Nonstationarity presents a variety of challenges for machine learning systems. One surprising pathology which can arise in nonstationary learning problems is plasticity loss, whereby making progress on new learning objectives becomes more difficult as training progresses. Networks which are unable to adapt in response to changes in their environment experience plateaus or even declines in performance in highly non-stationary domains such as reinforcement learning, where the learner must quickly adapt to new information even after hundreds of millions of optimization steps. The loss of plasticity manifests in a cluster of related empirical phenomena which have been identified by a number of recent works, including the primacy bias, implicit under-parameterization, rank collapse, and capacity loss. While this phenomenon is widely observed, it is still not fully understood. This talk will present exciting recent results which shed light on the mechanisms driving the loss of plasticity in a variety of learning problems and survey methods to maintain network plasticity in non-stationary tasks, with a particular focus on deep reinforcement learning.

One of the most prominent features of the human brain is the fabulous size of the cerebral cortex and its intricate folding, both of which emerge during development. Over the last few years, work from my lab has shown that specific cellular and genetic mechanisms play central roles in cortex folding, particularly linked to neural stem and progenitor cells. Key mechanisms include high rates of neurogenesis, high abundance of basal Radial Glia Cells (bRGCs), and neuron migration, all of which are intertwined during development. We have also shown that primary cortical folds follow highly stereotyped patterns, defined by a spatial-temporal protomap of gene expression within germinal layers of the developing cortex. I will present recent findings from my laboratory revealing novel cellular and genetic mechanisms that regulate cortex expansion and folding. We have uncovered the contribution of epigenetic regulation to the establishment of the cortex folding protomap, modulating the expression levels of key transcription factors that control progenitor cell proliferation and cortex folding. At the single cell level, we have identified an unprecedented diversity of cortical progenitor cell classes in the ferret and human embryonic cortex. These are differentially enriched in gyrus versus sulcus regions and establish parallel cell lineages, not observed in mouse. Our findings show that genetic and epigenetic mechanisms in gyrencephalic species diversify cortical progenitor cell types and implement parallel cell linages, driving the expansion of neurogenesis and patterning cerebral cortex folds.

The adult intestine is a major barrier epithelium and coordinator of multi-organ functions. Stem cells constantly repair the intestinal epithelium by adjusting their proliferation and differentiation to tissue intrinsic as well as micro- and macro-environmental signals. How these signals integrate to control intestinal and whole-body homeostasis is largely unknown. Addressing this gap in knowledge is central to an improved understanding of intestinal pathophysiology and its systemic consequences. Combining Drosophila and mammalian model systems my laboratory has discovered fundamental mechanisms driving intestinal regeneration and tumourigenesis and outlined complex inter-organ signaling regulating health and disease. During my talk, I will discuss inter-related areas of research from my lab, including:1- Interactions between the intestine and its microenvironment influencing intestinal regeneration and tumourigenesis. 2- Long-range signals from the intestine impacting whole-body in health and disease.

In this talk, I will focus on recent work that has uncovered the diversity of intrinsically photosensitive retinal ganglion cells (ipRGCs). These are a unique type of retinal ganglion cell that contains the photopigment melanopsin. ipRGCs are the retinal neurons responsible for driving non-imaging forming behaviors and reflexes, such as circadian entrainment and pupil constriction, amongst many others. My lab has recently focused on uncovering the diversity of ipRGCs, their distribution throughout the mammalian retina, and their axon projections in the brain.

Programmed axon death is a widespread and completely preventable mechanism in injury and disease. Mouse and Drosophila studies define a molecular pathway involving activation of SARM1 NA Dase and its prevention by NAD synthesising enzyme NMNAT2 . Loss of axonal NMNAT2 causes its substrate, NMN , to accumulate and activate SARM1 , driving loss of NAD and changes in ATP , ROS and calcium. Animal models caused by genetic mutation, toxins, viruses or metabolic defects can be alleviated by blocking programmed axon death, for example models of CMT1B , chemotherapy-induced peripheral neuropathy (CIPN), rabies and diabetic peripheral neuropathy (DPN). The perinatal lethality of NMNAT2 null mice is completely rescued, restoring a normal, healthy lifespan. Animal models lack the genetic and environmental diversity present in human populations and this is problematic for modelling gene-environment combinations, for example in CIPN and DPN , and identifying rare, pathogenic mutations. Instead, by testing human gene variants in WGS datasets for loss- and gain-of-function, we identified enrichment of rare SARM1 gain-of-function variants in sporadic ALS , despite previous negative findings in SOD1 transgenic mice. We have shown in mice that heterozygous SARM1 loss-of-function is protective from a range of axonal stresses and that naturally-occurring SARM1 loss-of-function alleles are present in human populations. This enables new approaches to identify disorders where blocking SARM1 may be therapeutically useful, and the existence of two dominant negative human variants in healthy adults is some of the best evidence available that drugs blocking SARM1 are likely to be safe. Further loss- and gain-of-function variants in SARM1 and NMNAT2 are being identified and used to extend and strengthen the evidence of association with neurological disorders. We aim to identify diseases, and specific patients, in whom SARM1 -blocking drugs are most likely to be effective.

Focal neocortical epilepsy is associated with intermittent brief population discharges (interictal spikes), which resemble sentinel spikes that often occur at the onset of seizures. Why interictal spikes self-terminate whilst seizures persist and propagate is incompletely understood, but is likely to relate to the intermittent collapse of feed-forward GABAergic inhibition. Inhibition could fail through multiple mechanisms, including (i) an attenuation or even reversal of the driving force for chloride in postsynaptic neurons because of intense activation of GABAA receptors, (ii) an elevation of potassium secondary to chloride influx leading to depolarization of neurons, or (iii) insufficient GABA release from interneurons. I shall describe the results of experiments using fluorescence imaging of calcium, glutamate or GABA in awake rodent models of neocortical epileptiform activity. Interictal spikes were accompanied by brief glutamate transients which were maximal at the initiation site and rapidly propagatedcentrifugally. GABA transients lasted longer than glutamate transients and were maximal ~1.5 mm from the focus. Prior to seizure initiation GABA transients were attenuated, whilst glutamate transients increased, consistent with a progressive failure of local inhibitory restraint. As seizures increased in frequency, there was a gradual increase in the spatial extent of spike-associated glutamate transients associated with interictal spikes. Neurotransmitter imaging thus reveals a progressive collapse of an annulus of feed-forward GABA release, allowing runaway recruitment of excitatory neurons as a fundamental mechanism underlying the escape of seizures from local inhibitory restraint.

