Changes in GABAB modulation during a theta cycle may be analogous to the fall of temperature during annealing

Cortical oscillations support sampling-based computations in spiking neural networks

Being permanently confronted with an uncertain world, brains have faced evolutionary pressure to represent this uncertainty in order to respond appropriately. Often, this requires visiting multiple interpretations of the available information or multiple solutions to an encountered problem. This gives rise to the so-called mixing problem: since all of these "valid" states represent powerful attractors, but between themselves can be very dissimilar, switching between such states can be difficult. We propose that cortical oscillations can be effectively used to overcome this challenge. By acting as an effective temperature, background spiking activity modulates exploration. Rhythmic changes induced by cortical oscillations can then be interpreted as a form of simulated tempering. We provide a rigorous mathematical discussion of this link and study some of its phenomenological implications in computer simulations. This identifies a new computational role of cortical oscillations and connects them to various phenomena in the brain, such as sampling-based probabilistic inference, memory replay, multisensory cue combination, and place cell flickering.

Modulation of motor cortex inhibition during motor imagery

Motor imagery (MI) is similar to overt movement, engaging common neural substrates and facilitating the corticomotor pathway; however, it does not result in excitatory descending motor output. Transcranial magnetic stimulation (TMS) can be used to assess inhibitory networks in the primary motor cortex via measures of 1-ms short-interval intracortical inhibition (SICI), long-interval intracortical inhibition (LICI), and late cortical disinhibition (LCD). These measures are thought to reflect extrasynaptic GABA_A tonic inhibition, postsynaptic GABA_B inhibition, and presynaptic GABA_B disinhibition, respectively. The behavior of 1-ms SICI, LICI, and LCD during MI has not yet been explored. This study aimed to investigate how 1-ms SICI, LICI, and LCD are modulated during MI and voluntary relaxation (VR) of a target muscle. Twenty-five healthy young adults participated. TMS was used to assess nonconditioned motor evoked potential (MEP) amplitude, 1-ms SICI, 100- (LICI₁₀₀) and 150-ms LICI, and LCD in the right abductor pollicis brevis (APB) and right abductor digiti minimi during rest, MI, and VR of the hand. Compared with rest, MEP amplitudes were facilitated in APB during MI. SICI was not affected by task or muscle. LICI₁₀₀ decreased in both muscles during VR but not MI, whereas LCD was recruited in both muscles during both tasks. This indicates that VR modulates postsynaptic GABA_B inhibition, whereas both tasks modulate presynaptic GABA_B inhibition in a non-muscle-specific way. This study highlights further neurophysiological parallels between actual and imagined movement, which may extend to voluntary relaxation.NEW & NOTEWORTHY This is the first study to investigate how 1-ms short-interval intracortical inhibition, long-interval intracortical inhibition, and late cortical disinhibition are modulated during motor imagery and voluntary muscle relaxation. We present novel findings of decreased 100-ms long-interval intracortical inhibition during voluntary muscle relaxation and increased late cortical disinhibition during both motor imagery and voluntary muscle relaxation.

Optimal recall from bounded metaplastic synapses: predicting functional adaptations in hippocampal area CA3

A venerable history of classical work on autoassociative memory has significantly shaped our understanding of several features of the hippocampus, and most prominently of its CA3 area, in relation to memory storage and retrieval. However, existing theories of hippocampal memory processing ignore a key biological constraint affecting memory storage in neural circuits: the bounded dynamical range of synapses. Recent treatments based on the notion of metaplasticity provide a powerful model for individual bounded synapses; however, their implications for the ability of the hippocampus to retrieve memories well and the dynamics of neurons associated with that retrieval are both unknown. Here, we develop a theoretical framework for memory storage and recall with bounded synapses. We formulate the recall of a previously stored pattern from a noisy recall cue and limited-capacity (and therefore lossy) synapses as a probabilistic inference problem, and derive neural dynamics that implement approximate inference algorithms to solve this problem efficiently. In particular, for binary synapses with metaplastic states, we demonstrate for the first time that memories can be efficiently read out with biologically plausible network dynamics that are completely constrained by the synaptic plasticity rule, and the statistics of the stored patterns and of the recall cue. Our theory organises into a coherent framework a wide range of existing data about the regulation of excitability, feedback inhibition, and network oscillations in area CA3, and makes novel and directly testable predictions that can guide future experiments.

