A computational neural model that incorporates both intrinsic dynamics and sensory feedback in the Aplysia feeding network

Studying the nervous system underlying animal motor control can shed light on how animals can adapt flexibly to a changing environment. We focus on the neural basis of feeding control in Aplysia californica. Using the Synthetic Nervous System framework, we developed a model of Aplysia feeding neural circuitry that balances neurophysiological plausibility and computational complexity. The circuitry includes neurons, synapses, and feedback pathways identified in existing literature. We organized the neurons into three layers and five subnetworks according to their functional roles. Simulation results demonstrate that the circuitry model can capture the intrinsic dynamics at neuronal and network levels. When combined with a simplified peripheral biomechanical model, it is sufficient to mediate three animal-like feeding behaviors (biting, swallowing, and rejection). The kinematic, dynamic, and neural responses of the model also share similar features with animal data. These results emphasize the functional roles of sensory feedback during feeding.

NSF DARE-Transforming modeling in neurorehabilitation: Four threads for catalyzing progress

We present an overview of the Conference on Transformative Opportunities for Modeling in Neurorehabilitation held in March 2023. It was supported by the Disability and Rehabilitation Engineering (DARE) program from the National Science Foundation's Engineering Biology and Health Cluster. The conference brought together experts and trainees from around the world to discuss critical questions, challenges, and opportunities at the intersection of computational modeling and neurorehabilitation to understand, optimize, and improve clinical translation of neurorehabilitation. We organized the conference around four key, relevant, and promising Focus Areas for modeling: Adaptation & Plasticity, Personalization, Human-Device Interactions, and Modeling 'In-the-Wild'. We identified four common threads across the Focus Areas that, if addressed, can catalyze progress in the short, medium, and long terms. These were: (i) the need to capture and curate appropriate and useful data necessary to develop, validate, and deploy useful computational models (ii) the need to create multi-scale models that span the personalization spectrum from individuals to populations, and from cellular to behavioral levels (iii) the need for algorithms that extract as much information from available data, while requiring as little data as possible from each client (iv) the insistence on leveraging readily available sensors and data systems to push model-driven treatments from the lab, and into the clinic, home, workplace, and community. The conference archive can be found at (dare2023.usc.edu). These topics are also extended by three perspective papers prepared by trainees and junior faculty, clinician researchers, and federal funding agency representatives who attended the conference.

NSF DARE-transforming modeling in neurorehabilitation: perspectives and opportunities from US funding agencies

In recognition of the importance and timeliness of computational models for accelerating progress in neurorehabilitation, the U.S. National Science Foundation (NSF) and the National Institutes of Health (NIH) sponsored a conference in March 2023 at the University of Southern California that drew global participation from engineers, scientists, clinicians, and trainees. This commentary highlights promising applications of computational models to understand neurorehabilitation ("Using computational models to understand complex mechanisms in neurorehabilitation" section), improve rehabilitation care in the context of digital twin frameworks ("Using computational models to improve delivery and implementation of rehabilitation care" section), and empower future interdisciplinary workforces to deliver higher-quality clinical care using computational models ("Using computational models in neurorehabilitation requires an interdisciplinary workforce" section). The authors describe near-term gaps and opportunities, all of which encourage interdisciplinary team science. Four major opportunities were identified including (1) deciphering the relationship between engineering figures of merit-a term commonly used by engineers to objectively quantify the performance of a device, system, method, or material relative to existing state of the art-and clinical outcome measures, (2) validating computational models from engineering and patient perspectives, (3) creating and curating datasets that are made publicly accessible, and (4) developing new transdisciplinary frameworks, theories, and models that incorporate the complexities of the nervous and musculoskeletal systems. This commentary summarizes U.S. funding opportunities by two Federal agencies that support computational research in neurorehabilitation. The NSF has funding programs that support high-risk/high-reward research proposals on computational methods in neurorehabilitation informed by theory- and data-driven approaches. The NIH supports the development of new interventions and therapies for a wide range of nervous system injuries and impairments informed by the field of computational modeling. The conference materials can be found at https://dare2023.usc.edu/ .

Understanding divergence: Placing developmental neuroscience in its dynamic context

Neurodevelopment is not merely a process of brain maturation, but an adaptation to constraints unique to each individual and to the environments we co-create. However, our theoretical and methodological toolkits often ignore this reality. There is growing awareness that a shift is needed that allows us to study divergence of brain and behaviour across conventional categorical boundaries. However, we argue that in future our study of divergence must also incorporate the developmental dynamics that capture the emergence of those neurodevelopmental differences. This crucial step will require adjustments in study design and methodology. If our ultimate aim is to incorporate the developmental dynamics that capture how, and ultimately when, divergence takes place then we will need an analytic toolkit equal to these ambitions. We argue that the over reliance on group averages has been a conceptual dead-end with regard to the neurodevelopmental differences. This is in part because any individual differences and developmental dynamics are inevitably lost within the group average. Instead, analytic approaches which are themselves new, or simply newly applied within this context, may allow us to shift our theoretical and methodological frameworks from groups to individuals. Likewise, methods capable of modelling complex dynamic systems may allow us to understand the emergent dynamics only possible at the level of an interacting neural system.

Personalized modeling of Alzheimer's disease progression estimates neurodegeneration severity from EEG recordings

INTRODUCTION

Early identification of Alzheimer's disease (AD) is necessary for a timely onset of therapeutic care. However, cortical structural alterations associated with AD are difficult to discern.

METHODS

We developed a cortical model of AD-related neurodegeneration accounting for slowing of local dynamics and global connectivity degradation. In a monocentric study we collected electroencephalography (EEG) recordings at rest from participants in healthy (HC, n = 17), subjective cognitive decline (SCD, n = 58), and mild cognitive impairment (MCI, n = 44) conditions. For each patient, we estimated neurodegeneration model parameters based on individual EEG recordings.

RESULTS

Our model outperformed standard EEG analysis not only in discriminating between HC and MCI conditions (F1 score 0.95 vs 0.75) but also in identifying SCD patients with biological hallmarks of AD in the cerebrospinal fluid (recall 0.87 vs 0.50).

DISCUSSION

Personalized models could (1) support classification of MCI, (2) assess the presence of AD pathology, and (3) estimate the risk of cognitive decline progression, based only on economical and non-invasive EEG recordings.

Highlights

Personalized cortical model estimating structural alterations from EEG recordings.Discrimination of Mild Cognitive Impairment (MCI) and Healthy (HC) subjects (95%)Prediction of biological markers of Alzheimer's in Subjective Decline (SCD) Subjects (87%)Transition correctly predicted for 3/3 subjects that converted from SCD to MCI after 1y.

Age-related variability in network engagement during music listening

Listening to music is an enjoyable behaviour that engages multiple networks of brain regions. As such, the act of music listening may offer a way to interrogate network activity, and to examine the reconfigurations of brain networks that have been observed in healthy aging. The present study is an exploratory examination of brain network dynamics during music listening in healthy older and younger adults. Network measures were extracted and analyzed together with behavioural data using a combination of hidden Markov modelling and partial least squares. We found age- and preference-related differences in fMRI data collected during music listening in healthy younger and older adults. Both age groups showed higher occupancy (the proportion of time a network was active) in a temporal-mesolimbic network while listening to self-selected music. Activity in this network was strongly positively correlated with liking and familiarity ratings in younger adults, but less so in older adults. Additionally, older adults showed a higher degree of correlation between liking and familiarity ratings consistent with past behavioural work on age-related dedifferentiation. We conclude that, while older adults do show network and behaviour patterns consistent with dedifferentiation, activity in the temporal-mesolimbic network is relatively robust to dedifferentiation. These findings may help explain how music listening remains meaningful and rewarding in old age.