Genome-wide association studies (GWAS) have identified multiple disease-associated genetic variations across different psychiatric disorders raising the question of how these genetic variants relate to the corresponding pharmacological treatments. In this talk, I will outline our work investigating whether functional information from a range of open bioinformatics datasets such as protein interaction network (PPI), brain eQTL, and gene expression pattern across the brain can uncover the relationship between GWAS-identified genetic variation and the genes targeted by current drugs for psychiatric disorders. Focusing on four psychiatric disorders---ADHD, bipolar disorder, schizophrenia, and major depressive disorder---we assess relationships between the gene targets of drug treatments and GWAS hits and show that while incorporating information derived from functional bioinformatics data, such as the PPI network and spatial gene expression, can reveal links for bipolar disorder, the overall correspondence between treatment targets and GWAS-implicated genes in psychiatric disorders rarely exceeds null expectations. This relatively low degree of correspondence across modalities suggests that the genetic mechanisms driving the risk for psychiatric disorders may be distinct from the pathophysiological mechanisms used for targeting symptom manifestations through pharmacological treatments and that novel approaches for understanding and treating psychiatric disorders may be required.

An equation of state is something you hear about in introductory thermodynamics, for example, the Ideal gas equation. The ideal gas equation relates the pressure, volume, and the number of particles of the gas, to its temperature, uniquely defining its state. This description is possible in physics when the system under investigation is in equilibrium or near equilibrium. In biology, a tissue is modeled as a fluid composed of cells. These cells are constantly interacting with each other through mechanical and chemical signaling, driving them far from equilibrium. Can an equation of state exist for such a messy interacting system? In this talk, I show that the presence of strong cell-cell interaction in tissues gives rise to a novel non-equilibrium, size-dependent surface tension, something unheard of for classical fluids. This surface tension, in turn, modifies the packing of cells inside the tissue generating a size-dependent density and pressure. Finally, we show that a combination of these non-equilibrium pressure and densities can yield an equation of state for biological tissues arbitrarily far from equilibrium. In the end, I discuss how this new paradigm of size-dependent biological properties gives rise to novel modes of cellular motion in tissues

How does our brain rapidly determine if something is good or bad? How do we know our place within a social group? How do we know how to behave appropriately in dynamic environments with ever-changing conditions? The Tye Lab is interested in understanding how neural circuits important for driving positive and negative motivational valence (seeking pleasure or avoiding punishment) are anatomically, genetically and functionally arranged. We study the neural mechanisms that underlie a wide range of behaviors ranging from learned to innate, including social, feeding, reward-seeking and anxiety-related behaviors. We have also become interested in “social homeostasis” -- how our brains establish a preferred set-point for social contact, and how this maintains stability within a social group. How are these circuits interconnected with one another, and how are competing mechanisms orchestrated on a neural population level? We employ optogenetic, electrophysiological, electrochemical, pharmacological and imaging approaches to probe these circuits during behavior.

Transcriptomics has revealed that cortical inhibitory neurons exhibit a great diversity of fine molecular subtypes, but it is not known whether these subtypes have correspondingly diverse activity patterns in the living brain. We show that inhibitory subtypes in primary visual cortex (V1) have diverse correlates with brain state, but that this diversity is organized by a single factor: position along their main axis of transcriptomic variation. We combined in vivo 2-photon calcium imaging of mouse V1 with a novel transcriptomic method to identify mRNAs for 72 selected genes in ex vivo slices. We classified inhibitory neurons imaged in layers 1-3 into a three-level hierarchy of 5 Subclasses, 11 Types, and 35 Subtypes using previously-defined transcriptomic clusters. Responses to visual stimuli differed significantly only across Subclasses, suppressing cells in the Sncg Subclass while driving cells in the other Subclasses. Modulation by brain state differed at all hierarchical levels but could be largely predicted from the first transcriptomic principal component, which also predicted correlations with simultaneously recorded cells. Inhibitory Subtypes that fired more in resting, oscillatory brain states have less axon in layer 1, narrower spikes, lower input resistance and weaker adaptation as determined in vitro and express more inhibitory cholinergic receptors. Subtypes firing more during arousal had the opposite properties. Thus, a simple principle may largely explain how diverse inhibitory V1 Subtypes shape state-dependent cortical processing.

Active matter describes a class of systems that are maintained far from equilibrium by driving forces acting on the constituent particles. Here I will focus on confined active nematics, which exhibit especially rich flow behavior, ranging from structured patterns in space and time to disordered turbulent flows. To understand this behavior, I will take a deterministic dynamical systems approach, beginning with the hydrodynamic equations for the active nematic. This approach reveals that the infinite-dimensional phase space of all possible flow configurations is populated by Exact Coherent Structures (ECS), which are exact solutions of the hydrodynamic equations with distinct and regular spatiotemporal structure; examples include unstable equilibria, periodic orbits, and traveling waves. The ECS are connected by dynamical pathways called invariant manifolds. The main hypothesis in this approach is that turbulence corresponds to a trajectory meandering in the phase space, transitioning between ECS by traveling on the invariant manifolds. Similar approaches have been successful in characterizing high Reynolds number turbulence of passive fluids. Here, I will present the first systematic study of active nematic ECS and their invariant manifolds and discuss their role in characterizing the phenomenon of active turbulence.

