Alexander S. Ecker

Decorrelated Neuronal Firing in Cortical Microcircuits

Columns, Connections, and Correlations What is the nature of interactions between neurons in neural circuits? The prevalent hypothesis suggests that dense local connectivity causes nearby cortical neurons to receive substantial amounts of common input, which in turn leads to strong correlations between them. Now two studies challenge this view, which impacts our fundamental understanding of coding in the cortex. Ecker et al. (p. 584) investigated the statistics of correlated firing in pairs of neurons from area V1 of awake macaque monkeys. In contrast to previous studies, correlations turned out to be very low, irrespective of the stimulus being shown to the animals, the distances of the recording sites, and the similarity of the neurons receptive fields or response properties. In an accompanying modeling and recording paper, Renart et al. (p. 587) demonstrate how it is possible to have zero noise correlation, even among cells with common input. Despite dense connectivity and shared input, the firing rates of nearby neurons are largely uncorrelated. Correlated trial-to-trial variability in the activity of cortical neurons is thought to reflect the functional connectivity of the circuit. Many cortical areas are organized into functional columns, in which neurons are believed to be densely connected and to share common input. Numerous studies report a high degree of correlated variability between nearby cells. We developed chronically implanted multitetrode arrays offering unprecedented recording quality to reexamine this question in the primary visual cortex of awake macaques. We found that even nearby neurons with similar orientation tuning show virtually no correlated variability. Our findings suggest a refinement of current models of cortical microcircuit architecture and function: Either adjacent neurons share only a few percent of their inputs or, alternatively, their activity is actively decorrelated.

Image Style Transfer Using Convolutional Neural Networks

Rendering the semantic content of an image in different styles is a difficult image processing task. Arguably, a major limiting factor for previous approaches has been the lack of image representations that explicitly represent semantic information and, thus, allow to separate image content from style. Here we use image representations derived from Convolutional Neural Networks optimised for object recognition, which make high level image information explicit. We introduce A Neural Algorithm of Artistic Style that can separate and recombine the image content and style of natural images. The algorithm allows us to produce new images of high perceptual quality that combine the content of an arbitrary photograph with the appearance of numerous wellknown artworks. Our results provide new insights into the deep image representations learned by Convolutional Neural Networks and demonstrate their potential for high level image synthesis and manipulation.

Principles of connectivity among morphologically defined cell types in adult neocortex