Transcranial stimulation of alpha oscillations modulates brain state dynamics in sustained attention

The brain operates an advanced complex system to support mental activities. Cognition is thought to emerge from dynamic states of the complex brain system, which are organized spatially through large-scale neural networks and temporally via neural synchrony. However, specific mechanisms underlying these processes remain obscure. Applying high-definition alpha-frequency transcranial alternating-current stimulation (HD α-tACS) in a continuous performance task (CPT) during functional resonance imaging (fMRI), we causally elucidate these major organizational architectures in a key cognitive operation-sustained attention. We demonstrated that α-tACS enhanced both electroencephalogram (EEG) alpha power and sustained attention, in a correlated fashion. Akin to temporal fluctuations inherent in sustained attention, our hidden Markov modeling (HMM) of fMRI timeseries uncovered several recurrent, dynamic brain states, which were organized through a few major neural networks and regulated by the alpha oscillation. Specifically, during sustain attention, α-tACS regulated the temporal dynamics of the brain states by suppressing a Task-Negative state (characterized by activation of the default mode network/DMN) and Distraction state (with activation of the ventral attention and visual networks). These findings thus linked dynamic states of major neural networks and alpha oscillations, providing important insights into systems-level mechanisms of attention. They also highlight the efficacy of non-invasive oscillatory neuromodulation in probing the functioning of the complex brain system and encourage future clinical applications to improve neural systems health and cognitive performance.

Single Neuron Modeling Identifies Potassium Channel Modulation as Potential Target for Repetitive Head Impacts

Traumatic brain injury (TBI) and repetitive head impacts can result in a wide range of neurological symptoms. Despite being the most common neurological disorder in the world, repeat head impacts and TBI do not have any FDA-approved treatments. Single neuron modeling allows researchers to extrapolate cellular changes in individual neurons based on experimental data. We recently characterized a model of high frequency head impact (HFHI) with a phenotype of cognitive deficits associated with decreases in neuronal excitability of CA1 neurons and synaptic changes. While the synaptic changes have been interrogated in vivo, the cause and potential therapeutic targets of hypoexcitability following repetitive head impacts are unknown. Here, we generated in silico models of CA1 pyramidal neurons from current clamp data of control mice and mice that sustained HFHI. We use a directed evolution algorithm with a crowding penalty to generate a large and unbiased population of plausible models for each group that approximated the experimental features. The HFHI neuron model population showed decreased voltage gated sodium conductance and a general increase in potassium channel conductance. We used partial least squares regression analysis to identify combinations of channels that may account for CA1 hypoexcitability after HFHI. The hypoexcitability phenotype in models was linked to A- and M-type potassium channels in combination, but not by any single channel correlations. We provide an open access set of CA1 pyramidal neuron models for both control and HFHI conditions that can be used to predict the effects of pharmacological interventions in TBI models.

Hodgkin's and Huxley's own assessments of their "quantitative description" of nerve membrane current

Alan Hodgkin's and Andrew Huxley's mid-20th century work on the ionic currents generating neuron action potentials stands among that century's great scientific achievements. Unsurprisingly, that case has attracted widespread attention from neuroscientists, historians and philosophers of science. In this paper, I do not propose to add any new insights into the vast historical treatment of Hodgkin's and Huxley's scientific discoveries in that much- discussed episode. Instead, I focus on an aspect of it that hasn't received much attention: Hodgkin's and Huxley's own assessments about what their famous "quantitative description" accomplished. The "Hodgkin-Huxley model" is now widely recognized as a foundation of contemporary computational neuroscience. Yet Hodgkin and Huxley expressed serious caveats about their model and what it added to their scientific discoveries, as far back as their (1952d), in which they first presented their model. They were even more critical of its accomplishments in their Nobel Prize addresses a decade later. Most notably, as I argue here, some worries they raised about their quantitative description seem still to be relevant to current work in ongoing computational neuroscience.

Episodic Memory formation: A review of complex Hippocampus input pathways

Memories of everyday experiences involve the encoding of a rich and dynamic representation of present objects and their contextual features. Traditionally, the resulting mnemonic trace is referred to as Episodic Memory, i.e. the "what", "where" and "when" of a lived episode. The journey for such memory trace encoding begins with the perceptual data of an experienced episode handled in sensory brain regions. The information is then streamed to cortical areas located in the ventral Medio Temporal Lobe, which produces multi-modal representations concerning either the objects (in the Perirhinal cortex) or the spatial and contextual features (in the parahippocampal region) of the episode. Then, this high-level data is gated through the Entorhinal Cortex and forwarded to the Hippocampal Formation, where all the pieces get bound together. Eventually, the resulting encoded neural pattern is relayed back to the Neocortex for a stable consolidation. This review will detail these different stages and provide a systematic overview of the major cortical streams toward the Hippocampus relevant for Episodic Memory encoding.

Rapid learning of spatial representations for goal-directed navigation based on a novel model of hippocampal place fields

The discovery of place cells and other spatially modulated neurons in the hippocampal complex of rodents has been crucial to elucidating the neural basis of spatial cognition. More recently, the replay of neural sequences encoding previously experienced trajectories has been observed during consummatory behavior-potentially with implications for rapid learning, quick memory consolidation, and behavioral planning. Several promising models for robotic navigation and reinforcement learning have been proposed based on these and previous findings. Most of these models, however, use carefully engineered neural networks, and sometimes require long learning periods. In this paper, we present a self-organizing model incorporating place cells and replay, and demonstrate its utility for rapid one-shot learning in non-trivial environments with obstacles.

The HARM models: Predicting longitudinal physical aggression in patients with schizophrenia at an individual level

The prediction and prevention of aggression in individuals with schizophrenia remains a top priority within forensic psychiatric settings. While risk assessment methods are well rooted in forensic psychiatry, there are no available tools to predict longitudinal physical aggression in patients with schizophrenia within forensic settings at an individual level. In the present study, we used evidence-based risk and protective factors, as well as variables related to course of treatment assessed at baseline, to predict prospective incidents of physical aggression (4-month, 12-month, and 18-month follow-up) among 151 patients with schizophrenia within the forensic mental healthcare system. Across our HARM models, the balanced accuracy (sensitivity + specificity/2) of predicting physical aggressive incidents in patients with schizophrenia ranged from 59.73 to 87.33% at 4-month follow-up, 68.31-80.10% at 12-month follow-up, and 46.22-81.63% at 18-month follow-up, respectively. Additionally, we developed separate models, using clinician rated clinical judgement of short term and immediate violent risk, as a measure of comparison. Several modifiable evidence-based predictors of prospective physical aggression in schizophrenia were identified, including impulse control, substance abuse, impulsivity, treatment non-adherence, mood and psychotic symptoms, substance abuse, and poor family support. To the best of our knowledge, our HARM models are the first to predict longitudinal physical aggression at an individual level in patients with schizophrenia in forensic settings. However, it is important to caution that since these machine learning models were developed in the context of forensic settings, they may not be generalisable to individuals with schizophrenia more broadly. Moreover, a low base rate of physical aggression was observed in the testing set (6.0-11.6% across timepoints). As such, larger cohorts will be required to determine the replicability of these findings.