While most experimental tasks aim at isolating simple cognitive processes to study their neural bases, naturalistic behaviour is often complex and multidimensional. I will present two studies revealing previously uncharacterised neural circuits for decision-making in macaques. This was possible thanks to innovative experimental tasks eliciting sophisticated behaviour, bridging the human and non-human primate research traditions. Firstly, I will describe a specialised medial frontal circuit for novel choice in macaques. Traditionally, monkeys receive extensive training before neural data can be acquired, while a hallmark of human cognition is the ability to act in novel situations. I will show how this medial frontal circuit can combine the values of multiple attributes for each available novel item on-the-fly to enable efficient novel choices. This integration process is associated with a hexagonal symmetry pattern in the BOLD response, consistent with a grid-like representation of the space of all available options. We prove the causal role played by this circuit by showing that focussed transcranial ultrasound neuromodulation impairs optimal choice based on attribute integration and forces the subjects to default to a simpler heuristic decision strategy. Secondly, I will present an ongoing project addressing the neural mechanisms driving behaviour shifts during an evidence accumulation task that requires subjects to trade speed for accuracy. While perceptual decision-making in general has been thoroughly studied, both cognitively and neurally, the reasons why speed and/or accuracy are adjusted, and the associated neural mechanisms, have received little attention. We describe two orthogonal dimensions in which behaviour can vary (traditional speed-accuracy trade-off and efficiency) and we uncover independent neural circuits concerned with changes in strategy and fluctuations in the engagement level. The former involves the frontopolar cortex, while the latter is associated with the insula and a network of subcortical structures including the habenula.

Almost everybody that has seen neurons under a microscope for the first time is fascinated by their beauty and their complex shape. Early on during development, however, there are hardly any signs of their future complexity, but the neurons look round and simple. How do neurons develop their sophisticated structure? How do they initially generate domains that later have distinct function within neuronal circuits, such as the axon? And, can a better understanding of the underlying developmental mechanisms help us in pathological conditions, such as a spinal cord injury, to induce axons to regenerate? Here, I will talk about the cytoskeleton as a driving force for neuronal polarization. We will then explore how cytoskeletal changes help to reactivate the growth program of injured CNS axons to elicit axon regeneration after a spinal cord injury. Finally, we will discuss whether axon growth and synapse formation may be processes in neurons that might exclude each other. Following this developmental hypothesis, it will help us to generate a novel perspective on regeneration failure in the adult CNS, and how we can overcome this failure to induce axon regeneration. Thus, this talk will describe how we can exploit developmental mechanisms to induce axon regeneration after a spinal cord injury.

Brain microstructure plays a key role in driving the transport of drug molecules directly administered to the brain tissue as in Convection-Enhanced Delivery procedures. This study reports the first systematic attempt to characterize the cytoarchitecture of commissural, long association and projection fiber, namely: the corpus callosum, the fornix and the corona radiata. Ovine samples from three different subjects have been imaged using scanning electron microscope combined with focused ion beam milling. Particular focus has been given to the axons. For each tract, a 3D reconstruction of relatively large volumes (including a significant number of axons) has been performed. Namely, outer axonal ellipticity, outer axonal cross-sectional area and its relative perimeter have been measured. This study [1] provides useful insight into the fibrous organization of the tissue that can be described as composite material presenting elliptical tortuous tubular fibers, leading to a workflow to enable accurate simulations of drug delivery which include well-resolved microstructural features.  As a demonstration of the use of these imaging and reconstruction techniques, our research analyses the hydraulic permeability of two white matter (WM) areas (corpus callosum and fornix) whose three-dimensional microstructure was reconstructed starting from the acquisition of the electron microscopy images. Considering that the white matter structure is mainly composed of elongated and parallel axons we computed the permeability along the parallel and perpendicular directions using computational fluid dynamics [2]. The results show a statistically significant difference between parallel and perpendicular permeability, with a ratio about 2 in both the white matter structures analysed, thus demonstrating their anisotropic behaviour. This is in line with the experimental results obtained using perfusion of brain matter [3]. Moreover, we find a significant difference between permeability in corpus callosum and fornix, which suggests that also the white matter heterogeneity should be considered when modelling drug transport in the brain. Our findings, that demonstrate and quantify the anisotropic and heterogeneous character of the white matter, represent a fundamental contribution not only for drug delivery modelling but also for shedding light on the interstitial transport mechanisms in the extracellular space. These and many other discoveries will be discussed during the talk." "1. https://www.researchsquare.com/article/rs-686577/v1, 2. https://www.pnas.org/content/118/36/e2105328118, 3. https://ieeexplore.ieee.org/abstract/document/9198110

Far-from-equilibrium processes constantly dissipate energy while converting a free-energy source to another form of energy. Living systems, for example, rely on an orchestra of molecular motors that consume chemical fuel to produce mechanical work. In this talk, I will describe two features of life, namely, time-irreversibility, and nonequilibrium self-assembly. Time irreversibility is the hallmark of nonequilibrium dissipative processes. Detecting dissipation is essential for our basic understanding of the underlying physical mechanism, however, it remains a challenge in the absence of observable directed motion, flows, or fluxes. Additional difficulty arises in complex systems where many internal degrees of freedom are inaccessible to an external observer. I will introduce a novel approach to detect time irreversibility and estimate the entropy production from time-series measurements, even in the absence of observable currents. This method can be implemented in scenarios where only partial information is available and thus provides a new tool for studying nonequilibrium phenomena. Further, I will explore the added benefits achieved by nonequilibrium driving for self-assembly, identify distinctive collective phenomena that emerge in a nonequilibrium self-assembly setting, and demonstrate the interplay between the assembly speed, kinetic stability, and relative population of dynamical attractors.

The ability to reach, grasp, and manipulate objects is a remarkable expression of motor skill, and the loss of this ability in injury, stroke, or disease can be devastating. These behaviors are controlled by the coordinated activity of tens of millions of neurons distributed across many CNS regions, including the primary motor cortex. While many studies have characterized the activity of single cortical neurons during reaching, the principles governing the dynamics of large, distributed neural populations remain largely unknown. Recent work in primates has suggested that during the execution of reaching, motor cortex may autonomously generate the neural pattern controlling the movement, much like the spinal central pattern generator for locomotion. In this seminar, I will describe recent work that tests this hypothesis using large-scale neural recording, high-resolution behavioral measurements, dynamical systems approaches to data analysis, and optogenetic perturbations in mice. We find, by contrast, that motor cortex requires strong, continuous, and time-varying thalamic input to generate the neural pattern driving reaching. In a second line of work, we demonstrate that the cortico-cerebellar loop is not critical for driving the arm towards the target, but instead fine-tunes movement parameters to enable precise and accurate behavior. Finally, I will describe my future plans to apply these experimental and analytical approaches to the adaptive control of locomotion in complex environments.