What is the function of hippocampal theta rhythm?--Linking behavioral data to phasic properties of field potential and unit recording data

The extensive physiological data on hippocampal theta rhythm provide an opportunity to evaluate hypotheses about the role of theta rhythm for hippocampal network function. Computational models based on these hypotheses help to link behavioral data with physiological measurements of different variables during theta rhythm. This paper reviews work on network models in which theta rhythm contributes to the following functions: (1) separating the dynamics of encoding and retrieval, (2) enhancing the context-dependent retrieval of sequences, (3) buffering of novel information in entorhinal cortex (EC) for episodic encoding, and (4) timing interactions between prefrontal cortex and hippocampus for memory-guided action selection. Modeling shows how these functional mechanisms are related to physiological data from the hippocampal formation, including (1) the phase relationships of synaptic currents during theta rhythm measured by current source density analysis of electroencephalographic data from region CA1 and dentate gyrus, (2) the timing of action potentials, including the theta phase precession of single place cells during running on a linear track, the context-dependent changes in theta phase precession across trials on each day, and the context-dependent firing properties of hippocampal neurons in spatial alternation (e.g., "splitter cells"), (3) the cholinergic regulation of sustained activity in entorhinal cortical neurons, and (4) the phasic timing of prefrontal cortical neurons relative to hippocampal theta rhythm.

Computational theories on the function of theta oscillations

Neural rhythms can be studied in terms of conditions for their generation, or in terms of their functional significance. The theta oscillation is a particularly prominent rhythm, reported to be present in many brain areas, and related to many important cognitive processes. The generating mechanisms of theta have extensively been studied and reviewed elsewhere; here we discuss ideas that have accumulated over the past decades on the computational roles it may subserve. Theories propose different aspects of theta oscillations as being relevant for their cognitive functions: limit cycle oscillations in neuronal firing rates, subthreshold membrane potential oscillations, periodic modulation of synaptic transmission and plasticity, and phase precession of hippocampal place cells. The relevant experimental data is briefly summarized in the light of these theories. Specific models proposing a function for theta in pattern recognition, memory, sequence learning and navigation are reviewed critically. Difficulties with testing and comparing alternative models are discussed, along with potentially important future research directions in the field.

Hippocampus

The hippocampus serves a critical role in declarative memory--our capacity to recall everyday facts and events. Recent studies using functional brain imaging in humans and neuropsychological analyses of humans and animals with hippocampal damage have revealed some of the elemental cognitive processes mediated by the hippocampus. In addition, recent characterizations of neuronal firing patterns in behaving animals and humans have suggested how neural representations in the hippocampus underlie those elemental cognitive processes in the service of declarative memory.

The hippocampus and disambiguation of overlapping sequences

Recent models of hippocampal function emphasize its potential role in disambiguating sequences of events that compose distinct episodic memories. In this study, rats were trained to distinguish two overlapping sequences of odor choices. The capacity to disambiguate the sequences was measured by the critical odor choice after the overlapping elements of the sequences. When the sequences were presented in rapid alternation, damage to the hippocampus, produced either by infusions of the neurotoxin ibotenic acid or by radiofrequency current, produced a severe deficit, although animals with radiofrequency lesions relearned the task. When the sequences were presented spaced apart and in random order, animals with radiofrequency hippocampal lesions could perform the task. However, they failed when a memory delay was imposed before the critical choice. These findings support the hypothesis that the hippocampus is involved in representing sequences of nonspatial events, particularly when interference between the sequences is high or when animals must remember across a substantial delay preceding items in a current sequence.

Critical role of the hippocampus in memory for sequences of events

Recent models of hippocampal function emphasize the potential role of this brain structure in encoding and retrieving sequences of events that compose episodic memories. Here we show that hippocampal lesions produce a severe and selective impairment in the capacity of rats to remember the sequential ordering of a series of odors, despite an intact capacity to recognize odors that recently occurred. These findings support the hypothesis that hippocampal networks mediate associations between sequential events that constitute elements of an episodic memory.