A dynamical model of visual motion processing for arbitrary stimuli including type II plaids

To explore the operating principle of visual motion processing in the brain underlying perception and eye movements, we model the information processing of velocity estimate of the visual stimulus at the algorithmic level using the dynamical system approach. In this study, we formulate the model as an optimization process of an appropriately defined objective function. The model is applicable to arbitrary visual stimuli. We find that our theoretical predictions qualitatively agree with time evolution of eye movement reported by previous works across various types of stimulus. Our results suggest that the brain implements the present framework as the internal model of motion vision. We anticipate our model to be a promising building block for more profound understanding of visual motion processing as well as for the development of robotics.

Optimization of data pre-processing methods for time-series classification of electroencephalography data

The performance of time-series classification of electroencephalographic data varies strongly across experimental paradigms and study participants. Reasons are task-dependent differences in neuronal processing and seemingly random variations between subjects, amongst others. The effect of data pre-processing techniques to ameliorate these challenges is relatively little studied. Here, the influence of spatial filter optimization methods and non-linear data transformation on time-series classification performance is analyzed by the example of high-frequency somatosensory evoked responses. This is a model paradigm for the analysis of high-frequency electroencephalography data at a very low signal-to-noise ratio, which emphasizes the differences of the explored methods. For the utilized data, it was found that the individual signal-to-noise ratio explained up to 74% of the performance differences between subjects. While data pre-processing was shown to increase average time-series classification performance, it could not fully compensate the signal-to-noise ratio differences between the subjects. This study proposes an algorithm to prototype and benchmark pre-processing pipelines for a paradigm and data set at hand. Extreme learning machines, Random Forest, and Logistic Regression can be used quickly to compare a set of potentially suitable pipelines. For subsequent classification, however, machine learning models were shown to provide better accuracy.

EBRAINS Live Papers - Interactive Resource Sheets for Computational Studies in Neuroscience

We present here an online platform for sharing resources underlying publications in neuroscience. It enables authors to easily upload and distribute digital resources, such as data, code, and notebooks, in a structured and systematic way. Interactivity is a prominent feature of the Live Papers, with features to download, visualise or simulate data, models and results presented in the corresponding publications. The resources are hosted on reliable data storage servers to ensure long term availability and easy accessibility. All data are managed via the EBRAINS Knowledge Graph, thereby helping maintain data provenance, and enabling tight integration with tools and services offered under the EBRAINS ecosystem.

Heteroclinic cycling and extinction in May-Leonard models with demographic stochasticity

May and Leonard (SIAM J Appl Math 29:243-253, 1975) introduced a three-species Lotka-Volterra type population model that exhibits heteroclinic cycling. Rather than producing a periodic limit cycle, the trajectory takes longer and longer to complete each "cycle", passing closer and closer to unstable fixed points in which one population dominates and the others approach zero. Aperiodic heteroclinic dynamics have subsequently been studied in ecological systems (side-blotched lizards; colicinogenic Escherichia coli), in the immune system, in neural information processing models ("winnerless competition"), and in models of neural central pattern generators. Yet as May and Leonard observed "Biologically, the behavior (produced by the model) is nonsense. Once it is conceded that the variables represent animals, and therefore cannot fall below unity, it is clear that the system will, after a few cycles, converge on some single population, extinguishing the other two." Here, we explore different ways of introducing discrete stochastic dynamics based on May and Leonard's ODE model, with application to ecological population dynamics, and to a neuromotor central pattern generator system. We study examples of several quantitatively distinct asymptotic behaviors, including total extinction of all species, extinction to a single species, and persistent cyclic dominance with finite mean cycle length.

A large and rich EEG dataset for modeling human visual object recognition

The human brain achieves visual object recognition through multiple stages of linear and nonlinear transformations operating at a millisecond scale. To predict and explain these rapid transformations, computational neuroscientists employ machine learning modeling techniques. However, state-of-the-art models require massive amounts of data to properly train, and to the present day there is a lack of vast brain datasets which extensively sample the temporal dynamics of visual object recognition. Here we collected a large and rich dataset of high temporal resolution EEG responses to images of objects on a natural background. This dataset includes 10 participants, each with 82,160 trials spanning 16,740 image conditions. Through computational modeling we established the quality of this dataset in five ways. First, we trained linearizing encoding models that successfully synthesized the EEG responses to arbitrary images. Second, we correctly identified the recorded EEG data image conditions in a zero-shot fashion, using EEG synthesized responses to hundreds of thousands of candidate image conditions. Third, we show that both the high number of conditions as well as the trial repetitions of the EEG dataset contribute to the trained models' prediction accuracy. Fourth, we built encoding models whose predictions well generalize to novel participants. Fifth, we demonstrate full end-to-end training of randomly initialized DNNs that output EEG responses for arbitrary input images. We release this dataset as a tool to foster research in visual neuroscience and computer vision.

ABOT: an open-source online benchmarking tool for machine learning-based artefact detection and removal methods from neuronal signals

Brain signals are recorded using different techniques to aid an accurate understanding of brain function and to treat its disorders. Untargeted internal and external sources contaminate the acquired signals during the recording process. Often termed as artefacts, these contaminations cause serious hindrances in decoding the recorded signals; hence, they must be removed to facilitate unbiased decision-making for a given investigation. Due to the complex and elusive manifestation of artefacts in neuronal signals, computational techniques serve as powerful tools for their detection and removal. Machine learning (ML) based methods have been successfully applied in this task. Due to ML’s popularity, many articles are published every year, making it challenging to find, compare and select the most appropriate method for a given experiment. To this end, this paper presents ABOT (Artefact removal Benchmarking Online Tool) as an online benchmarking tool which allows users to compare existing ML-driven artefact detection and removal methods from the literature. The characteristics and related information about the existing methods have been compiled as a knowledgebase (KB) and presented through a user-friendly interface with interactive plots and tables for users to search it using several criteria. Key characteristics extracted from over 120 articles from the literature have been used in the KB to help compare the specific ML models. To comply with the FAIR (Findable, Accessible, Interoperable and Reusable) principle, the source code and documentation of the toolbox have been made available via an open-access repository.

[Multidimensional and computational theory of mood]

What is mood? Despite its crucial place in psychiatric nosography and cognitive science, it is still difficult to delimit its conceptual ground. The distinction between emotion and mood is ambiguous: mood is often presented as an affective state that is more prolonged and less intense than emotion, or as an affective polarity distinguishing high and low mood swinging around a baseline. However, these definitions do not match the clinical reality of mood disorders such as unipolar depression and bipolar disorder, and do not allow us to understand the effect of mood on behaviour, perception and cognition. In this paper, we propose a multidimensional and computational theory of mood inspired by contemporary hypotheses in theoretical neuroscience and philosophy of emotion. After suggesting an operational distinction between emotion and mood, we show how a succession of emotions can cumulatively generate congruent mood over time, making mood an emerging state from emotion. We then present how mood determines mental and behavioral states when interacting with the environment, constituting a dispositional state of emotion, perception, belief, and action. Using this theoretical framework, we propose a computational representation of the emerging and dispositional dimensions of mood by formalizing mood as a layer of third-order Bayesian beliefs encoding the precision of emotion, and regulated by prediction errors associated with interoceptive predictions. Finally, we show how this theoretical framework sheds light on the processes involved in mood disorders, the emergence of mood congruent beliefs, or the mechanisms of antidepressant treatments in clinical psychiatry.