Predictive learning hypotheses posit that the neocortex learns a hierarchical model of the structure of features in the environment. Under these hypotheses, expected or predictable features are differentiated from unexpected ones by comparing bottom-up and top-down streams of data, with unexpected features then driving changes in the representation of incoming stimuli. This is supported by numerous studies in early sensory cortices showing that pyramidal neurons respond particularly strongly to unexpected stimulus events. However, it remains unknown how their responses govern subsequent changes in stimulus representations, and thus, govern learning. Here, I present results from our study of layer 2/3 and layer 5 pyramidal neurons imaged in primary visual cortex of awake, behaving mice using two-photon calcium microscopy at both the somatic and distal apical planes. Our data reveals that individual neurons and distal apical dendrites show distinct, but predictable changes in unexpected event responses when tracked over several days. Considering existing evidence that bottom-up information is primarily targeted to somata, with distal apical dendrites receiving the bulk of top-down inputs, our findings corroborate hypothesized complementary roles for these two neuronal compartments in hierarchical computing. Altogether, our work provides novel evidence that the neocortex indeed instantiates a predictive hierarchical model in which unexpected events drive learning.

This talk will focus on the idea that the kind of similarity driving analogical retrieval is determined by the kind of features encoded regarding the source and the target cue situations. Emphasis will be put on educational perspectives in order to show the influence of world semantics on learners’ problem representations and solving strategies, as well as the difficulties arising from semantic incongruence between representations and strategies. Special attention will be given to the recoding of semantically incongruent representations, a crucial step that learners struggle with, in order to illustrate a promising path for going beyond informal strategies.

Filaments (slender, microscopic elastic bodies) are prevalent in biological and industrial settings. In the biological case, the filaments are often active, in that they are driven internally by motor proteins, with the prime examples being cilia and flagella. For cilia in particular, which can appear in dense arrays, their resulting motions are coupled through the surrounding fluid, as well as through surfaces to which they are attached. In this talk, I present numerical simulations exploring the coordinated motion of active filaments and how it depends on the driving force, density of filaments, as well as the attached surface. In particular, we find that when the surface is spherical, its topology introduces local defects in coordinated motion which can then feedback and alter the global state. This is particularly true when the surface is not held fixed and is free to move in the surrounding fluid. These simulations take advantage of a computational framework we developed for fully 3D filament motion that combines unit quaternions, implicit geometric time integration, quasi-Newton methods, and fast, matrix-free methods for hydrodynamic interactions and it will also be presented.

Eyes abound in the animal kingdom. Some are large as basketballs and others are just fractions of a millimetre. Eyes also come in many different types, such as the compound eyes of insects, the mirror eyes of scallopsor our own camera-like eyes. Common to all animal eyes is that they serve the same fundamental role of collecting external information for guidingthe animal’s behaviour. But behaviours vary tremendously across the animal kingdom, and it turns outthis is the key to understand how eyes evolved. The lecture will take a tour from the first animals that could only sense the presence of light, to those that saw the first crude image of the world and finally to animals that use acute vision for interacting with otheranimals. Amazingly, all these stages of eye evolution still exist in animals living today, and this is how we can unravel the evolution of behaviours that has been the driving force behind eye evolution

The last decade in the field of neuroscience has been marked by intense debate on the meaning of the term fear. Whereas some have argued that fear (as well as other emotions) relies on cognitive capacities that are unique to humans, others view it as a negative state constructed from essential building blocks. This latter definition posits that fear states are associated with varying readouts that one could consider to be parallel processes or serial events tied to a specific hierarchy. Within this framework, innate defensive behaviors are considered to be common displays of fear states that lie under the control of hard-wired brain circuits. As a general rule, these defensive behaviors can be classified as either reactive or cognitive based on a thread imminence continuum. However, while evidence of the neuronal circuits that lead to these divergent behavioral strategies has accrued over the last decades, most literature has considered these responses in isolation. As a result, important misconceptions have arisen regarding how fear circuits are distributed in the brain and the contribution of specific nodes within these circuits to defensive behaviors. To mitigate the status quo, I will conduct a systematic comparison of brain circuits driving the expression of freezing and active avoidance behavior, which I will use as well-studied proxies of reactive and cognitive fear, respectively. In addition, I propose that by integrating associative information with interoceptive and exteroceptive signals the central nucleus of the amygdala plays a crucial role in biasing the selection of defensive behaviors.

The human brain is a complex organ characterized by heterogeneous patterns of interconnections. Non-invasive imaging techniques now allow for these patterns to be carefully and comprehensively mapped in individual humans, paving the way for a better understanding of how wiring supports cognitive processes. While a large body of work now focuses on descriptive statistics to characterize these wiring patterns, a critical open question lies in how the organization of these networks constrains the potential repertoire of brain dynamics. In this talk, I will describe an approach for understanding how perturbations to brain dynamics propagate through complex wiring patterns, driving the brain into new states of activity. Drawing on a range of disciplinary tools – from graph theory to network control theory and optimization – I will identify control points in brain networks and characterize trajectories of brain activity states following perturbation to those points. Finally, I will describe how these computational tools and approaches can be used to better understand the brain's intrinsic control mechanisms and their alterations in psychiatric conditions.

