A Novel Form of Local Plasticity in Tuft Dendrites of Neocortical Somatosensory Layer 5 Pyramidal Neurons

Role of medial prefrontal cortex voltage-dependent potassium 4.3 channels in nicotine-induced enhancement of object recognition memory in male mice

Nicotine has been shown to enhance object recognition memory in the novel object recognition (NOR) test by activating excitatory neurons in the medial prefrontal cortex (mPFC). However, the exact neuronal mechanisms underlying the nicotine-induced activation of mPFC neurons and the resultant memory enhancement remain poorly understood. To address this issue, we performed brain-slice electrophysiology and the NOR test in male C57BL/6J mice. Whole-cell patch-clamp recordings from layer V pyramidal neurons in the mPFC revealed that nicotine augments the summation of evoked excitatory postsynaptic potentials (eEPSPs) and that this effect was suppressed by N-[3,5-Bis(trifluoromethyl)phenyl]-N'-[2,4-dibromo-6-(2H-tetrazol-5-yl)phenyl]urea (NS5806), a voltage-dependent potassium (Kv) 4.3 channel activator. In line with these findings, intra-mPFC infusion of NS5806 suppressed systemically administered nicotine-induced memory enhancement in the NOR test. Additionally, miRNA-mediated knockdown of Kv4.3 channels in mPFC pyramidal neurons enhanced object recognition memory. Furthermore, inhibition of A-type Kv channels by intra-mPFC infusion of 4-aminopyridine was found to enhance object recognition memory, while this effect was abrogated by prior intra-mPFC NS5806 infusion. These results suggest that nicotine augments the summation of eEPSPs via the inhibition of Kv4.3 channels in mPFC layer V pyramidal neurons, resulting in the enhancement of object recognition memory.

A dendritic mechanism for balancing synaptic flexibility and stability

Biological and artificial neural networks learn by modifying synaptic weights, but it is unclear how these systems retain previous knowledge and also acquire new information. Here, we show that cortical pyramidal neurons can solve this plasticity-versus-stability dilemma by differentially regulating synaptic plasticity at distinct dendritic compartments. Oblique dendrites of adult mouse layer 5 cortical pyramidal neurons selectively receive monosynaptic thalamic input, integrate linearly, and lack burst-timing synaptic potentiation. In contrast, basal dendrites, which do not receive thalamic input, exhibit conventional NMDA receptor (NMDAR)-mediated supralinear integration and synaptic potentiation. Congruently, spiny synapses on oblique branches show decreased structural plasticity in vivo. The selective decline in NMDAR activity and expression at synapses on oblique dendrites is controlled by a critical period of visual experience. Our results demonstrate a biological mechanism for how single neurons can safeguard a set of inputs from ongoing plasticity by altering synaptic properties at distinct dendritic domains.

Design of Ultrapotent Genetically Encoded Inhibitors of Kv4.2 for Gating Neural Plasticity

The Kv4.2 potassium channel plays established roles in neuronal excitability, while also being implicated in plasticity. Current means to study the roles of Kv4.2 are limited, motivating us to design a genetically encoded membrane tethered Heteropodatoxin-2 (MetaPoda). We find that MetaPoda is an ultrapotent and selective gating-modifier of Kv4.2. We narrow its site of contact with the channel to two adjacent residues within the voltage sensitive domain (VSD) and, with docking simulations, suggest that the toxin binds the VSD from within the membrane. We also show that MetaPoda does not require an external linker of the channel for its activity. In neurons (obtained from female and male rat neonates), MetaPoda specifically, and potently, inhibits all Kv4 currents, leaving all other A-type currents unaffected. Inhibition of Kv4 in hippocampal neurons does not promote excessive excitability, as is expected from a simple potassium channel blocker. We do find that MetaPoda's prolonged expression (1 week) increases expression levels of the immediate early gene cFos and prevents potentiation. These findings argue for a major role of Kv4.2 in facilitating plasticity of hippocampal neurons. Lastly, we show that our engineering strategy is suitable for the swift engineering of another potent Kv4.2-selective membrane-tethered toxin, Phrixotoxin-1, denoted MetaPhix. Together, we provide two uniquely potent genetic tools to study Kv4.2 in neuronal excitability and plasticity.