Probabilistic solvers enable a straight-forward exploration of numerical uncertainty in neuroscience models

Understanding neural computation on the mechanistic level requires models of neurons and neuronal networks. To analyze such models one typically has to solve coupled ordinary differential equations (ODEs), which describe the dynamics of the underlying neural system. These ODEs are solved numerically with deterministic ODE solvers that yield single solutions with either no, or only a global scalar error indicator on precision. It can therefore be challenging to estimate the effect of numerical uncertainty on quantities of interest, such as spike-times and the number of spikes. To overcome this problem, we propose to use recently developed sampling-based probabilistic solvers, which are able to quantify such numerical uncertainties. They neither require detailed insights into the kinetics of the models, nor are they difficult to implement. We show that numerical uncertainty can affect the outcome of typical neuroscience simulations, e.g. jittering spikes by milliseconds or even adding or removing individual spikes from simulations altogether, and demonstrate that probabilistic solvers reveal these numerical uncertainties with only moderate computational overhead.

Human-in-the-loop optimization of visual prosthetic stimulation

Retinal prostheses are a promising strategy to restore sight to patients with retinal degenerative diseases. These devices compensate for the loss of photoreceptors by electrically stimulating neurons in the retina. Currently, the visual function that can be recovered with such devices is very limited. This is due, in part, to current spread, unintended axonal activation, and the limited resolution of existing devices. Here we show, using a recent model of prosthetic vision, that optimizing how visual stimuli are encoded by the device can help overcome some of these limitations, leading to dramatic improvements in visual perception.

APPROACH

We propose a strategy to do this in practice, using patients' feedback in a visual task. The main challenge of our approach comes from the fact that, typically, one only has access to a limited number of noisy responses from patients. We propose two ways to deal with this: first, we use a model of prosthetic vision to constrain and simplify the optimization. We show that, if one knew the parameters of this model for a given patient, it would be possible to greatly improve their perceptual performance. Second we propose a preferential Bayesian optimization to efficiently learn these model parameters for each patient, using minimal trials.

MAIN RESULTS

To test our approach, we presented healthy subjects with visual stimuli generated by a recent model of prosthetic vision, to replicate the perceptual experience of patients fitted with an implant. Our optimization procedure led to significant and robust improvements in perceived image quality, that transferred to increased performance in other tasks.

SIGNIFICANCE

Importantly, our strategy is agnostic to the type of prosthesis and thus could readily be implemented in existing implants.

A step-by-step tutorial on active inference and its application to empirical data

The active inference framework, and in particular its recent formulation as a partially observable Markov decision process (POMDP), has gained increasing popularity in recent years as a useful approach for modeling neurocognitive processes. This framework is highly general and flexible in its ability to be customized to model any cognitive process, as well as simulate predicted neuronal responses based on its accompanying neural process theory. It also affords both simulation experiments for proof of principle and behavioral modeling for empirical studies. However, there are limited resources that explain how to build and run these models in practice, which limits their widespread use. Most introductions assume a technical background in programming, mathematics, and machine learning. In this paper we offer a step-by-step tutorial on how to build POMDPs, run simulations using standard MATLAB routines, and fit these models to empirical data. We assume a minimal background in programming and mathematics, thoroughly explain all equations, and provide exemplar scripts that can be customized for both theoretical and empirical studies. Our goal is to provide the reader with the requisite background knowledge and practical tools to apply active inference to their own research. We also provide optional technical sections and multiple appendices, which offer the interested reader additional technical details. This tutorial should provide the reader with all the tools necessary to use these models and to follow emerging advances in active inference research.

Sensation to navigation: a computational neuroscience approach to magnetic field navigation

Diverse taxa use Earth's magnetic field (i.e., magnetoreception) as a guide during long-distance navigation. However, despite decades of research, specific sensory mechanisms of magnetoreception remain unconfirmed. Necessarily, this has led to theoretical and computational work developing hypotheses of how animals may navigate using magnetoreception. One hypothesized strategy relies on an animal using combinations of magnetic intensity and inclination as a kind of signature to identify a specific region or location. Using these signatures, animals could use a waypoint-based navigation strategy. We show that this navigation strategy is biologically plausible using a close approximation of neural processing to successfully guide an agent in a simulated magnetic field. Moreover, we accomplish this strategy using a processing approach previously utilized for mechanoreception, suggesting processing of Earth's magnetic field may share features with the processing of other, more well-understood sensory systems. Taken together, our results suggest that both for the engineering of novel navigation systems and the study of animal magnetoreception, we should take lessons from other sensory systems.

Major depression disorder diagnosis and analysis based on structural magnetic resonance imaging and deep learning

Major depression disorder is one of the diseases with the highest rate of disability and morbidity and is associated with numerous structural and functional differences in neural systems. However, it is difficult to analyze digital medical imaging data without computational intervention. A voxel-wise densely connected convolutional neural network, Three-dimensional Densenet (3D-DenseNet), is proposed to mine the feature differences. In addition, a novel transfer learning method, called Alzheimer's Disease Neuroimaging Initiative Transfer (ADNI-Transfer), is designed and combined with the proposed 3D-DenseNet. The experimental results on a database that contains 174 subjects, including 99 patients with major depression disorder and 75 healthy controls, show that large changes in brain structures between major depressive disorder patients and healthy controls mainly are located in the regions including superior frontal gyrus, dorsolateral, middle temporal gyrus, middle frontal gyrus, postcentral gyrus, inferior temporal gyrus. In addition, the proposed deep learning network can better extract different features of brain structures between major depressive disorder patients and healthy controls and achieve excellent classification results of major depressive disorder. At the same time, the designed transfer learning method can further improve classification performance. These results verify that our proposed method is feasible and valid for diagnosing and analyzing major depression disorder.

Assessing Computational Methods to Quantify Mother-Child Brain Synchrony in Naturalistic Settings Based on fNIRS Signals

Mother-child brain-to-brain synchrony captures the temporal similarities in brain signals between dyadic partners, and has been shown to emerge during the display of joint behaviours. Despite the rise in the number of studies that investigate synchrony in naturalistic contexts, the use of varying methodological approaches to compute synchrony remains a central problem. When dyads engage in unstructured social interactions, the wide range of behavioural cues they display contribute to the use of varying lengths of signals to compute synchrony. The present functional Near-infrared Spectroscopy (fNIRS) study investigates how different methods to quantify brain signals during joint and non-joint portions of dyadic play affect the outcome of brain-to-brain synchrony. Three strategies to cope with unstructured data are tested and different signal lengths of 15, 20, 25, 30, 35, 40, 45s were used to determine the optimal method to sensitively capture synchrony. Results showed that using all available portions of the signals generated a greater number of less conservative results compared to the other two strategies, which were to compute the average synchrony for the joint and non-joint signals portions and to compute the difference between the average synchrony of joint and non-joint portions. From the different signal durations, only length portions of 25s to 35s generated significant results. These findings demonstrate that differences in computational approaches and signal lengths affect synchrony measurements and should be considered in naturalistic synchrony studies.