In vivo, the human genome folds into a characteristic ensemble of 3D structures. The mechanism driving the folding process remains unknown. A theoretical model for chromatin (the minimal chromatin model) explains the folding of interphase chromosomes and generates chromosome conformations consistent with experimental data is presented. The energy landscape of the model was derived by using the maximum entropy principle and relies on two experimentally derived inputs: a classification of loci into chromatin types and a catalog of the positions of chromatin loops. This model was generalized by utilizing a neural network to infer these chromatin types using epigenetic marks present at a locus, as assayed by ChIP-Seq. The ensemble of structures resulting from these simulations completely agree with HI-C data and exhibits unknotted chromosomes, phase separation of chromatin types, and a tendency for open chromatin to lie at the periphery of chromosome territories. Although this theoretical methodology was trained in one cell line, the human GM12878 lymphoblastoid cells, it has successfully predicted the structural ensembles of multiple human cell lines. Finally, going beyond Hi-C, our predicted structures are also consistent with microscopy measurements. Analysis of both structures from simulation and microscopy reveals that short segments of chromatin make two-state transitions between closed conformations and open dumbbell conformations. For gene active segments, the vast majority of genes appear clustered in the linker region of the chromatin segment, allowing us to speculate possible mechanisms by which chromatin structure and dynamics may be involved in controlling gene expression. * Supported by the NSF

Though the term has many definitions, computational neuroscience is mainly about applying mathematics to the study of the brain. The brain—a jumble of all different kinds of neurons interconnected in countless ways that somehow produce consciousness—has been described as “the most complex object in the known universe”. Physicists for centuries have turned to mathematics to properly explain some of the most seemingly simple processes in the universe—how objects fall, how water flows, how the planets move. Equations have proved crucial in these endeavors because they capture relationships and make precise predictions possible. How could we expect to understand the most complex object in the universe without turning to mathematics? — The answer is we can’t, and that is why I wrote this book. While I’ve been studying and working in the field for over a decade, most people I encounter have no idea what “computational neuroscience” is or that it even exists. Yet a desire to understand how the brain works is a common and very human interest. I wrote this book to let people in on the ways in which the brain will ultimately be understood: through mathematical and computational theories. — At the same time, I know that both mathematics and brain science are on their own intimidating topics to the average reader and may seem downright prohibitory when put together. That is why I’ve avoided (many) equations in the book and focused instead on the driving reasons why scientists have turned to mathematical modeling, what these models have taught us about the brain, and how some surprising interactions between biologists, physicists, mathematicians, and engineers over centuries have laid the groundwork for the future of neuroscience. — Each chapter of Models of the Mind covers a separate topic in neuroscience, starting from individual neurons themselves and building up to the different populations of neurons and brain regions that support memory, vision, movement and more. These chapters document the history of how mathematics has woven its way into biology and the exciting advances this collaboration has in store.

The journey of development begins with sperm swimming through the female reproductive tract en-route to the egg. In order to successfully complete this journey sperm must beat a single flagellum, propelling themselves through a wide range of fluids, from liquified semen to viscous cervical mucus. It is well-known that the beating tail is driven by an array of 9 microtubule doublets surrounding a central pair, with interconnecting dynein motors generating shear forces and driving elastic wave propagation. Despite this knowledge, the exact mechanism by which coordination of these motors drives oscillating waves along the flagellum remains unknown; hypothesised mechanisms include curvature control, sliding control, and geometric clutch. In this talk we will discuss the mechanisms of flagellar bending, and present a simple model of active curvature that is able to produce many of the various sperm waveforms that are seen experimentally, including those in low and high viscosity fluids and after a cell has ‘hyperactivated’ (a chemical process thought to be key for fertilization). We will show comparisons between these simulated waveforms and sperm that have been experimentally tracked, and discuss methods for fitting simulated mechanistic parameters to these real cells.

Visual attention refers to a set of processes allowing selection of relevant and filtering out of irrelevant information in the visual environment. A large amount of research on visual attention has involved visual search paradigms, where observers are asked to report whether a single target is present or absent. However, recent studies have revealed that these classic single-target visual search tasks only provide a snapshot of how attention is allocated in the visual environment, and that multitarget visual foraging tasks may capture the dynamics visual attention more accurately. In visual foraging, observers are asked to select multiple instances of multiple target types, as fast as they can. A critical question in foraging research concerns the factors driving the next target selection. Most likely, this would require two steps: (1) identifying a set of candidates for the next selection, and (2) selecting the best option among these candidates. After having briefly described the advantage of visual foraging over visual search, I will review recent visual foraging studies testing the influence of several manipulations (e.g., target crypticity, number of items, selection modality) on foraging behaviour. Overall, these studies revealed that the next target selection during visual foraging is determined by the competition between three factors: target value, target proximity, and priming of features. I will explain how the analysis of individual differences in foraging behaviour can provide important information, with the idea that individuals show by-default internal biases toward value, proximity and priming that determine their search strategy and behaviour.

Dr. Rajan and her lab design neural network models based on experimental data, and reverse-engineer them to figure out how brain circuits function in health and disease. They recently developed a powerful framework for tracing neural paths across multiple brain regions— called Current-Based Decomposition (CURBD). This new approach enables the computation of excitatory and inhibitory input currents that drive a given neuron, aiding in the discovery of how entire populations of neurons behave across multiple interacting brain regions. Dr. Rajan’s team has applied this method to studying the neural underpinnings of behavior. As an example, when CURBD was applied to data gathered from an animal model often used to study depression- and anxiety-like behaviors (i.e., learned helplessness) the underlying biology driving adaptive and maladaptive behaviors in the face of stress was revealed. With this framework Dr. Rajan's team probes for mechanisms at work across brain regions that support both healthy and disease states-- as well as identify key divergences from multiple different nervous systems, including zebrafish, mice, non-human primates, and humans.