Synaptic configuration and reconfiguration in the neocortex are spatiotemporally selective

Brain computation relies on the neural networks. Neurons extend the neurites such as dendrites and axons, and the contacts of these neurites that form chemical synapses are the biological basis of signal transmissions in the central nervous system. Individual neuronal outputs can influence the other neurons within the range of the axonal spread, while the activities of single neurons can be affected by the afferents in their somatodendritic fields. The morphological profile, therefore, binds the functional role each neuron can play. In addition, synaptic connectivity among neurons displays preference based on the characteristics of presynaptic and postsynaptic neurons. Here, the author reviews the "spatial" and "temporal" connection selectivity in the neocortex. The histological description of the neocortical circuitry depends primarily on the classification of cell types, and the development of gene engineering techniques allows the cell type-specific visualization of dendrites and axons as well as somata. Using genetic labeling of particular cell populations combined with immunohistochemistry and imaging at a subcellular spatial resolution, we revealed the "spatial selectivity" of cortical wirings in which synapses are non-uniformly distributed on the subcellular somatodendritic domains in a presynaptic cell type-specific manner. In addition, cortical synaptic dynamics in learning exhibit presynaptic cell type-dependent "temporal selectivity": corticocortical synapses appear only transiently during the learning phase, while learning-induced new thalamocortical synapses persist, indicating that distinct circuits may supervise learning-specific ephemeral synapse and memory-specific immortal synapse formation. The selectivity of spatial configuration and temporal reconfiguration in the neural circuitry may govern diverse functions in the neocortex.

A Biologically Inspired Neural Network Model to Gain Insight Into the Mechanisms of Post-Traumatic Stress Disorder and Eye Movement Desensitization and Reprocessing Therapy

Eye movement desensitization and reprocessing (EMDR) therapy is a well-established therapeutic method to treat post-traumatic stress disorder (PTSD). However, how EMDR exerts its therapeutic action has been studied in many types of research but still needs to be completely understood. This is in part due to limited knowledge of the neurobiological mechanisms underlying EMDR, and in part to our incomplete understanding of PTSD. In order to model PTSD, we used a biologically inspired computational model based on firing rate units, encompassing the cortex, hippocampus, and amygdala. Through the modulation of its parameters, we fitted real data from patients treated with EMDR or classical exposure therapy. This allowed us to gain insights into PTSD mechanisms and to investigate how EMDR achieves trauma remission.

Classical-Contextual Interactions in V1 May Rely on Dendritic Computations

A signature feature of the neocortex is the dense network of horizontal connections (HCs) through which pyramidal neurons (PNs) exchange "contextual" information. In primary visual cortex (V1), HCs are thought to facilitate boundary detection, a crucial operation for object recognition, but how HCs modulate PN responses to boundary cues within their classical receptive fields (CRF) remains unknown. We began by "asking" natural images, through a structured data collection and ground truth labeling process, what function a V1 cell should use to compute boundary probability from aligned edge cues within and outside its CRF. The "answer" was an asymmetric 2-D sigmoidal function, whose nonlinear form provides the first normative account for the "multiplicative" center-flanker interactions previously reported in V1 neurons (Kapadia et al., 1995, 2000; Polat et al., 1998). Using a detailed compartmental model, we then show that this boundary-detecting classical-contextual interaction function can be computed by NMDAR-dependent spatial synaptic interactions within PN dendrites - the site where classical and contextual inputs first converge in the cortex. In additional simulations, we show that local interneuron circuitry activated by HCs can powerfully leverage the nonlinear spatial computing capabilities of PN dendrites, providing the cortex with a highly flexible substrate for integration of classical and contextual information.

Dendritic Excitability and Synaptic Plasticity In Vitro and In Vivo

Much of our understanding of dendritic and synaptic physiology comes from in vitro experimentation, where the afforded mechanical stability and convenience of applying drugs allowed patch-clamping based recording techniques to investigate ion channel distributions, their gating kinetics, and to uncover dendritic integrative and synaptic plasticity rules. However, with current efforts to study these questions in vivo, there is a great need to translate existing knowledge between in vitro and in vivo experimental conditions. In this review, we identify discrepancies between in vitro and in vivo ionic composition of extracellular media and discuss how changes in ionic composition alter dendritic excitability and plasticity induction. Here, we argue that under physiological in vivo ionic conditions, dendrites are expected to be more excitable and the threshold for synaptic plasticity induction to be lowered. Consequently, the plasticity rules described in vitro vary significantly from those implemented in vivo.

