In recent years, brain research has indisputably entered a new epoch, driven by substantial methodological advances and digitally enabled data integration and modelling at multiple scales—from molecules to the whole brain. Major advances are emerging at the intersection of neuroscience with technology and computing. This new science of the brain combines high-quality research, data integration across multiple scales, a new culture of multidisciplinary large-scale collaboration, and translation into applications. As pioneered in Europe’s Human Brain Project (HBP), a systematic approach will be essential for meeting the coming decade’s pressing medical and technological challenges. The aims of this paper are to: develop a concept for the coming decade of digital brain research, discuss this new concept with the research community at large, identify points of convergence, and derive therefrom scientific common goals; provide a scientific framework for the current and future development of EBRAINS, a research infrastructure resulting from the HBP’s work; inform and engage stakeholders, funding organisations and research institutions regarding future digital brain research; identify and address the transformational potential of comprehensive brain models for artificial intelligence, including machine learning and deep learning; outline a collaborative approach that integrates reflection, dialogues, and societal engagement on ethical and societal opportunities and challenges as part of future neuroscience research.

IntroductionInformation transmission and representation in both natural and artificial networks is dependent on connectivity between units. Biological neurons, in addition, modulate synaptic dynamics and post-synaptic membrane properties, but how these relate to information transmission in a population of neurons is still poorly understood. A recent study investigated local learning rules and showed how a spiking neural network can learn to represent continuous signals. Our study builds on their model to explore how basic membrane properties and synaptic delays affect information transfer.MethodsThe system consisted of three input and output units and a hidden layer of 300 excitatory and 75 inhibitory leaky integrate-and-fire (LIF) or adaptive integrate-and-fire (AdEx) units. After optimizing the connectivity to accurately replicate the input patterns in the output units, we transformed the model to more biologically accurate units and included synaptic delay and concurrent action potential generation in distinct neurons. We examined three different parameter regimes which comprised either identical physiological values for both excitatory and inhibitory units (Comrade), more biologically accurate values (Bacon), or the Comrade regime whose output units were optimized for low reconstruction error (HiFi). We evaluated information transmission and classification accuracy of the network with four distinct metrics: coherence, Granger causality, transfer entropy, and reconstruction error.ResultsBiophysical parameters showed a major impact on information transfer metrics. The classification was surprisingly robust, surviving very low firing and information rates, whereas information transmission overall and particularly low reconstruction error were more dependent on higher firing rates in LIF units. In AdEx units, the firing rates were lower and less information was transferred, but interestingly the highest information transmission rates were no longer overlapping with the highest firing rates.DiscussionOur findings can be reflected on the predictive coding theory of the cerebral cortex and may suggest information transfer qualities as a phenomenological quality of biological cells.

BACKGROUND:Resting-state functional magnetic resonance imaging (rsfMRI) reflects spontaneous activation of cortical networks. After stroke, these networks reorganize, both due to structural lesion and reorganization of functional connectivity (FC).OBJECTIVE:We studied FC in chronic phase occipital stroke patients with homonymous visual field defects before and after repetitive transorbital alternating current stimulation (rtACS).METHODS:This spin-off study, embedded in the randomized, sham-controlled REVIS trial, comprised 16 chronic occipital stroke patients with visual field defect and 12 healthy control subjects. The patients underwent rsfMRI at baseline, after two weeks of rtACS or sham treatment, and after two months of treatment-free follow-up, whereas the control subjects were measured once. We used a multivariate regression connectivity model to determine mutual prediction accuracy between 74 cortical regions of interest. Additionally, the model parameters were included into a graph to analyze average path length, centrality eigenvector, centrality degree, and clustering of the network. The patients and controls at baseline and the two treatment groups were compared with multilevel modeling.RESULTS:Before treatment, the patients and controls had similar whole-network prediction accuracy and network parameters, whereas centrality eigenvector differed in perilesional regions, indicating local modification in connectivity. In line with behavioral results, neither prediction accuracy nor any network parameter changed systematically as a result of rtACS rehabilitation compared to sham.CONCLUSIONS:Whole-network FC showed no difference between occipital stroke patients and healthy population, congruent with the peripheral location of the visual network in relation to the high-density cortical core. rtACS treatment in the given setting did not affect FC.