Broadly, the Mauro Costa-Mattioli laboratory (The MCM Lab) encompasses two complementary lines of research. The first one, more traditional but very important, aims at unraveling the molecular mechanisms underlying memory formation (e.g., using state-of-the-art molecular and cell-specific genetic approaches). Learning and memory disorders can strike the brain during development (e.g., Autism Spectrum Disorders and Down Syndrome), as well as during adulthood (e.g., Alzheimer’s disease). We are interested in understanding the specific circuits and molecular pathways that are primarily targeted in these disorders and how they can be restored. To tackle these questions, we use a multidisciplinary, convergent and cross-species approach that combines mouse and fly genetics, molecular biology, electrophysiology, stem cell biology, optogenetics and behavioral techniques. The second line of research, more recent and relatively unexplored, is focused on understanding how gut microbes control CNS driven-behavior and brain function. Our recent discoveries, that microbes in the gut could modulate brain function and behavior in a very powerful way, have added a whole new dimension to the classic view of how complex behaviors are controlled. The unexpected findings have opened new avenues of study for us and are currently driving my lab to answer a host of new and very interesting questions: - What are the gut microbes (and metabolites) that regulate CNS-driven behaviors? Would it be possible to develop an unbiased screening method to identify specific microbes that regulate different behaviors? - If this is the case, can we identify how members of the gut microbiome (and their metabolites) mechanistically influence brain function? - What is the communication channel between the gut microbiota and the brain? Do different gut microbes use different ways to interact with the brain? - Could disruption of the gut microbial ecology cause neurodevelopmental dysfunction? If so, what is the impact of disruption in young and adult animals? - More importantly, could specific restoration of selected bacterial strains (new generation probiotics) represent a novel therapeutic approach for the targeted treatment of neurodevelopmental disorders? - Finally, can we develop microbiota-directed therapeutic foods to repair brain dysfunction in a variety of neurological disorders?

My research uses a range of methods to better understand how the brain’s natural chemicals control complex behaviours, thoughts and perceptions. I also have a particular fascination about the factors that determine the contents of an individual’s conscious experience. In this talk I will present work that sits at the intersection of these two research areas looking at the role of different neurotransmitter systems in driving changes in conscious state. Specifically, I will discuss a series of studies using ambiguous stimuli to explore the neuropharmacological processes that underly alternations in perceptual awareness. By comparing different methods and neurotransmitter systems including: serotonin (psychedelics), noradrenaline (pupillometry) and Glutamate/GABA (Magnetic Resonance Spectroscopy MRS) we can start to tease apart the distinct role that different neurotransmitter systems play in coordinating conscious experience across time.

Developmental and epileptic encephalopathies (DEEs) are a broad spectrum of genetic epilepsies associated with impaired neurological development as a direct consequence of a genetic mutation, in addition to the effect of the frequent epileptic activity on brain. Compelling genetic studies indicate that heterozygous de novo mutations represent the most common underlying genetic mechanism, in accordance with the sporadic presentation of DEE. De novo mutations may exert a loss-of-function (LOF) on the protein by decrementing expression level and/or activity, leading to functional haploinsufficiency. These diseases share several features: severe and frequent refractory seizures, diffusely abnormal background activity on EEG, intellectual disability often profound, and severe consequences on global development. One of major causes of early onset DEE are de novo heterozygous mutations in syntaxin-binding-protein-1 gene STXBP1, which encodes a membrane trafficking protein playing critical role in vesicular docking and fusion. LOF STXBP1 mutations lead to a failure of neurotransmitter secretion from synaptic vesicles. Core clinical features of STXBP1 encephalopathy include early-onset epilepsy with hypsarrhythmic EEG, or burst-suppression pattern, or multifocal epileptiform activity. Seizures are often resistant to standard treatments and patients typically show intellectual disability, mostly severe to profound. Additional neurologic features may include autistic traits, movement disorders (dyskinesia, dystonia, tremor), axial hypotonia, and ataxia, indicating a broader neurologic impairment. Patients with severe neuro-cognitive features but without epilepsy have been reported. Recently, a new class of natural and synthetic non-coding RNAs have been identified, enabling upregulation of protein translation in a gene-specific way (SINEUPs), without any increase in mRNA of the target gene. SINEUPs are translational activators composed by a Binding Domain (BD) that overlaps, in antisense orientation, to the sense protein-coding mRNA, and determines target selection; and an Effector Domain (ED), that is essential for protein synthesis up regulation. SINEUPs have been shown to restore the physiological expression of a protein in case of haploinsufficiency, without driving excessive overexpression out of the physiological range. This technology brings many advantages, as it mainly acts on endogenous target mRNAs produced in situ by the wild-type allele; this action is limited to mRNA under physiological regulation, therefore no off-site effects can be expected in cells and tissues that do not express the target transcript; by acting only on a posttranscriptional level, SINEUPs do not trigger hereditable genome editing. After bioinformatic analysis of the promoter region of interest, we designed SINEUPs with 3 different BD for STXBP1. Human neurons from iPSCs were treated and STXBP1 levels showed a 1.5-fold increase compared to the Negative control. RNA levels of STXBP1 after the administration of SINEUPs remained stable as expected. These preliminary results proved the SINEUPs potential to specifically increase the protein levels without impacting on the genome. This is an extremely flexible approach to target many developmental and epileptic encephalopathies caused by haploinsufficiency, and therefore to address these diseases in a more tailored and radical way.

The Brainstorms Festival is the No1 online neuroscience and AI event for scientists, businesses, investors and startups. Join and listen to talks from leading scientists, take part in interactive discussions, and network with the people driving neurotech and AI innovation globally. The festival provides a digital playground for visionaries with dozens of medical innovations, panel discussions, workshops, a hackathon, and a neuroethics panel discussion which is crucial topic for neurodiversity and disability rights. Register now and be part of our amazing crowd!

Understanding the properties of active matter is a challenge which is currently driving a rapid growth in soft- and bio-physics. Some of the most important examples of active matter are at the microscale, and include active colloids and suspensions of microorganisms, both as a simple active fluid (single species) and as mixed suspensions of active and passive elements. In this last class of systems, recent experimental and theoretical work has started to provide a window into new phenomena including activity-induced depletion interactions, phase separation, and the possibility to extract net work from active suspensions. Here I will present our work on a paradigmatic example of mixed active-passive system, where the activity is provided by swimming microalgae. Macro- and micro-scopic experiments reveal that microorganism-colloid interactions are dominated by rare close encounters leading to large displacements through direct entrainment. Simulations and theoretical modelling show that the ensuing particle dynamics can be understood in terms of a simple jump-diffusion process, combining standard diffusion with Poisson-distributed jumps. Entrainment length can be understood within the framework of Taylor dispersion as a competition between advection by the no-slip surface of the cell body and microparticle diffusion. Building on these results, we then ask how external control of the dynamics of the active component (e.g. induced microswimmer anisotropy/inhomogeneity) can be used to alter the transport of passive cargo. As a first step in this direction, we study the behaviour of mixed active-passive systems in confinement. The resulting spatial inhomogeneity in swimmers’ distribution and orientation has a dramatic effect on the spatial distribution of passive particles, with the colloids accumulating either towards the boundaries or towards the bulk of the sample depending on the size of the container. We show that this can be used to induce the system to de-mix spontaneously.

