Sleeping our way to weight normalization and stable learning

Half a century legacy of long-term potentiation

In 1973, two papers from Bliss and Lømo and from Bliss and Gardner-Medwin reported that high-frequency synaptic stimulation in the dentate gyrus of rabbits resulted in a long-lasting increase in synaptic strength. This form of synaptic plasticity, commonly referred to as long-term potentiation (LTP), was immediately considered as an attractive mechanism accounting for the ability of the brain to store information. In this historical piece looking back over the past 50 years, we discuss how these two landmark contributions directly motivated a colossal research effort and detail some of the resulting milestones that have shaped our evolving understanding of the molecular and cellular underpinnings of LTP. We highlight the main features of LTP, cover key experiments that defined its induction and expression mechanisms, and outline the evidence supporting a potential role of LTP in learning and memory. We also briefly explore some ramifications of LTP on network stability, consider current limitations of LTP as a model of associative memory, and entertain future research orientations.

Computational models of memory consolidation and long-term synaptic plasticity during sleep

The brain stores memories by persistently changing the connectivity between neurons. Sleep is known to be critical for these changes to endure. Research on the neurobiology of sleep and the mechanisms of long-term synaptic plasticity has provided data in support of various theories of how brain activity during sleep affects long-term synaptic plasticity. The experimental findings - and therefore the theories - are apparently quite contradictory, with some evidence pointing to a role of sleep in the forgetting of irrelevant memories, whereas other results indicate that sleep supports the reinforcement of the most valuable recollections. A unified theoretical framework is in need. Computational modeling and simulation provide grounds for the quantitative testing and comparison of theoretical predictions and observed data, and might serve as a strategy to organize the rather complicated and diverse pool of data and methodologies used in sleep research. This review article outlines the emerging progress in the computational modeling and simulation of the main theories on the role of sleep in memory consolidation.

Sleep smart-optimizing sleep for declarative learning and memory

The last decade has witnessed a spurt of new publications documenting sleep's essential contribution to the brains ability to form lasting memories. For the declarative memory domain, slow wave sleep (the deepest sleep stage) has the greatest beneficial effect on the consolidation of memories acquired during preceding wakefulness. The finding that newly encoded memories become reactivated during subsequent sleep fostered the idea that reactivation leads to the strengthening and transformation of the memory trace. According to the active system consolidation account, trace reactivation leads to the redistribution of the transient memory representations from the hippocampus to the long-lasting knowledge networks of the cortex. Apart from consolidating previously learned information, sleep also facilitates the encoding of new memories after sleep, which probably relies on the renormalization of synaptic weights during sleep as suggested by the synaptic homeostasis theory. During wakefulness overshooting potentiation causes an imbalance in synaptic weights that is countered by synaptic downscaling during subsequent sleep. This review briefly introduces the basic concepts and central findings of the research on sleep and memory, and discusses implications of this lab-based work for everyday applications to make the best possible use of sleep's beneficial effect on learning and memory.

Tuning into diversity of homeostatic synaptic plasticity

Neurons are endowed with the remarkable ability to integrate activity levels over time and tune their excitability such that action potential firing is maintained within a computationally optimal range. These feedback mechanisms, collectively referred to as "homeostatic plasticity", enable neurons to respond and adapt to prolonged alterations in neuronal activity by regulating several determinants of cellular excitability. Perhaps the best-characterized of these homeostatic responses involves the regulation of excitatory glutamatergic transmission. This homeostatic synaptic plasticity (HSP) operates bidirectionally, thus providing a means for neurons to tune cellular excitability in response to either elevations or reductions in net activity. The last decade has seen rapid growth in interest and efforts to understand the mechanistic underpinnings of HSP in part because of the theoretical stabilization that HSP confers to neural network function. Since the initial reports describing HSP in central neurons, innovations in experimental approaches have permitted the mechanistic dissection of this cellular adaptive response and, as a result, key advances have been made in our understanding of the cellular and molecular basis of HSP. Here, we review recent evidence that outline the presence of distinct forms of HSP at excitatory glutamatergic synapses which operate at different sub-cellular levels. We further present theoretical considerations on the potential computational roles afforded by local, synapse-specific homeostatic regulation. This article is part of the Special Issue entitled 'Homeostatic Synaptic Plasticity'.

Transcranial electrical stimulation accelerates human sleep homeostasis

The sleeping brain exhibits characteristic slow-wave activity which decays over the course of the night. This decay is thought to result from homeostatic synaptic downscaling. Transcranial electrical stimulation can entrain slow-wave oscillations (SWO) in the human electro-encephalogram (EEG). A computational model of the underlying mechanism predicts that firing rates are predominantly increased during stimulation. Assuming that synaptic homeostasis is driven by average firing rates, we expected an acceleration of synaptic downscaling during stimulation, which is compensated by a reduced drive after stimulation. We show that 25 minutes of transcranial electrical stimulation, as predicted, reduced the decay of SWO in the remainder of the night. Anatomically accurate simulations of the field intensities on human cortex precisely matched the effect size in different EEG electrodes. Together these results suggest a mechanistic link between electrical stimulation and accelerated synaptic homeostasis in human sleep.

Hebbian plasticity and homeostasis in a model of hypercolumn of the visual cortex

Neurons in the nervous system display a wide variety of plasticity processes. Among them are covariance-based rules and homeostatic plasticity. By themselves, the first ones tend to generate instabilities because of the unbounded potentiation of synapses. The second ones tend to stabilize the system by setting a target for the postsynaptic firing rate. In this work, we analyze the combined effect of these two mechanisms in a simple model of hypercolumn of the visual cortex. We find that the presence of homeostatic plasticity together with nonplastic uniform inhibition stabilizes the effect of Hebbian plasticity. The system can reach nontrivial solutions, where the recurrent intracortical connections are strongly modulated. The modulation is strong enough to generate contrast invariance. Moreover, this state can be reached even beginning from a weakly modulated initial condition.

Sleep function: current questions and new approaches

The mammalian brain oscillates through three distinct global activity states: wakefulness, non-rapid eye movement (NREM) sleep and REM sleep. The regulation and function of these 'vigilance' or 'behavioural' states can be investigated over a broad range of temporal and spatial scales and at different levels of functional organization, i.e. from gene expression to memory, in single neurons, cortical columns or the whole brain and organism. We summarize some basic questions that have arisen from recent approaches in the quest for the functions of sleep. Whereas traditionally sleep was viewed to be regulated through top-down control mechanisms, recent approaches have emphasized that sleep is emerging locally and regulated in a use-dependent (homeostatic) manner. Traditional markers of sleep homeostasis, such as the electroencephalogram slow-wave activity, have been linked to changes in connectivity and plasticity in local neuronal networks. Thus waking experience-induced local network changes may be sensed by the sleep homeostatic process and used to mediate sleep-dependent events, benefiting network stabilization and memory consolidation. Although many questions remain unanswered, the available data suggest that sleep function will best be understood by an analysis which integrates sleep's many functional levels with its local homeostatic regulation.