppression of inhibitory synaptic transmission. 2. Electroencephalographic (EEG) recordings and field potential recordings from the piriform cortex were modeled with a simplified network in which cortical pyramidal cells were represented by excitatory input/output functions with gain parameters dependent on previous activity. The model incorporated distributed excitatory afferent input and excitatory connections between units. In addition, the model contained two sets of inhibitory units mediating inhibition with different time constants and different reversal potentials. This model can match effectively the patterns of cortical EEG and field potentials, showing oscillatory dynamics in both the gamma (30-80 Hz) and theta (3-10 Hz) frequency range. 3. Cholinergic suppression of neuronal adaptation was modeled by reducing the change in gain associated with previous activity. This caused an increased number of oscillations within the network in response to shock stimulation of the lateral olfactory tract, effectively replicating the effect of carbachol on the field potential response in physiological experiments. 4. Cholinergic suppression of intrinsic excitatory synaptic transmission decreased the prominence of gamma oscillations within the network, allowing theta oscillations to predominate. Coupled with the cholinergic suppression of neuronal adaptation, this caused the network to shift from a nonoscillatory state into an oscillatory state of predominant theta oscillations. This replicates the longer term effect of carbachol in experimental preparations on the EEG potential recorded from the cortex in vivo and from brain-slice preparations of the hippocampus in vitro. Analysis of the model suggests that these oscillations depend upon the time constant of neuronal adaptation rather than the time constant of inhibition or the activity of bursting neurons. 5. Cholinergic modulation may be involved in switching the dynamics of this cortical region between those appropriate for learning and those appropriate for recall. During recall, the spread of activity along intrinsic excitatory connections allows associative memory function, whereas neuronal adaptation prevents the spread of activity between different patterns. During learning, the recall of previously stored patterns is prevented by suppression of intrinsic excitatory connections, whereas the response to the new patterns is enhanced by suppression of neuronal adaptation.
The effect of activation of cholinergic receptors on long-term potentiation (LTP) in rat piriform cortex pyramidal cells was studied using extracellular and intracellular recordings in brain slice preparations. The functional role of this modulation was studied in a realistic network biophysical stimulation. Repetitive stimuli were applied in two paradigms: one in which the recorded cell was held at its resting potential and one in which synaptic activity was superimposed on a depolarizing pulse strong enough to evoke four action potentials. In the absence of cholinergic modulation, stimulation at 5 Hz induced LTP primarily in the second condition (13.7%, n = 6 out of 9, measured at 10 min after tetanus). When stimuli were applied in the presence of the muscarinic agonist carbachol (20 microM), LTP of greater amplitude was induced in both paradigms (resting: 41.5%, n = 11 out of 16, depolarized: 36%, n = 5 out of 7, measured at 10 min after tetanus). Increases in excitatory postsynaptic potential (EPSP) amplitudes in the presence of carbachol were gradual, starting at the time 5 Hz stimuli were applied and continuing until an action potential was evoked synaptically. In the presence of the NMDA receptor antagonist 2-amino-5-phosphonovaleric acid (APV), LTP could not be induced. The muscarinic antagonist atropine also prevented LTP induction in the presence of carbachol. Cholinergic modulation of synaptic plasticity was examined in a previously developed realistic biophysical network simulation. In simulations, use of a gradual rate of synaptic modification prevented excessive strengthening of synapses, which could cause interference between stored patterns. The effect of excess synaptic strengthening can be avoided by introducing activity dependent depression of synaptic strength. Coactivation of learning and depression rules results in a stable system where no interference occurs, at any rate of learning. Implementing the depression rule only during recall does not improve the network's performance. This implies that reduction in the strength of synaptic connections should occur in the presence of ACh, more than in normal conditions. We propose that two effects of ACh--enhancement of LTP and enhancement of LTD--should act together to increase the stability of the cortical network in the process of acquiring information.