A key challenge in modern soft matter is to identify the principles that govern the organisation and functionality in non-equilibrium systems. Current research efforts largely focus on non-equilibrium processes that occur either at the single-molecule scale (e.g. protein and DNA conformations under driving forces), or at the scale of whole tissues, organisms, and active colloidal and microscopic objects. However, the range of the scales in-between — from molecules to large-scaled molecular assemblies that consume energy and perform work — remains under-explored. This is, nevertheless, the scale that is crucial for the function of a living cell, where molecular self-assembly driven far from equilibrium produces mechanical work needed for cell reshaping, transport, motility, division, and healing. Today I will discuss physical modelling of active elastic filaments, called ESCRT-III filaments, that dynamically assemble and disassemble on cell membranes. This dynamic assembly changes the filaments’ shape and mechanical properties and leads to the remodelling and cutting of cells. I will present a range of experimental comparisons of our simulation results: from ESCRT-III-driven trafficking in eukaryotes to division of evolutionary simple archaeal cells.

Our work concerns the general problem of adaptive behaviour in response to predatory threats, and of the neural mechanisms underlying a choice between strategies. When faced with a threat, an animal must decide whether to freeze, reducing its chances of being noticed, or to flee to the safety of a refuge. Animals from fish to primates choose between these two alternatives when confronted by an attacking predator, a choice that largely depends on the context in which the threat occurs. Recent work has made strides identifying the pre-motor circuits, and their inputs, which control freezing behaviour in rodents, but how contextual information is integrated to guide this choice is still far from understood. The social environment is a potent contextual modulator of defensive behaviours of animals in a group. Indeed, anti-predation strategies are believed to be a major driving force for the evolution of sociality. We recently found that fruit flies in response to visual looming stimuli, simulating a large object on collision course, make rapid freeze/flee choices accompanied by lasting changes in the fly’s internal state, reflected in altered cardiac activity. In this talk, I will discuss our work on how flies process contextual cues, focusing on the social environment, to guide their behavioural response to a threat. We have identified a social safety cue, resumption of activity, and visual projection neurons involved in processing this cue. Given the knowledge regarding sensory detection of looming threats and descending neuron involved in the expression of freezing, we are now in a unique position to understand how information about a threat is integrated with cues from the social environment to guide the choice of whether to freeze.

We are interested in understanding how microbes impact the behavior of host animals. Animal nervous systems likely evolved in environments richly surrounded by microbes, yet the impact of bacteria on nervous system function has been relatively under-studied. A challenge has been to identify systems in which both host and microbe are amenable to genetic manipulation, and which enable high-throughput behavioral screening in response to defined and naturalistic conditions. To accomplish these goals, we use an animal host — the roundworm C. elegans, which feeds on bacteria — in combination with its natural gut microbiome to identify inter-organismal signals driving host-microbe interactions and decision-making. C. elegans has some of the most extensive molecular, neurobiological and genetic tools of any multicellular eukaryote, and, coupled with the ease of gnotobiotic culture in these worms, represents a highly attractive system in which to study microbial influence on host behavior. Using this system, we discovered that commensal bacterial metabolites directly modulate nervous system function of their host. Beneficial gut microbes of the genus Providencia produce the neuromodulator tyramine in the C. elegans intestine. Using a combination of behavioral analysis, neurogenetics, metabolomics and bacterial genetics we established that bacterially produced tyramine is converted to octopamine in C. elegans, which acts directly in sensory neurons to reduce odor aversion and increase sensory preference for Providencia. We think that this type of sensory modulation may increase association of C. elegans with these microbes, increasing availability of this nutrient-rich food source for the worm and its progeny, while facilitating dispersal of the bacteria.

Learning by analogy is a powerful tool children’s developmental repertoire, as well as in educational contexts such as mathematics, where the key knowledge base involves building flexible schemas. However, noticing and learning from analogies develops over time and is cognitively resource intensive. I review studies that provide insight into the relationship between mechanisms driving children’s developing analogy skills, highlighting environmental inputs (parent talk and prior experiences priming attention to relations) and neuro-cognitive factors (Executive Functions and brain injury). I then note implications for mathematics learning, reviewing experimental findings that show analogy can improve learning, but also that both individual differences in EFs and environmental factors that reduce available EFs such as performance pressure can predict student learning.

The amyloid cascade hypothesis for Alzheimer disease ((Hardy and Selkoe, 2002; Hardy and Higgins, 1992; Selkoe, 1991), updated in (Karran et al., 2011) provides a linear model for the pathogenesis of AD with Aβ accumulation upstream and Tau pathology, inflammation, synaptic dysfunction, neuronal loss and dementia downstream, all interlinked, initiated and driven by Aβ42 peptides or oligomers. The genetic mutations causing familial Alzheimer disease seem to support this model. The nagging problem remains however that the postulated causal, and especially the ’driving’ role of abnormal Aβ aggregation or Aβ oligomer formation could not be convincingly demonstrated until now. Indeed, many questions (e.g. what causes Aβ toxicity, what is the relation between Aβ and Tau pathology, what causes neuronal death, why is amyloid deposition not correlated with dementia etc…) were already raised when the amyloid hypothesis was conceived 25 years ago. These questions remain in essence unanswered. It seems that the old paradigm is not tenable: the amyloid cascade is too linear, too neurocentric, and does not take into account the long time lag between the biochemical phase i.e. the appearance of amyloid plaques and neuronal tangles and the ultimate clinical phase, i.e. the manifestation of dementia. The pathways linking these two phases must be complex and tortuous. We have called this the cellular phase of AD (De Strooper and Karran, 2016) to suggest that a long period of action and reaction involving neurons, neuronal circuitry but also microglia, astroglia, oligodendrocytes, and the vasculature underlies the disease. In fact it is this long disease process that should be studied in the coming years. While microglia are part of this process, they should not be considered as the only component of the cellular phase. We expect that further clinical investigations and novel tools will allow to diagnose the effects of the cellular changes in the brain and provide clinical signs for this so called preclinical or prodromal AD. Furthermore the better understanding of this phase will lead to completely novel drug targets and treatments and will lead to an era where patients will receive an appropriate therapy according to their clinical stage. In this view anti-amyloid therapy is probably only effective and useful in the very early stage of the disease and AD does no longer equal to dementia. We will discuss in our talk how single cell technology and transplantation of human iPS cells into mouse brain allow to start to map in a systematic way the cellular phase of Alzheimer’s Disease.