Methylphenidate augmentation of escitalopram to enhance adherence to antidepressant treatment: a pilot randomized controlled trial

BACKGROUND

Adherence to treatment, i.e. the extent to which a patient's therapeutic engagement coincides with the prescribed treatment, is among the most important problems in mental health care. The current study investigated the influence of pairing an acute positive reinforcing dopaminergic/noradrenergic effect (methylphenidate, MPH) with a standard antidepressant on the rates of adherence to medication treatment. The primary objective of this study was to determine whether MPH + escitalopram resulted in higher rates of medication adherence relative to placebo + escitalopram.

METHODS

Twenty participants with moderate to severe depression were 1-1 randomized to either (1) 5 mg MPH + 10 mg escitalopram or (2) placebo + 10 mg escitalopram with the possibility for a dose increase at 4 weeks. A Bayesian analysis was conducted to evaluate the outcomes.

RESULTS

First, neither percent Pill count nor Medication Electronic Monitoring System adherence showed that MPH was superior to placebo. In fact, placebo showed slightly higher adherence rates on the primary (7.82% better than MPH) and secondary (7.07% better than MPH) outcomes. There was a less than 25% chance of MPH augmentation showing at least as good or better adherence than placebo. Second, both groups showed a significant effect of treatment on the QIDS-SR with a median effect of an 8.6-point score reduction. Third, neither subjective measures of adherence attitudes nor socio-demographic covariates had a significant influence on the primary or secondary outcome variables.

CONCLUSIONS

These data do not support the use of MPH to increase adherence to antidepressant medication in individuals with moderate to severe depression. CLINICALTRIALS.

GOV IDENTIFIER

NCT03388164 , registered on 01/02/2018.

Computational benefits of structural plasticity, illustrated in songbirds

The plasticity of nervous systems allows animals to quickly adapt to a changing environment. In particular, the structural plasticity of brain networks is often critical to the development of the central nervous system and the acquisition of complex behaviors. As an example, structural plasticity is central to the development of song-related brain circuits and may be critical for song acquisition in juvenile songbirds. Here, we review current evidences for structural plasticity and their significance from a computational point of view. We start by reviewing evidence for structural plasticity across species and categorizing them along the spatial axes as well as the along the time course during development. We introduce the vocal learning circuitry in zebra finches, as a useful example of structural plasticity, and use this specific case to explore the possible contributions of structural plasticity to computational models. Finally, we discuss current modeling studies incorporating structural plasticity and unexplored questions which are raised by such models.

Rendering neuronal state equations compatible with the principle of stationary action

The principle of stationary action is a cornerstone of modern physics, providing a powerful framework for investigating dynamical systems found in classical mechanics through to quantum field theory. However, computational neuroscience, despite its heavy reliance on concepts in physics, is anomalous in this regard as its main equations of motion are not compatible with a Lagrangian formulation and hence with the principle of stationary action. Taking the Dynamic Causal Modelling (DCM) neuronal state equation as an instructive archetype of the first-order linear differential equations commonly found in computational neuroscience, we show that it is possible to make certain modifications to this equation to render it compatible with the principle of stationary action. Specifically, we show that a Lagrangian formulation of the DCM neuronal state equation is facilitated using a complex dependent variable, an oscillatory solution, and a Hermitian intrinsic connectivity matrix. We first demonstrate proof of principle by using Bayesian model inversion to show that both the original and modified models can be correctly identified via in silico data generated directly from their respective equations of motion. We then provide motivation for adopting the modified models in neuroscience by using three different types of publicly available in vivo neuroimaging datasets, together with open source MATLAB code, to show that the modified (oscillatory) model provides a more parsimonious explanation for some of these empirical timeseries. It is our hope that this work will, in combination with existing techniques, allow people to explore the symmetries and associated conservation laws within neural systems - and to exploit the computational expediency facilitated by direct variational techniques.

Neuronal excitation/inhibition imbalance: core element of a translational perspective on Alzheimer pathophysiology

Our incomplete understanding of the link between Alzheimer's Disease pathology and symptomatology is a crucial obstacle for therapeutic success. Recently, translational studies have begun to connect the dots between protein alterations and deposition, brain network dysfunction and cognitive deficits. Disturbance of neuronal activity, and in particular an imbalance in underlying excitation/inhibition (E/I), appears early in AD, and can be regarded as forming a central link between structural brain pathology and cognitive dysfunction. While there are emerging (non-)pharmacological options to influence this imbalance, the complexity of human brain dynamics has hindered identification of an optimal approach. We suggest that focusing on the integration of neurophysiological aspects of AD at the micro-, meso- and macroscale, with the support of computational network modeling, can unite fundamental and clinical knowledge, provide a general framework, and suggest rational therapeutic targets.

Organization of areal connectivity in the monkey frontoparietal network

Activity observed in biological neural networks is determined by anatomical connectivity between cortical areas. The monkey frontoparietal network facilitates cognitive functions, but the organization of its connectivity is unknown. Here, a new connectivity matrix is proposed which shows that the network utilizes a small-world architecture and the 3-node M9 motif. Its areas exhibit relatively homogeneous connectivity with no suggestion of the hubs seen in scale-free networks. Crucially, its M9 dynamical relay motif is optimally arranged for near-zero and non-zero phase synchrony to arise in support of cognition, serving as a candidate topological mechanism for previously reported findings. These results can serve as a benchmark to be used in the treatment of neurological disorders where the types of cognition the frontoparietal network supports are impaired.

Assessment of eye-tracking scanpath outliers using fractal geometry

Outlier scanpaths identification is a crucial preliminary step in designing visual software, digital media analysis, radiology training and clustering participants in eye-tracking experiments. However, the task is challenging due to the visual irregularity of the scanpath shapes and the difficulty in dimensionality reduction due to geometric complexity. Conventional approaches have used heat maps to exclude scanpaths that lack a similarity pattern. However, the typically-used packages, such as ScanMatch and MultiMatch often generate discordant results when outlier identification is done empirically. This paper introduces a novel outlier evaluation approach by integrating the fractal dimension (FD), capturing the geometrical complexity of patterns, as an additional parameter with the heat map. This additional parameter is used to evaluate the degree of influence of a scanpath within a dataset. More specifically, the 2D Cartesian coordinates of a scanpath are fitted to a space filling 1D fractal curve to characterise its temporal FD. The FDs of the scanpaths are then compared to match their geometric complexity to one another. The findings indicate that the FD can be a beneficial additional parameter when evaluating the candidacy of poorly matching scanpaths as outliers and performs better at identifying unusual scanpaths than using other methods, including scanpath matching, Jaccard, or bounding box methods alone.

Nonlinear System Identification of Neural Systems from Neurophysiological Signals

The human nervous system is one of the most complicated systems in nature. Complex nonlinear behaviours have been shown from the single neuron level to the system level. For decades, linear connectivity analysis methods, such as correlation, coherence and Granger causality, have been extensively used to assess the neural connectivities and input-output interconnections in neural systems. Recent studies indicate that these linear methods can only capture a certain amount of neural activities and functional relationships, and therefore cannot describe neural behaviours in a precise or complete way. In this review, we highlight recent advances in nonlinear system identification of neural systems, corresponding time and frequency domain analysis, and novel neural connectivity measures based on nonlinear system identification techniques. We argue that nonlinear modelling and analysis are necessary to study neuronal processing and signal transfer in neural systems quantitatively. These approaches can hopefully provide new insights to advance our understanding of neurophysiological mechanisms underlying neural functions. These nonlinear approaches also have the potential to produce sensitive biomarkers to facilitate the development of precision diagnostic tools for evaluating neurological disorders and the effects of targeted intervention.