Plasticity of olfactory bulb inputs mediated by dendritic NMDA-spikes in rodent piriform cortex

The piriform cortex (PCx) is essential for learning of odor information. The current view postulates that odor learning in the PCx is mainly due to plasticity in intracortical (IC) synapses, while odor information from the olfactory bulb carried via the lateral olfactory tract (LOT) is ‘hardwired.’ Here, we revisit this notion by studying location- and pathway-dependent plasticity rules. We find that in contrast to the prevailing view, synaptic and optogenetically activated LOT synapses undergo strong and robust long-term potentiation (LTP) mediated by only a few local NMDA-spikes delivered at theta frequency, while global spike timing-dependent plasticity (STDP) protocols failed to induce LTP in these distal synapses. In contrast, IC synapses in apical and basal dendrites undergo plasticity with both NMDA-spikes and STDP protocols but to a smaller extent compared with LOT synapses. These results are consistent with a self-potentiating mechanism of odor information via NMDA-spikes that can form branch-specific memory traces of odors that can further associate with contextual IC information via STDP mechanisms to provide cognitive and emotional value to odors.

Unifying Long-Term Plasticity Rules for Excitatory Synapses by Modeling Dendrites of Cortical Pyramidal Neurons

A large number of experiments have indicated that precise spike times, firing rates, and synapse locations crucially determine the dynamics of long-term plasticity induction in excitatory synapses. However, it remains unknown how plasticity mechanisms of synapses distributed along dendritic trees cooperate to produce the wide spectrum of outcomes for various plasticity protocols. Here, we propose a four-pathway plasticity framework that is well grounded in experimental evidence and apply it to a biophysically realistic cortical pyramidal neuron model. We show in computer simulations that several seemingly contradictory experimental landmark studies are consistent with one unifying set of mechanisms when considering the effects of signal propagation in dendritic trees with respect to synapse location. Our model identifies specific spatiotemporal contributions of dendritic and axo-somatic spikes as well as of subthreshold activation of synaptic clusters, providing a unified parsimonious explanation not only for rate and timing dependence but also for location dependence of synaptic changes.

•

A phenomenological synaptic plasticity rule is applied to a pyramidal neuron model

•

Model reproduces rate-, timing-, and location-dependent plasticity results

•

Active dendrites allow plasticity via dendritic spikes and subthreshold events

•

Cooperative plasticity exists across the dendritic tree and within single branches

Experience and sleep-dependent synaptic plasticity: from structure to activity

Synaptic plasticity is important for learning and memory. With increasing evidence linking sleep states to changes in synaptic strength, an emerging view is that sleep promotes learning and memory by facilitating experience-induced synaptic plasticity. In this review, we summarize the recent progress on the function of sleep in regulating cortical synaptic plasticity. Specifically, we outline the electroencephalogram signatures of sleep states (e.g. slow-wave sleep, rapid eye movement sleep, spindles), sleep state-dependent changes in gene and synaptic protein expression, synaptic morphology, and neuronal and network activity. We highlight studies showing that post-experience sleep potentiates experience-induced synaptic changes and discuss the potential mechanisms that may link sleep-related brain activity to synaptic structural remodelling. We conclude that both synapse formation or strengthening and elimination or weakening occur across sleep. This sleep-dependent synaptic plasticity plays an important role in neuronal circuit refinement during development and after learning, while sleep disorders may contribute to or exacerbate the development of common neurological diseases. This article is part of the Theo Murphy meeting issue 'Memory reactivation: replaying events past, present and future'.

High and asymmetric somato-dendritic coupling of V1 layer 5 neurons independent of visual stimulation and locomotion

Active dendrites impact sensory processing and behaviour. However, it remains unclear how active dendritic integration relates to somatic output in vivo. We imaged semi-simultaneously GCaMP6s signals in the soma, trunk and distal tuft dendrites of layer 5 pyramidal neurons in the awake mouse primary visual cortex. We found that apical tuft signals were dominated by widespread, highly correlated calcium transients throughout the tuft. While these signals were highly coupled to trunk and somatic transients, the frequency of calcium transients was found to decrease in a distance-dependent manner from soma to tuft. Ex vivo recordings suggest that low-frequency back-propagating action potentials underlie the distance-dependent loss of signals, while coupled somato-dendritic signals can be triggered by high-frequency somatic bursts or strong apical tuft depolarization. Visual stimulation and locomotion increased neuronal activity without affecting somato-dendritic coupling. High, asymmetric somato-dendritic coupling is therefore a widespread feature of layer 5 neurons activity in vivo.