A census of neocortical neurons Despite the importance of the brains neocortex, we still do not completely understand the diversity and functional connections of its cell types. Jiang et al. recorded, labeled, and classified over 1200 interneurons and more than 400 pyramidal neurons in the mature mouse visual cortex. Fifteen major classes of interneurons fell into three types: some connect to all neurons, some connect to other interneurons, and some form synapses with pyramidal neurons. Science, this issue p. 10.1126/science.aac9462 The connections between more than 10,000 pairs of individually classified neurons in the visual cortex of adult mice are mapped. INTRODUCTION The intricate microcircuitry of the cerebral cortex is thought to be a critical substrate from which arise the impressive capabilities of the mammalian brain. Until now, our knowledge of the stereotypical connectivity in neocortical microcircuits has been pieced together from individual studies of the connectivity between small numbers of neuronal cell types. Here, we provide unbiased, large-scale profiling of neuronal cell types and connections to reveal the essential building blocks of the cortex and the principles governing their assembly into cortical circuits. Using advanced techniques for tissue slicing, multiple simultaneous whole-cell recording, and morphological reconstruction, we are able to provide a comprehensive view of the connectivity between diverse types of neurons, particularly among types of γ-aminobutyric acid–releasing (GABAergic) interneurons, in the adult animal. RATIONALE We took advantage of a method for preparing high-quality slices of adult tissue and combined this technique with octuple simultaneous, whole-cell recordings followed by an improved staining method that allowed detailed recovery of axonal and dendritic arbor morphology. These data allowed us to perform a census of morphologically and electrophysiologically defined neuronal types (primarily GABAergic interneurons) in neocortical layers 1, 2/3, and 5 (L1, L23, and L5, respectively) and to observe their connectivity patterns in adult animals. RESULTS Our large-scale, comprehensive profiling of neocortical neurons differentiated 15 major types of interneurons, in addition to two lamina-defined types of pyramidal neurons (L23 and L5). Cortical interneurons comprise two types in L1 (eNGC and SBC-like), seven in L23 (L23MC, L23NGC, BTC, BPC, DBC, L23BC, and ChC), and six in L5 (L5MC, L5NGC, L5BC, SC, HEC, and DC) (see the figure). Each type has stereotypical electrophysiological properties and morphological features and can be differentiated from all others by cell type–specific axonal geometry and axonal projection patterns. Importantly, each type of neuron has its own characteristic input-output connectivity profile, connecting with other constituent neuronal types with varying degrees of specificity in postsynaptic targets, laminar location, and synaptic characteristics. Despite specific connection patterns for each cell type, we found that a small number of simple connectivity motifs are repeated across layers and cell types defining a canonical cortical microcircuit. CONCLUSION Our comprehensive profiling of neuronal cell types and connections in adult neocortex provides the most complete wiring diagram of neocortical microcircuits to date. Compared with current genetic labels for cell class, which paint the cortex in broad strokes, our analysis of morphological and electrophysiological properties revealed new cell classes and allowed us to derive a small number of simple connectivity rules that were repeated across layers and cell types. This detailed blueprint of cortical wiring should aid efforts to identify specific circuit abnormalities in animal models of brain disease and may eventually provide a path toward the development of comprehensive circuit-based, cell type–specific interventions. Connectivity among morphologically defined cell types in adult neocortex. (A) Simultaneous octuple whole-cell recording to study connectivity followed by morphological reconstruction. (B) Synaptic connectivity between morphologically distinct types of neurons, including pyramidal (P) neurons

State Dependence of Noise Correlations in Macaque Primary Visual Cortex

Shared, trial-to-trial variability in neuronal populations has a strong impact on the accuracy of information processing in the brain. Estimates of the level of such noise correlations are diverse, ranging from 0.01 to 0.4, with little consensus on which factors account for these differences. Here we addressed one important factor that varied across studies, asking how anesthesia affects the population activity structure in macaque primary visual cortex. We found that under opioid anesthesia, activity was dominated by strong coordinated fluctuations on a timescale of 1-2 Hz, which were mostly absent in awake, fixating monkeys. Accounting for these global fluctuations markedly reduced correlations under anesthesia, matching those observed during wakefulness and reconciling earlier studies conducted under anesthesia and in awake animals. Our results show that internal signals, such as brain state transitions under anesthesia, can induce noise correlations but can also be estimated and accounted for based on neuronal population activity.

Generating spike trains with specified correlation coefficients

Spike trains recorded from populations of neurons can exhibit substantial pairwise correlations between neurons and rich temporal structure. Thus, for the realistic simulation and analysis of neural systems, it is essential to have efficient methods for generating artificial spike trains with specified correlation structure. Here we show how correlated binary spike trains can be simulated by means of a latent multivariate gaussian model. Sampling from the model is computationally very efficient and, in particular, feasible even for large populations of neurons. The entropy of the model is close to the theoretical maximum for a wide range of parameters. In addition, this framework naturally extends to correlations over time and offers an elegant way to model correlated neural spike counts with arbitrary marginal distributions.

Spike sorting for large, dense electrode arrays

Developments in microfabrication technology have enabled the production of neural electrode arrays with hundreds of closely spaced recording sites, and electrodes with thousands of sites are under development. These probes in principle allow the simultaneous recording of very large numbers of neurons. However, use of this technology requires the development of techniques for decoding the spike times of the recorded neurons from the raw data captured from the probes. Here we present a set of tools to solve this problem, implemented in a suite of practical, user-friendly, open-source software. We validate these methods on data from the cortex, hippocampus and thalamus of rat, mouse, macaque and marmoset, demonstrating error rates as low as 5%.