Shaping a data-driven era in dementia care pathway through computational neurology approaches

BACKGROUND

Dementia is caused by a variety of neurodegenerative diseases and is associated with a decline in memory and other cognitive abilities, while inflicting an enormous socioeconomic burden. The complexity of dementia and its associated comorbidities presents immense challenges for dementia research and care, particularly in clinical decision-making.

MAIN BODY

Despite the lack of disease-modifying therapies, there is an increasing and urgent need to make timely and accurate clinical decisions in dementia diagnosis and prognosis to allow appropriate care and treatment. However, the dementia care pathway is currently suboptimal. We propose that through computational approaches, understanding of dementia aetiology could be improved, and dementia assessments could be more standardised, objective and efficient. In particular, we suggest that these will involve appropriate data infrastructure, the use of data-driven computational neurology approaches and the development of practical clinical decision support systems. We also discuss the technical, structural, economic, political and policy-making challenges that accompany such implementations.

CONCLUSION

The data-driven era for dementia research has arrived with the potential to transform the healthcare system, creating a more efficient, transparent and personalised service for dementia.

Control for multifunctionality: bioinspired control based on feeding in Aplysia californica

Animals exhibit remarkable feats of behavioral flexibility and multifunctional control that remain challenging for robotic systems. The neural and morphological basis of multifunctionality in animals can provide a source of bioinspiration for robotic controllers. However, many existing approaches to modeling biological neural networks rely on computationally expensive models and tend to focus solely on the nervous system, often neglecting the biomechanics of the periphery. As a consequence, while these models are excellent tools for neuroscience, they fail to predict functional behavior in real time, which is a critical capability for robotic control. To meet the need for real-time multifunctional control, we have developed a hybrid Boolean model framework capable of modeling neural bursting activity and simple biomechanics at speeds faster than real time. Using this approach, we present a multifunctional model of Aplysia californica feeding that qualitatively reproduces three key feeding behaviors (biting, swallowing, and rejection), demonstrates behavioral switching in response to external sensory cues, and incorporates both known neural connectivity and a simple bioinspired mechanical model of the feeding apparatus. We demonstrate that the model can be used for formulating testable hypotheses and discuss the implications of this approach for robotic control and neuroscience.

The predictive global neuronal workspace: A formal active inference model of visual consciousness

The global neuronal workspace (GNW) model has inspired over two decades of hypothesis-driven research on the neural basis of consciousness. However, recent studies have reported findings that are at odds with empirical predictions of the model. Further, the macro-anatomical focus of current GNW research has limited the specificity of predictions afforded by the model. In this paper we present a neurocomputational model - based on Active Inference - that captures central architectural elements of the GNW and is able to address these limitations. The resulting 'predictive global workspace' casts neuronal dynamics as approximating Bayesian inference, allowing precise, testable predictions at both the behavioural and neural levels of description. We report simulations demonstrating the model's ability to reproduce: 1) the electrophysiological and behavioural results observed in previous studies of inattentional blindness; and 2) the previously introduced four-way taxonomy predicted by the GNW, which describes the relationship between consciousness, attention, and sensory signal strength. We then illustrate how our model can reconcile/explain (apparently) conflicting findings, extend the GNW taxonomy to include the influence of prior expectations, and inspire novel paradigms to test associated behavioural and neural predictions.

Dynamics of sensory integration of olfactory and mechanical stimuli within the response patterns of moth antennal lobe neurons

Odors emanating from a biologically relevant source are rapidly embedded within a windy, turbuluent medium that folds and spins filaments into fragmented strands of varying sizes. Environmental odor plumes therefore exhibit complex spatiotemporal dynamics, and rarely yield an easily discernible concentration gradient marking an unambiguous trail to an odor source. Thus, sensory integration of chemical input, encoding odor identity or concentration, and mechanosensory input, encoding wind speed, is a critical task that animals face in resolving the complex dynamics of odor plumes and tracking an odor source. In insects, who employ olfactory navigation as their primary means of foraging for food and finding mates, the antennal lobe (AL) is the first brain structure that processes sensory odor information. Although the importance of chemosensory and mechanosensory integration is widely recognized, the AL itself has traditionally been viewed purely from the perspective of odor encoding, with little attention given to its role as a bimodal integrator. In this work, we seek to explore the AL as a model for studying sensory integration - it boasts well-understood architecture, well-studied olfactory responses, and easily measurable cells. Using a moth model, we present experimental data that clearly demonstrates that AL neurons respond, in dynamically distinct ways, to both chemosensory and mechanosensory input; mechanosensory responses are transient and temporally precise, while olfactory responses are long-lasting but lack temporal precision. We further develop a computational model of the AL, show that our model captures odor response dynamics reported in the literature, and examine the dynamics of our model with the inclusion of mechanosensory input; we then use our model to pinpoint dynamical mechanisms underlying the bimodal AL responses revealed in our experimental work. Finally, we propose a novel hypothesis about the role of mechanosensory input in sculpting AL dynamics and the implications for biological odor tracking.

Testing Psychosis Phenotypes From Bipolar-Schizophrenia Network for Intermediate Phenotypes for Clinical Application: Biotype Characteristics and Targets

BACKGROUND

Psychiatry aspires to the molecular understanding of its disorders and, with that knowledge, to precision medicine. Research supporting such goals in the dimension of psychosis has been compromised, in part, by using phenomenology alone to estimate disease entities. To this end, we are proponents of a deep phenotyping approach in psychosis, using computational strategies to discover the most informative phenotypic fingerprint as a promising strategy to uncover mechanisms in psychosis.

METHODS

Doing this, the Bipolar-Schizophrenia Network for Intermediate Phenotypes (B-SNIP) has used biomarkers to identify distinct subtypes of psychosis with replicable biomarker characteristics. While we have presented these entities as relevant, their potential utility in clinical practice has not yet been demonstrated.

RESULTS

Here we carried out an analysis of clinical features that characterize biotypes. We found that biotypes have unique and defining clinical characteristics that could be used as initial screens in the clinical and research settings. Differences in these clinical features appear to be consistent with biotype biomarker profiles, indicating a link between biological features and clinical presentation. Clinical features associated with biotypes differ from those associated with DSM diagnoses, indicating that biotypes and DSM syndromes are not redundant and are likely to yield different treatment predictions. We highlight 3 predictions based on biotype that are derived from individual biomarker features and cannot be obtained from DSM psychosis syndromes.

CONCLUSIONS

In the future, biotypes may prove to be useful for targeting distinct molecular, circuit, cognitive, and psychosocial therapies for improved functional outcomes.

Simulating Large-scale Models of Brain Neuronal Circuits using Google Cloud Platform

Biophysically detailed modeling provides an unmatched method to integrate data from many disparate experimental studies, and manipulate and explore with high precision the resultin brain circuit simulation. We developed a detailed model of the brain motor cortex circuits, simulating over 10,000 biophysically detailed neurons and 30 million synaptic connections. Optimization and evaluation of the cortical model parameters and responses was achieved via parameter exploration using grid search parameter sweeps and evolutionary algorithms. This involves running tens of thousands of simulations requiring significant computational resources. This paper describes our experience in setting up and using Google Compute Platform (GCP) with Slurm to run these large-scale simulations. We describe the best practices and solutions to the issues that arose during the process, and present preliminary results from running simulations on GCP.