Understanding individual and macroscopic transport properties of motile micro-organisms in complex environments is a timely question, relevant to many ecological, medical and technological situations. At the fundamental level, this question is also receiving a lot of attention as fluids loaded with swimming micro-organisms has become a rich domain of applications and a conceptual playground for the statistical physics of “active matter”. The existence of microscopic sources of energy borne by the motile character of these micro-swimmers is driving self-organization processes at the origin of original emergent phases and unconventional macroscopic properties leading to revisit many standard concepts in the physics of suspensions. In this presentation, I will report on a recent exploration on the question of spontaneous formation of large scale collective motion in relation with the rheological response of active suspensions. I will also present new experiments showing how the motility of bacteria can be controlled such as to extract work macroscopically.

Engineered pattern-recognition systems strive for short time-to-solution and low energy-to-solution characteristics. This represents one of the main driving forces behind the development of neuromorphic devices. For both them and their biological archetypes, this corresponds to using as few spikes as early as possible. The concept of few and early spikes is used as the founding principle in the time-to-first-spike coding scheme. Within this framework, we have developed a spike-timing-based learning algorithm, which we used to train neuronal networks on the mixed-signal neuromorphic platform BrainScaleS-2. We derive, from first principles, error-backpropagation-based learning in networks of leaky integrate-and-fire (LIF) neurons relying only on spike times, for specific configurations of neuronal and synaptic time constants. We explicitly examine applicability to neuromorphic substrates by studying the effects of reduced weight precision and range, as well as of parameter noise. We demonstrate the feasibility of our approach on continuous and discrete data spaces, both in software simulations and on BrainScaleS-2. This narrows the gap between previous models of first-spike-time learning and biological neuronal dynamics and paves the way for fast and energy-efficient neuromorphic applications.

The potential of machine learning and deep learning to advance artificial intelligence is driving a quest to build dedicated computers, such as neuromorphic hardware that emulate the biological processes of the brain. While the hardware technologies already exist, their application to real-world tasks is hindered by the lack of suitable programming methods. Advances at the interface of neural computation and machine learning showed that key aspects of deep learning models and tools can be transferred to biologically plausible neural circuits. Building on these advances, I will show that differentiable programming can address many challenges of programming spiking neural networks for solving real-world tasks, and help devise novel continual and local learning algorithms. In turn, these new algorithms pave the road towards systematically synthesizing machine intelligence in neuromorphic hardware without detailed knowledge of the hardware circuits.

Temporal lobe epilepsy is associated with memory deficits but the circuit mechanisms underlying these cognitive disabilities are not understood. We used electrophysiological recordings, open-source wire-free miniaturized microscopy and computational modeling to probe these deficits in a model of temporal lobe epilepsy. We find desynchronization of dentate gyrus interneurons with CA1 interneurons during theta oscillations and a loss of precision and stability of place fields. We also find that emergence of place cell dysfunction is delayed, providing a potential temporal window for treatments. Computation modeling shows that desynchronization rather than interneuron cell loss can drive place cell dysfunction. Future studies will uncover cell types driving these changes and transcriptional changes that may be driving dysfunction.

How does our brain rapidly determine if something is good or bad? How do we know our place within a social group? How do we know how to behave appropriately in dynamic environments with ever-changing conditions? The Tye Lab is interested in understanding how neural circuits important for driving positive and negative motivational valence (seeking pleasure or avoiding punishment) are anatomically, genetically and functionally arranged. We study the neural mechanisms that underlie a wide range of behaviours ranging from learned to innate, including social, feeding, reward-seeking and anxiety-related behaviours. We have also become interested in “social homeostasis” -- how our brains establish a preferred set-point for social contact, and how this maintains stability within a social group. How are these circuits interconnected with one another, and how are competing mechanisms orchestrated on a neural population level? We employ optogenetic, electrophysiological, electrochemical, pharmacological and imaging approaches to probe these circuits during behaviour.

During periods of persistent and inescapable stress, animals can switch from active to passive coping strategies to manage effort-expenditure. Such normally adaptive behavioural state transitions can become maladaptive in disorders such as depression. We developed a new class of multi-region recurrent neural network (RNN) models to infer brain-wide interactions driving such maladaptive behaviour. The models were trained to match experimental data across two levels simultaneously: brain-wide neural dynamics from 10-40,000 neurons and the realtime behaviour of the fish. Analysis of the trained RNN models revealed a specific change in inter-area connectivity between the habenula (Hb) and raphe nucleus during the transition into passivity. We then characterized the multi-region neural dynamics underlying this transition. Using the interaction weights derived from the RNN models, we calculated the input currents from different brain regions to each Hb neuron. We then computed neural manifolds spanning these input currents across all Hb neurons to define subspaces within the Hb activity that captured communication with each other brain region independently. At the onset of stress, there was an immediate response within the Hb/raphe subspace alone. However, RNN models identified no early or fast-timescale change in the strengths of interactions between these regions. As the animal lapsed into passivity, the responses within the Hb/raphe subspace decreased, accompanied by a concomitant change in the interactions between the raphe and Hb inferred from the RNN weights. This innovative combination of network modeling and neural dynamics analysis points to dual mechanisms with distinct timescales driving the behavioural state transition: early response to stress is mediated by reshaping the neural dynamics within a preserved network architecture, while long-term state changes correspond to altered connectivity between neural ensembles in distinct brain regions.