A proposed function for hippocampal theta rhythm: separate phases of encoding and retrieval enhance reversal of prior learning

The theta rhythm appears in the rat hippocampal electroencephalogram during exploration and shows phase locking to stimulus acquisition. Lesions that block theta rhythm impair performance in tasks requiring reversal of prior learning, including reversal in a T-maze, where associations between one arm location and food reward need to be extinguished in favor of associations between the opposite arm location and food reward. Here, a hippocampal model shows how theta rhythm could be important for reversal in this task by providing separate functional phases during each 100-300 msec cycle, consistent with physiological data. In the model, effective encoding of new associations occurs in the phase when synaptic input from entorhinal cortex is strong and long-term potentiation (LTP) of excitatory connections arising from hippocampal region CA3 is strong, but synaptic currents arising from region CA3 input are weak (to prevent interference from prior learned associations). Retrieval of old associations occurs in the phase when entorhinal input is weak and synaptic input from region CA3 is strong, but when depotentiation occurs at synapses from CA3 (to allow extinction of prior learned associations that do not match current input). These phasic changes require that LTP at synapses arising from region CA3 should be strongest at the phase when synaptic transmission at these synapses is weakest. Consistent with these requirements, our recent data show that synaptic transmission in stratum radiatum is weakest at the positive peak of local theta, which is when previous data show that induction of LTP is strongest in this layer.

GABA(B) presynaptic inhibition has an in vivo time constant sufficiently rapid to allow modulation at theta frequency

Cyclical activity of GABAergic interneurons during theta rhythm could mediate phasic changes in strength of glutamatergic synaptic transmission in the hippocampal formation if presynaptic inhibition from activation of GABA(B) receptors is sufficiently rapid to change within a theta cycle. The experiments described here analyzed the time course of GABA(B) modulation using a heterosynaptic depression paradigm in anesthetized rats at physiological temperatures. Heterosynaptic depression of the slope of evoked potentials decayed with a time constant that would allow significant changes in transmission across different phases of the theta cycle. This heterosynaptic depression was significantly reduced by local infusion of the GABA(B) receptor antagonist CGP55845A.

Initial state randomness improves sequence learning in a model hippocampal network

Randomness can be a useful component of computation. Using a computationally minimal, but still biologically based model of the hippocampus, we evaluate the effects of initial state randomization on learning a cognitive problem that requires this brain structure. Greater randomness of initial states leads to more robust performance in simulations of the cognitive task called transverse patterning, a context-dependent discrimination task that we code as a sequence prediction problem. At the conclusion of training, greater initial randomness during training trials also correlates with increased, repetitive firing of select individual neurons, previously named local context neurons. In essence, such repetitively firing neurons recognize subsequences, and previously their presence has been correlated with solving the transverse patterning problem. A more detailed analysis of the simulations across training trials reveals more about initial state randomization. The beneficial effects of initial state randomization derive from enhanced variation, across training trials, of the sequential states of a network. This greater variation is not uniformly present during training; it is largely restricted to the beginning of training and when novel sequences are introduced. Little such variation occurs after extensive or even moderate amounts of training. We explain why variation is high early in training, but not later. This automatic modulation of the initial-state-driven random variation through state space is reminiscent of simulated annealing where modulated randomization encourages a selectively broad search through state space. In contrast to an annealing schedule, the selective occurrence of such a random search here is an emergent property, and the critical randomization occurs during training rather than testing.

Differences in time course of ACh and GABA modulation of excitatory synaptic potentials in slices of rat hippocampus

Activation of muscarinic receptors and GABA(B) receptors causes presynaptic inhibition of glutamatergic synaptic potentials at excitatory feedback connections in cortical structures. These effects may regulate dynamics in cortical structures, with presynaptic inhibition allowing extrinsic afferent input to dominate during encoding, while the absence of presynaptic inhibition allows stronger excitatory feedback during retrieval or consolidation. However, proposals for a functional role of such modulatory effects strongly depend on the time course of these modulatory effects; how rapidly can they turn off and on? In brain slice preparations of hippocampal region CA1, we have explored the time course of suppression of extracellularly recorded synaptic potentials after pressure pulse application of acetylcholine and GABA. Acetylcholine causes suppression of extracellular potentials with onset time constants between 1 and 2 s, and decay constants ranging between 10 and 20 s, even with very brief injection pulses. GABA causes suppression of extracellular potentials with onset time constants between 0.2 and 0.7 s, and decay time constants that decrease to values shorter than 2 s for very brief injection pulses. These techniques do not give an exact measure of the physiological time course in vivo, but they give a notion of the relative time course of the two modulators. The slow changes due to activation of muscarinic acetylcholine receptors may alter the dynamics of cortical circuits over longer intervals (e.g., between different stages of waking and sleep), setting dynamics appropriate for encoding versus consolidation processes. The faster changes in synaptic potentials caused by GABA could cause changes within each cycle of the theta rhythm, rapidly switching between encoding and retrieval dynamics during exploration.