Endocannabinoid-Dependent Long-Term Potentiation of Synaptic Transmission at Rat Barrel Cortex

Brain-derived neurotrophic factor (BDNF) plays a critical role in modulating plasticity in sensory cortices. Indeed, a BDNF-dependent long-term potentiation (LTP) at distal basal excitatory synapses of Layer 5 pyramidal neurons (L5PNs) has been demonstrated in disinhibited rat barrel cortex slices. Although it is well established that this LTP requires the pairing of excitatory postsynaptic potentials (PSPs) with Ca2+ spikes, its induction when synaptic inhibition is working remains unexplored. Here we show that low-frequency stimulation at basal dendrites of L5PNs is able to trigger a PSP followed by an action potential (AP) and a slow depolarization (termed PSP-Ca2+ response) in thalamocortical slices without blocking synaptic inhibition. We demonstrate that AP barrage-mediated release of endocannabinoids (eCBs) from the recorded L5PNs induces PSP-Ca2+ response facilitation and BDNF-dependent LTP. Indeed, this LTP requires the type 1 cannabinoid receptors activation, is prevented by postsynaptic intracellular 1,2-bis(2-aminophenoxy) ethane-N,N,N,N'-tetraacetic acid (BAPTA) or the anandamide membrane transporter inhibitor AM404, and only occurs in L5PNs neurons showing depolarization-induced suppression of inhibition. Additionally, electrical stimulation at the posteromedial thalamic nucleus induced similar response and LTP. These results reveal a novel form of eCB-dependent LTP at L5PNs that could be relevant in the processing of sensory information in the barrel cortex.

How Dendrites Affect Online Recognition Memory

In order to record the stream of autobiographical information that defines our unique personal history, our brains must form durable memories from single brief exposures to the patterned stimuli that impinge on them continuously throughout life. However, little is known about the computational strategies or neural mechanisms that underlie the brain's ability to perform this type of "online" learning. Based on increasing evidence that dendrites act as both signaling and learning units in the brain, we developed an analytical model that relates online recognition memory capacity to roughly a dozen dendritic, network, pattern, and task-related parameters. We used the model to determine what dendrite size maximizes storage capacity under varying assumptions about pattern density and noise level. We show that over a several-fold range of both of these parameters, and over multiple orders-of-magnitude of memory size, capacity is maximized when dendrites contain a few hundred synapses-roughly the natural number found in memory-related areas of the brain. Thus, in comparison to entire neurons, dendrites increase storage capacity by providing a larger number of better-sized learning units. Our model provides the first normative theory that explains how dendrites increase the brain's capacity for online learning; predicts which combinations of parameter settings we should expect to find in the brain under normal operating conditions; leads to novel interpretations of an array of existing experimental results; and provides a tool for understanding which changes associated with neurological disorders, aging, or stress are most likely to produce memory deficits-knowledge that could eventually help in the design of improved clinical treatments for memory loss.

Primed to Sleep: The Dynamics of Synaptic Plasticity Across Brain States

It is commonly accepted that brain plasticity occurs in wakefulness and sleep. However, how these different brain states work in concert to create long-lasting changes in brain circuitry is unclear. Considering that wakefulness and sleep are profoundly different brain states on multiple levels (e.g., cellular, molecular and network activation), it is unlikely that they operate exactly the same way. Rather it is probable that they engage different, but coordinated, mechanisms. In this article we discuss how plasticity may be divided across the sleep-wake cycle, and how synaptic changes in each brain state are linked. Our working model proposes that waking experience triggers short-lived synaptic events that are necessary for transient plastic changes and mark (i.e., 'prime') circuits and synapses for further processing in sleep. During sleep, synaptic protein synthesis at primed synapses leads to structural changes necessary for long-term information storage.

