David Beniaguev’s research while affiliated with Hebrew University of Jerusalem and other places

Humans exhibit unique cognitive abilities within the animal kingdom, but the neural mechanisms driving these advanced capabilities remain poorly understood. Human cortical neurons differ from those of other species, such as rodents, in both their morphological and physiological characteristics. Could the distinct properties of human cortical neurons help explain the superior cognitive capabilities of humans? Understanding this relationship requires a metric to quantify how neuronal properties contribute to the functional complexity of single neurons, yet no such standardized measure currently exists. Here, we propose the Functional Complexity Index (FCI), a generalized, deep learning-based framework to assess the input-output complexity of neurons. By comparing the FCI of cortical pyramidal neurons from different layers in rats and humans, we identified key morpho-electrical factors that underlie functional complexity. Human cortical pyramidal neurons were found to be significantly more functionally complex than their rat counterparts, primarily due to differences in dendritic membrane area and branching pattern, as well as density and nonlinearity of NMDA-mediated synaptic receptors. These findings reveal the structural-biophysical basis for the enhanced functional properties of human neurons.

Dynamic changes in sensory representations have been basic tenants of studies in neural coding and plasticity. In olfaction, relatively little is known about the dynamic range of changes in odor representations under different brain states and over time. Here, we used time-lapse in vivo two-photon calcium imaging to describe changes in odor representation by mitral cells, the output neurons of the mouse olfactory bulb. Using anesthetics as a gross manipulation to switch between different brain states (wakefulness and under anesthesia), we found that odor representations by mitral cells undergo significant re-shaping across states but not over time within state. Odor representations were well balanced across the population in the awake state yet highly diverse under anesthesia. To evaluate differences in odor representation across states, we used linear classifiers to decode odor identity in one state based on training data from the other state. Decoding across states resulted in nearly chance-level accuracy. In contrast, repeating the same procedure for data recorded within the same state but in different time points, showed that time had a rather minor impact on odor representations. Relative to the differences across states, odor representations remained stable over months. Thus, single mitral cells can change dynamically across states but maintain robust representations across months. These findings have implications for sensory coding and plasticity in the mammalian brain.

A cortical neuron typically makes multiple synaptic contacts on the dendrites of a post-synaptic target neuron. The functional implications of this apparent redundancy are unclear. The dendritic location of a synaptic contact affects the time-course of the somatic post-synaptic potential (PSP) due to dendritic cable filtering. Consequently, a single pre-synaptic axonal spike results with a PSP composed of multiple temporal profiles. Here, we developed a "filter-and-fire" (F&F) neuron model that captures these features and show that the memory capacity of this neuron is threefold larger than that of a leaky integrate-and-fire (I&F) neuron, when trained to emit precisely timed output spikes for specific input patterns. Furthermore, the F&F neuron can learn to recognize spatio-temporal input patterns, e.g., MNIST digits, where the I&F model completely fails. Multiple synaptic contacts between pairs of cortical neurons are therefore an important feature rather than a bug and can serve to reduce axonal wiring requirements.

Utilizing recent advances in machine learning, we introduce a systematic approach to characterize neurons’ input/output (I/O) mapping complexity. Deep neural networks (DNNs) were trained to faithfully replicate the I/O function of various biophysical models of cortical neurons at millisecond (spiking) resolution. A temporally convolutional DNN with five to eight layers was required to capture the I/O mapping of a realistic model of a layer 5 cortical pyramidal cell (L5PC). This DNN generalized well when presented with inputs widely outside the training distribution. When NMDA receptors were removed, a much simpler network (fully connected neural network with one hidden layer) was sufficient to fit the model. Analysis of the DNNs’ weight matrices revealed that synaptic integration in dendritic branches could be conceptualized as pattern matching from a set of spatiotemporal templates. This study provides a unified characterization of the computational complexity of single neurons and suggests that cortical networks therefore have a unique architecture, potentially supporting their computational power.

We propose a novel approach based on modern deep artificial neural networks (DNNs) for understanding how the morpho-electrical complexity of neurons shapes their input/output (I/O) properties at the millisecond resolution in response to massive synaptic input. The I/O of integrate and fire point neuron is accurately captured by a DNN with a single unit and one hidden layer. A fully connected DNN with one hidden layer faithfully replicated the I/O relationship of a detailed model of Layer 5 cortical pyramidal cell (L5PC) receiving AMPA and GABAA synapses. However, when adding voltage-gated NMDA-conductances, a temporally-convolutional DNN with seven layers was required. Analysis of the DNN filters provides new insights into dendritic processing shaping the I/O properties of neurons. This work proposes a systematic approach for characterizing the functional "depth" of a biological neurons, suggesting that cortical pyramidal neurons and the networks they form are computationally much more powerful than previously assumed.