The Effect of Noise Correlations in Populations of Diversely Tuned Neurons

The amount of information encoded by networks of neurons critically depends on the correlation structure of their activity. Neurons with similar stimulus preferences tend to have higher noise correlations than others. In homogeneous populations of neurons, this limited range correlation structure is highly detrimental to the accuracy of a population code. Therefore, reduced spike count correlations under attention, after adaptation, or after learning have been interpreted as evidence for a more efficient population code. Here, we analyze the role of limited range correlations in more realistic, heterogeneous population models. We use Fisher information and maximum-likelihood decoding to show that reduced correlations do not necessarily improve encoding accuracy. In fact, in populations with more than a few hundred neurons, increasing the level of limited range correlations can substantially improve encoding accuracy. We found that this improvement results from a decrease in noise entropy that is associated with increasing correlations if the marginal distributions are unchanged. Surprisingly, for constant noise entropy and in the limit of large populations, the encoding accuracy is independent of both structure and magnitude of noise correlations.

Comparing the feature selectivity of the gamma-band of the local field potential and the underlying spiking activity in primate visual cortex

The local field potential (LFP), comprised of low-frequency extra-cellular voltage fluctuations, has been used extensively to study the mechanisms of brain function. In particular, oscillations in the gamma-band (30–90 Hz) are ubiquitous in the cortex of many species during various cognitive processes. Surprisingly little is known about the underlying biophysical processes generating this signal. Here, we examine the relationship of the local field potential to the activity of localized populations of neurons by simultaneously recording spiking activity and LFP from the primary visual cortex (V1) of awake, behaving macaques. The spatial organization of orientation tuning and ocular dominance in this area provides an excellent opportunity to study this question, because orientation tuning is organized at a scale around one order of magnitude finer than the size of ocular dominance columns. While we find a surprisingly weak correlation between the preferred orientation of multi-unit activity and gamma-band LFP recorded on the same tetrode, there is a strong correlation between the ocular preferences of both signals. Given the spatial arrangement of orientation tuning and ocular dominance, this leads us to conclude that the gamma-band of the LFP seems to sample an area considerably larger than orientation columns. Rather, its spatial resolution lies at the scale of ocular dominance columns.

Feature selectivity of the gamma-band of the local field potential in primate primary visual cortex.

Extracellular voltage fluctuations (local field potentials, LFPs) reflecting neural mass action are ubiquitous across species and brain regions. Numerous studies have characterized the properties of LFP signals in the cortex to study sensory and motor computations as well as cognitive processes like attention, perception and memory. In addition, its extracranial counterpart – the electroencephalogram – is widely used in clinical applications. However, the link between LFP signals and the underlying activity of local populations of neurons remains largely elusive. Here, we review recent work elucidating the relationship between spiking activity of local neural populations and LFP signals. We focus on oscillations in the gamma-band (30–90 Hz) of the LFP in the primary visual cortex (V1) of the macaque that dominate during visual stimulation. Given that in area V1 much is known about the properties of single neurons and the cortical architecture, it provides an excellent opportunity to study the mechanisms underlying the generation of the LFP.

Population code in mouse V1 facilitates readout of natural scenes through increased sparseness

Neural codes are believed to have adapted to the statistical properties of the natural environment. However, the principles that govern the organization of ensemble activity in the visual cortex during natural visual input are unknown. We recorded populations of up to 500 neurons in the mouse primary visual cortex and characterized the structure of their activity, comparing responses to natural movies with those to control stimuli. We found that higher order correlations in natural scenes induced a sparser code, in which information is encoded by reliable activation of a smaller set of neurons and can be read out more easily. This computationally advantageous encoding for natural scenes was state-dependent and apparent only in anesthetized and active awake animals, but not during quiet wakefulness. Our results argue for a functional benefit of sparsification that could be a general principle governing the structure of the population activity throughout cortical microcircuits.