A model of theta rhythm production in the septal-hippocampal system and its modulation by ascending brain stem pathways

Recent experimental observations have disclosed the existence of a septal-hippocampal feedback circuit, composed of medial septum diagonal band of Broca (ms-dbB) GABAergic projections to the inhibitory interneurons of the hippocampus, and hippocampal GABAergic projections to the ms-dbB, the major targets of which are the GABAergic septo-hippocampal projection cells. We propose that this feedback circuit provides the mechanism for the rhythmic suppression of interneuronal activity in the hippocampus, which is observed as low-level GABAergic-mediated theta activity. We also propose that this circuit may be the mechanism by which ascending brain stem pathways to the ms-dbB, in particular from the reticular formation, can influence hippocampal information processing in response to particular behavioral states, by exercising control over the level and frequency of theta activity in the hippocampus. In support of these proposals, we describe a minimal computational model of the feedback circuit which uses a set of four coupled differential equations describing the average dynamic activity of the populations of excitatory and inhibitory cells involved in the circuit. We demonstrate through simulations the inherently robust 4-6-Hz oscillatory dynamics of the circuit, and show that manipulation of internal connection strengths and external modulatory influences on this circuit changes the dynamics in a way which closely mimics corresponding manipulations in recent neurophysiological experiments investigating theta activity.

Computational modeling of entorhinal cortex

Computational modeling provides a means for linking the physiological and anatomical characteristics of entorhinal cortex at a cellular level to the functional role of this region in behavior. We have developed detailed simulations of entorhinal cortical neurons and networks, with an emphasis on the role of acetylcholine in entorhinal cortical function. Computational modeling suggests that when acetylcholine levels are high, this sets appropriate dynamics for the storage of stimuli during performance of delayed matching tasks. In particular, acetylcholine activates a calcium-sensitive nonspecific cation current which provides an intrinsic cellular mechanism which could maintain neuronal activity across a delay period. Simulations demonstrate how this phenomena could underlie entorhinal cortex delay activity as described in previous unit recordings. Acetylcholine also induces theta rhythm oscillations which may be appropriate for timing of afferent input to be encoded in hippocampus and for extraction of individual stored sequences from multiple stored sequences. Lower levels of acetylcholine may allow sharp wave dynamics which can reactivate associations encoded in hippocampus and drive the formation of additional traces in hippocampus and entorhinal cortex during consolidation.

Size of CA1-evoked synaptic potentials is related to theta rhythm phase in rat hippocampus

Cholinergic and GABAergic neurons projecting to the hippocampus fire with specific phase relations to theta rhythm oscillations in the electroencephalogram (EEG). To determine if this phasic input has an impact on synaptic transmission within the hippocampus, we recorded evoked population excitatory postsynaptic potential (EPSPs) during different phases of theta rhythm by using techniques similar to those described in Rudell and Fox. Synaptic potentials elicited by stimulation of region CA3 of the contralateral hippocampus were recorded in region CA1 and CA3. In these experiments, the initial slope of evoked potentials showed a change in magnitude during different phases of the theta rhythm recorded in the dentate fissure, with individual trials showing an average of 9.5% change in slope of potentials, and the average across all experiments showing a change of 7.8%. Evoked potentials were maximal 18 degrees after the positive peak of the dentate fissure theta EEG. These potentials were also smaller by 18.2% during theta as opposed to non-theta states. Phasic changes in modulation of synaptic transmission could contribute to phase precession of hippocampal place cells and could enhance storage of new sequences of activity as demonstrated by computational models.

Neural models of memory

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.

GABA(B) modulation improves sequence disambiguation in computational models of hippocampal region CA3

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.