Potassium Channel Gain of Function in Epilepsy: An Unresolved Paradox

Exome and targeted sequencing have revolutionized clinical diagnosis. This has been particularly striking in epilepsy and neurodevelopmental disorders, for which new genes or new variants of preexisting candidate genes are being continuously identified at increasing rates every year. A surprising finding of these efforts is the recognition that gain of function potassium channel variants are actually associated with certain types of epilepsy, such as malignant migrating partial seizures of infancy or early-onset epileptic encephalopathy. This development has been difficult to understand as traditionally potassium channel loss-of-function, not gain-of-function, has been associated with hyperexcitability disorders. In this article, we describe the current state of the field regarding the gain-of-function potassium channel variants associated with epilepsy (KCNA2, KCNB1, KCND2, KCNH1, KCNH5, KCNJ10, KCNMA1, KCNQ2, KCNQ3, and KCNT1) and speculate on the possible cellular mechanisms behind the development of seizures and epilepsy in these patients. Understanding how potassium channel gain-of-function leads to epilepsy will provide new insights into the inner working of neural circuits and aid in developing new therapies.

Timed Synaptic Inhibition Shapes NMDA Spikes, Influencing Local Dendritic Processing and Global I/O Properties of Cortical Neurons

The NMDA spike is a long-lasting nonlinear phenomenon initiated locally in the dendritic branches of a variety of cortical neurons. It plays a key role in synaptic plasticity and in single-neuron computations. Combining dynamic system theory and computational approaches, we now explore how the timing of synaptic inhibition affects the NMDA spike and its associated membrane current. When impinging on its early phase, individual inhibitory synapses strongly, but transiently, dampen the NMDA spike; later inhibition prematurely terminates it. A single inhibitory synapse reduces the NMDA-mediated Ca²⁺ current, a key player in plasticity, by up to 45%. NMDA spikes in distal dendritic branches/spines are longer-lasting and more resilient to inhibition, enhancing synaptic plasticity at these branches. We conclude that NMDA spikes are highly sensitive to dendritic inhibition; sparse weak inhibition can finely tune synaptic plasticity both locally at the dendritic branch level and globally at the level of the neuron's output.

The Roles of Cortical Slow Waves in Synaptic Plasticity and Memory Consolidation

Sleep plays important roles in sensory and motor memory consolidation. Sleep oscillations, reflecting neural population activity, involve the reactivation of learning-related neurons and regulate synaptic strength and, thereby affect memory consolidation. Among sleep oscillations, slow waves (0.5–4 Hz) are closely associated with memory consolidation. For example, slow-wave power is regulated in an experience-dependent manner and correlates with acquired memory. Furthermore, manipulating slow waves can enhance or impair memory consolidation. During slow wave sleep, inter-areal interactions between the cortex and hippocampus (HC) have been proposed to consolidate declarative memory; however, interactions for non-declarative (HC-independent) memory remain largely uninvestigated. We recently showed that the directional influence in a slow-wave range through a top-down cortical long-range circuit is involved in the consolidation of non-declarative memory. At the synaptic level, the average cortical synaptic strength is known to be potentiated during wakefulness and depressed during sleep. Moreover, learning causes plasticity in a subset of synapses, allocating memory to them. Sleep may help to differentiate synaptic strength between allocated and non-allocated synapses (i.e., improving the signal-to-noise ratio, which may facilitate memory consolidation). Herein, we offer perspectives on inter-areal interactions and synaptic plasticity for memory consolidation during sleep.

Modeling somatic and dendritic spike mediated plasticity at the single neuron and network level

Synaptic plasticity is thought to be the principal neuronal mechanism underlying learning. Models of plastic networks typically combine point neurons with spike-timing-dependent plasticity (STDP) as the learning rule. However, a point neuron does not capture the local non-linear processing of synaptic inputs allowed for by dendrites. Furthermore, experimental evidence suggests that STDP is not the only learning rule available to neurons. By implementing biophysically realistic neuron models, we study how dendrites enable multiple synaptic plasticity mechanisms to coexist in a single cell. In these models, we compare the conditions for STDP and for synaptic strengthening by local dendritic spikes. We also explore how the connectivity between two cells is affected by these plasticity rules and by different synaptic distributions. Finally, we show that how memory retention during associative learning can be prolonged in networks of neurons by including dendrites.

Synaptic plasticity is the neuronal mechanism underlying learning. Here the authors construct biophysical models of pyramidal neurons that reproduce observed plasticity gradients along the dendrite and show that dendritic spike dependent LTP which is predominant in distal sections can prolong memory retention.