NeuroConstruct-based implementation of structured-light stimulated retinal circuitry

BACKGROUND

Retinal circuitry provides a fundamental window to neural networks, featuring widely investigated visual phenomena ranging from direction selectivity to fast detection of approaching motion. As the divide between experimental and theoretical visual neuroscience is fading, neuronal modeling has proven to be important for retinal research. In neuronal modeling a delicate balance is maintained between bio-plausibility and model tractability, giving rise to myriad modeling frameworks. One biologically detailed framework for neuro modeling is NeuroConstruct, which facilitates the creation, visualization and analysis of neural networks in 3D.

RESULTS

Here, we extended NeuroConstruct to support the generation of structured visual stimuli, to feature different synaptic dynamics, to allow for heterogeneous synapse distribution and to enable rule-based synaptic connectivity between cell populations. We utilized this framework to demonstrate a simulation of a dense plexus of biologically realistic and morphologically detailed starburst amacrine cells. The amacrine cells were connected to a ganglion cell and stimulated with expanding and collapsing rings of light.

CONCLUSIONS

This framework provides a powerful toolset for the investigation of the yet elusive underlying mechanisms of retinal computations such as direction selectivity. Particularly, we showcased the way NeuroConstruct can be extended to support advanced field-specific neuro-modeling.

Measuring directed functional connectivity using non-parametric directionality analysis: Validation and comparison with non-parametric Granger Causality

Background

‘Non-parametric directionality’ (NPD) is a novel method for estimation of directed functional connectivity (dFC) in neural data. The method has previously been verified in its ability to recover causal interactions in simulated spiking networks in Halliday et al. (2015).

Methods

This work presents a validation of NPD in continuous neural recordings (e.g. local field potentials). Specifically, we use autoregressive models to simulate time delayed correlations between neural signals. We then test for the accurate recovery of networks in the face of several confounds typically encountered in empirical data. We examine the effects of NPD under varying: a) signal-to-noise ratios, b) asymmetries in signal strength, c) instantaneous mixing, d) common drive, e) data length, and f) parallel/convergent signal routing. We also apply NPD to data from a patient who underwent simultaneous magnetoencephalography and deep brain recording.

Results

We demonstrate that NPD can accurately recover directed functional connectivity from simulations with known patterns of connectivity. The performance of the NPD measure is compared with non-parametric estimators of Granger causality (NPG), a well-established methodology for model-free estimation of dFC. A series of simulations investigating synthetically imposed confounds demonstrate that NPD provides estimates of connectivity that are equivalent to NPG, albeit with an increased sensitivity to data length. However, we provide evidence that: i) NPD is less sensitive than NPG to degradation by noise; ii) NPD is more robust to the generation of false positive identification of connectivity resulting from SNR asymmetries; iii) NPD is more robust to corruption via moderate amounts of instantaneous signal mixing.

Conclusions

The results in this paper highlight that to be practically applied to neural data, connectivity metrics should not only be accurate in their recovery of causal networks but also resistant to the confounding effects often encountered in experimental recordings of multimodal data. Taken together, these findings position NPD at the state-of-the-art with respect to the estimation of directed functional connectivity in neuroimaging.

•

Non-parametric directionality (NPD) is a novel directed connectivity measure.

•

NPD estimates are equivalent to estimates of Granger causality but are more robust to signal confounds.

•

Multivariate extensions of NPD can correctly identify signal routing.

Modeling awake hippocampal reactivations with model-based bidirectional search

Hippocampal offline reactivations during reward-based learning, usually categorized as replay events, have been found to be important for performance improvement over time and for memory consolidation. Recent computational work has linked these phenomena to the need to transform reward information into state-action values for decision making and to propagate it to all relevant states of the environment. Nevertheless, it is still unclear whether an integrated reinforcement learning mechanism could account for the variety of awake hippocampal reactivations, including variety in order (forward and reverse reactivated trajectories) and variety in the location where they occur (reward site or decision-point). Here, we present a model-based bidirectional search model which accounts for a variety of hippocampal reactivations. The model combines forward trajectory sampling from current position and backward sampling through prioritized sweeping from states associated with large reward prediction errors until the two trajectories connect. This is repeated until stabilization of state-action values (convergence), which could explain why hippocampal reactivations drastically diminish when the animal's performance stabilizes. Simulations in a multiple T-maze task show that forward reactivations are prominently found at decision-points while backward reactivations are exclusively generated at reward sites. Finally, the model can generate imaginary trajectories that are not allowed to the agent during task performance. We raise some experimental predictions and implications for future studies of the role of the hippocampo-prefronto-striatal network in learning.

Breaking human social decision making into multiple components and then putting them together again

Most of our waking time as human beings is spent interacting with other individuals. In order to make good decisions in this social milieu, it is often necessary to make inferences about the internal states, traits and intentions of others. Recently, some progress has been made toward uncovering the neural computations underlying human social decision-making by combining functional magnetic resonance neuroimaging (fMRI) with computational modeling of behavior. Modeling of behavioral data allows us to identify the key computations necessary for social decision-making and to determine how these computations are integrated. Furthermore, by correlating these variables against neuroimaging data, it has become possible to elucidate where in the brain various computations are implemented. Here we review the current state of knowledge in the domain of social computational neuroscience. Findings to date have emphasized that social decisions are driven by multiple computations conducted in parallel, and implemented in distinct brain regions. We suggest that further progress is going to depend on identifying how and where such variables get integrated in order to yield a coherent behavioral output.

Artificial intelligence outperforms human students in conducting neurosurgical audits

OBJECTIVES

Neurosurgical audits are an important part of improving the safety, efficiency and quality of care but require considerable resources, time, and funding. To that end, the advent of the Artificial Intelligence-based algorithms offered a novel, more economically viable solution. The aim of the study was to evaluate whether the algorithm can indeed outperform humans in that task.

PATIENTS & METHODS

Forty-six human students were invited to inspect the clinical notes of 45 medical outliers on a neurosurgical ward. The aim of the task was to produce a report containing a quantitative analysis of the scale of the problem (e.g. time to discharge) and a qualitative list of suggestions on how to improve the patient flow, quality of care, and healthcare costs. The Artificial Intelligence-based Frideswide algorithm (FwA) was used to analyse the same dataset.

RESULTS

The FwA produced 44 recommendations whilst human students reported an average of 3.89. The mean time to deliver the final report was 5.80 s for the FwA and 10.21 days for humans. The mean relative error for factual inaccuracy for humans was 14.75 % for total waiting times and 81.06 % for times between investigations. The report produced by the FwA was entirely factually correct. 13 out of 46 students submitted an unfinished audit, 3 out of 46 made an overdue submission. Thematic analysis revealed numerous internal contradictions of the recommendations given by human students.

CONCLUSION

The AI-based algorithm can produce significantly more recommendations in shorter time. The audits conducted by the AI are more factually accurate (0 % error rate) and logically consistent (no thematic contradictions). This study shows that the algorithm can produce reliable neurosurgical audits for a fraction of the resources required to conduct it by human means.