Modelling plasticity in dendrites: from single cells to networks

One of the key questions in neuroscience is how our brain self-organises to efficiently process information. To answer this question, we need to understand the underlying mechanisms of plasticity and their role in shaping synaptic connectivity. Theoretical neuroscience typically investigates plasticity on the level of neural networks. Neural network models often consist of point neurons, completely neglecting neuronal morphology for reasons of simplicity. However, during the past decades it became increasingly clear that inputs are locally processed in the dendrites before they reach the cell body. Dendritic properties enable local interactions between synapses and location-dependent modulations of inputs, rendering the position of synapses on dendrites highly important. These insights changed our view of neurons, such that we now think of them as small networks of nearly independent subunits instead of a simple point. Here, we propose that understanding how the brain processes information strongly requires that we consider the following properties: which plasticity mechanisms are present in the dendrites and how do they enable the self-organisation of synapses across the dendritic tree for efficient information processing? Ultimately, dendritic plasticity mechanisms can be studied in networks of neurons with dendrites, possibly uncovering unknown mechanisms that shape the connectivity in our brains.

Synaptic plasticity in dendrites: complications and coping strategies

The elaborate morphology, nonlinear membrane mechanisms and spatiotemporally varying synaptic activation patterns of dendrites complicate the expression, compartmentalization and modulation of synaptic plasticity. To grapple with this complexity, we start with the observation that neurons in different brain areas face markedly different learning problems, and dendrites of different neuron types contribute to the cell's input-output function in markedly different ways. By committing to specific assumptions regarding a neuron's learning problem and its input-output function, specific inferences can be drawn regarding the synaptic plasticity mechanisms and outcomes that we 'ought' to expect for that neuron. Exploiting this assumption-driven approach can help both in interpreting existing experimental data and designing future experiments aimed at understanding the brain's myriad learning processes.

Bidirectional Hebbian Plasticity Induced by Low-Frequency Stimulation in Basal Dendrites of Rat Barrel Cortex Layer 5 Pyramidal Neurons

According to Hebb's original hypothesis (Hebb, 1949), synapses are reinforced when presynaptic activity triggers postsynaptic firing, resulting in long-term potentiation (LTP) of synaptic efficacy. Long-term depression (LTD) is a use-dependent decrease in synaptic strength that is thought to be due to synaptic input causing a weak postsynaptic effect. Although the mechanisms that mediate long-term synaptic plasticity have been investigated for at least three decades not all question have as yet been answered. Therefore, we aimed at determining the mechanisms that generate LTP or LTD with the simplest possible protocol. Low-frequency stimulation of basal dendrite inputs in Layer 5 pyramidal neurons of the rat barrel cortex induces LTP. This stimulation triggered an EPSP, an action potential (AP) burst, and a Ca²⁺ spike. The same stimulation induced LTD following manipulations that reduced the Ca²⁺ spike and Ca²⁺ signal or the AP burst. Low-frequency whisker deflections induced similar bidirectional plasticity of action potential evoked responses in anesthetized rats. These results suggest that both in vitro and in vivo similar mechanisms regulate the balance between LTP and LTD. This simple induction form of bidirectional hebbian plasticity could be present in the natural conditions to regulate the detection, flow, and storage of sensorimotor information.

Toward an Integration of Deep Learning and Neuroscience

Neuroscience has focused on the detailed implementation of computation, studying neural codes, dynamics and circuits. In machine learning, however, artificial neural networks tend to eschew precisely designed codes, dynamics or circuits in favor of brute force optimization of a cost function, often using simple and relatively uniform initial architectures. Two recent developments have emerged within machine learning that create an opportunity to connect these seemingly divergent perspectives. First, structured architectures are used, including dedicated systems for attention, recursion and various forms of short- and long-term memory storage. Second, cost functions and training procedures have become more complex and are varied across layers and over time. Here we think about the brain in terms of these ideas. We hypothesize that (1) the brain optimizes cost functions, (2) the cost functions are diverse and differ across brain locations and over development, and (3) optimization operates within a pre-structured architecture matched to the computational problems posed by behavior. In support of these hypotheses, we argue that a range of implementations of credit assignment through multiple layers of neurons are compatible with our current knowledge of neural circuitry, and that the brain's specialized systems can be interpreted as enabling efficient optimization for specific problem classes. Such a heterogeneously optimized system, enabled by a series of interacting cost functions, serves to make learning data-efficient and precisely targeted to the needs of the organism. We suggest directions by which neuroscience could seek to refine and test these hypotheses.