Synchronization of Hindmarsh Rose Neurons

Modeling and implementation of biological neurons are key to the fundamental understanding of neural network architectures in the brain and its cognitive behavior. Synchronization of neuronal models play a significant role in neural signal processing as it is very difficult to identify the actual interaction between neurons in living brain. Therefore, the synchronization study of these neuronal architectures has received extensive attention from researchers. Higher biological accuracy of these neuronal units demands more computational overhead and requires more hardware resources for implementation. This paper presents a two coupled hardware implementation of Hindmarsh Rose neuron model which is mathematically simpler model and yet mimics several behaviors of a real biological neuron. These neurons are synchronized using an exponential function. The coupled system shows several behaviors depending upon the parameters of HR model and coupling function. An approximation of coupling function is also provided to reduce the hardware cost. Both simulations and a low cost hardware implementations of exponential synaptic coupling function and its approximation are carried out for comparison. Hardware implementation on field programmable gate array (FPGA) of approximated coupling function shows that the coupled network produces different dynamical behaviors with acceptable error. Hardware implementation shows that the approximated coupling function has significantly lower implementation cost. A spiking neural network based on HR neuron is also shown as a practical application of this coupled HR neural networks. The spiking network successfully encodes and decodes a time varying input.

Greedy low-rank algorithm for spatial connectome regression

Recovering brain connectivity from tract tracing data is an important computational problem in the neurosciences. Mesoscopic connectome reconstruction was previously formulated as a structured matrix regression problem (Harris et al. in Neural Information Processing Systems, 2016), but existing techniques do not scale to the whole-brain setting. The corresponding matrix equation is challenging to solve due to large scale, ill-conditioning, and a general form that lacks a convergent splitting. We propose a greedy low-rank algorithm for the connectome reconstruction problem in very high dimensions. The algorithm approximates the solution by a sequence of rank-one updates which exploit the sparse and positive definite problem structure. This algorithm was described previously (Kressner and Sirković in Numer Lin Alg Appl 22(3):564-583, 2015) but never implemented for this connectome problem, leading to a number of challenges. We have had to design judicious stopping criteria and employ efficient solvers for the three main sub-problems of the algorithm, including an efficient GPU implementation that alleviates the main bottleneck for large datasets. The performance of the method is evaluated on three examples: an artificial "toy" dataset and two whole-cortex instances using data from the Allen Mouse Brain Connectivity Atlas. We find that the method is significantly faster than previous methods and that moderate ranks offer a good approximation. This speedup allows for the estimation of increasingly large-scale connectomes across taxa as these data become available from tracing experiments. The data and code are available online.

Dopamine affects short-term memory corruption over time in Parkinson's disease

Cognitive deficits are a recognised component of Parkinson's disease (PD). However, particularly within the domain of short-term memory, it is unclear whether these impairments are masked, or caused, by patients' dopaminergic medication. The effect of medication on pure maintenance in PD patients has rarely been explored, with most assessments examining maintenance intercalated between other executive tasks. Moreover, few studies have utilised methods that can measure the quality of mental representations, which can enable the decomposition of recall errors into their underlying neurocognitive components. Here, we fill this gap by examining pure maintenance in PD patients in high and low dopaminergic states. Participants had to encode the orientation of two stimuli and reproduce these orientations after a short (2 s) or long (8 s) delay. In addition, we also examined the performance of healthy, age-matched older adults to contextualise these effects and determine whether PD represents an exacerbation of the normal ageing process. Patients showed improved recall OFF compared to ON their dopaminergic medication, but only for long-duration trials. Moreover, PD patients OFF their medication actually performed at a level superior to age-matched controls, indicative of a paradoxical enhancement of memory in the low dopaminergic state. The application of a probabilistic model of response selection suggested that PD patients made fewer misbinding errors in the low, compared with high, dopaminergic state for longer-delay trials. Thus, unexpectedly, the mechanisms that prevent memoranda from being corrupted by misbinding over time appear to be enhanced in PD patients OFF dopaminergic medication. Possible explanations for this paradoxical effect are discussed.

Voxelwise encoding models with non-spherical multivariate normal priors

Predictive models for neural or fMRI data are often fit using regression methods that employ priors on the model parameters. One widely used method is ridge regression, which employs a spherical multivariate normal prior that assumes equal and independent variance for all parameters. However, a spherical prior is not always optimal or appropriate. There are many cases where expert knowledge or hypotheses about the structure of the model parameters could be used to construct a better prior. In these cases, non-spherical multivariate normal priors can be employed using a generalized form of ridge known as Tikhonov regression. Yet Tikhonov regression is only rarely used in neuroscience. In this paper we discuss the theoretical basis for Tikhonov regression, demonstrate a computationally efficient method for its application, and show several examples of how Tikhonov regression can improve predictive models for fMRI data. We also show that many earlier studies have implicitly used Tikhonov regression by linearly transforming the regressors before performing ridge regression.

Spatially informed voxelwise modeling for naturalistic fMRI experiments

Voxelwise modeling (VM) is a powerful framework to predict single voxel responses evoked by a rich set of stimulus features present in complex natural stimuli. However, because VM disregards correlations across neighboring voxels, its sensitivity in detecting functional selectivity can be diminished in the presence of high levels of measurement noise. Here, we introduce spatially-informed voxelwise modeling (SPIN-VM) to take advantage of response correlations in spatial neighborhoods of voxels. To optimally utilize shared information, SPIN-VM performs regularization across spatial neighborhoods in addition to model features, while still generating single-voxel response predictions. We demonstrated the performance of SPIN-VM on a rich dataset from a natural vision experiment. Compared to VM, SPIN-VM yields higher prediction accuracies and better capture locally congruent information representations across cortex. These results suggest that SPIN-VM offers improved performance in predicting single-voxel responses and recovering coherent information representations.

Computational neuroscience across the lifespan: Promises and pitfalls

In recent years, the application of computational modeling in studies on age-related changes in decision making and learning has gained in popularity. One advantage of computational models is that they provide access to latent variables that cannot be directly observed from behavior. In combination with experimental manipulations, these latent variables can help to test hypotheses about age-related changes in behavioral and neurobiological measures at a level of specificity that is not achievable with descriptive analysis approaches alone. This level of specificity can in turn be beneficial to establish the identity of the corresponding behavioral and neurobiological mechanisms. In this paper, we will illustrate applications of computational methods using examples of lifespan research on risk taking, strategy selection and reinforcement learning. We will elaborate on problems that can occur when computational neuroscience methods are applied to data of different age groups. Finally, we will discuss potential targets for future applications and outline general shortcomings of computational neuroscience methods for research on human lifespan development.

An information-theoretic perspective on the costs of cognition

In statistics and machine learning, model accuracy is traded off with complexity, which can be viewed as the amount of information extracted from the data. Here, we discuss how cognitive costs can be expressed in terms of similar information costs, i.e. as a function of the amount of information required to update a person's prior knowledge (or internal model) to effectively solve a task. We then examine the theoretical consequences that ensue from this assumption. This framework naturally explains why some tasks - for example, unfamiliar or dual tasks - are costly and permits to quantify these costs using information-theoretic measures. Finally, we discuss brain implementation of this principle and show that subjective cognitive costs can originate either from local or global capacity limitations on information processing or from increased rate of metabolic alterations. These views shed light on the potential adaptive value of cost-avoidance mechanisms.