The cerebral cortex of primates encompasses multiple anatomically and physiologically distinct areas processing visual information. Areas V1, V2, and V5/MT are conserved across mammals and are central for visual behavior. To facilitate the generation of biologically accurate computational models of primate early visual processing, here we provide an overview of over 350 published studies of these three areas in the genus Macaca, whose visual system provides the closest model for human vision. The literature reports 14 anatomical connection types from the lateral geniculate nucleus of the thalamus to V1 having distinct layers of origin or termination, and 194 connection types between V1, V2, and V5, forming multiple parallel and interacting visual processing streams. Moreover, within V1, there are reports of 286 and 120 types of intrinsic excitatory and inhibitory connections, respectively. Physiologically, tuning of neuronal responses to 11 types of visual stimulus parameters has been consistently reported. Overall, the optimal spatial frequency (SF) of constituent neurons decreases with cortical hierarchy. Moreover, V5 neurons are distinct from neurons in other areas for their higher direction selectivity, higher contrast sensitivity, higher temporal frequency tuning, and wider SF bandwidth. We also discuss currently unavailable data that could be useful for biologically accurate models.

Simulation of the cerebral cortex requires a combination of extensive domain-specific knowledge and efficient software. However, when the complexity of the biological system is combined with that of the software, the likelihood of coding errors increases, which slows model adjustments. Moreover, few life scientists are familiar with software engineering and would benefit from simplicity in form of a high-level abstraction of the biological model. Our primary aim was to build a scalable cortical simulation framework for personal computers. We isolated an adjustable part of the domain-specific knowledge from the software. Next, we designed a framework that reads the model parameters from comma-separated value files and creates the necessary code for Brian2 model simulation. This separation allows rapid exploration of complex cortical circuits while decreasing the likelihood of coding errors and automatically using efficient hardware devices. Next, we tested the system on a simplified version of the neocortical microcircuit proposed by Markram and colleagues (2015). Our results indicate that the framework can efficiently perform simulations using Python, C++, and GPU devices. The most efficient device varied with computer hardware and the duration and scale of the simulated system. The speed of Brian2 was retained despite an overlying layer of software. However, the Python and C++ devices inherited the single core limitation of Brian2. The CxSystem framework supports exploration of complex models on personal computers and thus has the potential to facilitate research on cortical networks and systems.

Recently, Markram et al. (2015) presented a model of the rat somatosensory microcircuit (Markram model). Their model is high in anatomical and physiological detail, and its simulation requires supercomputers. The lack of neuroinformatics and computing power is an obstacle for using a similar approach to build models of other cortical areas or larger cortical systems. Simplified neuron models offer an attractive alternative to high-fidelity Hodgkin-Huxley-type neuron models, but their validity in modeling cortical circuits is unclear. We simplified the Markram model to a network of exponential integrate-and-fire (EIF) neurons that runs on a single CPU core in reasonable time. We analyzed the electrophysiology and the morphology of the Markram model neurons with eFel and NeuroM tools, provided by the Blue Brain Project. We then constructed neurons with few compartments and averaged parameters from the reference model. We used the CxSystem simulation framework to explore the role of short-term plasticity and GABAB and NMDA synaptic conductances in replicating oscillatory phenomena in the Markram model. We show that having a slow inhibitory synaptic conductance (GABAB) allows replication of oscillatory behavior in the high-calcium state. Furthermore, we show that qualitatively similar dynamics are seen even with a reduced number of cell types (from 55 to 17 types). This reduction halved the computation time. Our results suggest that qualitative dynamics of cortical microcircuits can be studied using limited neuroinformatics and computing resources supporting parameter exploration and simulation of cortical systems. The simplification procedure can easily be adapted to studying other microcircuits for which sparse electrophysiological and morphological data are available.

Despite evoked potentials' (EP) ubiquity in research and clinical medicine, insights are limited to gross brain dynamics as it remains challenging to map surface potentials to their sources in specific cortical regions. Multiple sources cancellation due to cortical folding and cross-talk obscures close sources, e.g. between visual areas V1 and V2. Recently retinotopic functional magnetic resonance imaging (fMRI) responses were used to constrain source locations to assist separating close sources and to determine cortical current generators. However, an fMRI is largely infeasible for routine EP investigation. We developed a novel method that replaces the fMRI derived retinotopic layout (RL) by an approach where the retinotopy and current estimates are generated from EEG or MEG signals and a standard clinical T1-weighted anatomical MRI. Using the EEG-RL, sources were localized to within 2 mm of the fMRI-RL constrained localized sources. The EEG-RL also produced V1 and V2 current waveforms that closely matched the fMRI-RL's (n = 2) r((1,198))=0.99, P < 0.0001. Applying the method to subjects without fMRI (n = 4) demonstrates it generates waveforms that agree closely with the literature. Our advance allows investigators with their current EEG or MEG systems to create a library of brain models tuned to individual subjects' cortical folding in retinotopic maps, and should be applicable to auditory and somatosensory maps. The novel method developed expands EP's ability to study specific brain areas, revitalizing this well-worn technique. Hum Brain Mapp 37:1696-1709, 2016. (c) 2016 Wiley Periodicals, Inc.

In the visual cortex, stimuli outside the classical receptive field (CRF) modulate the neural firing rate, without driving the neuron by themselves. In the primary visual cortex (V1), such contextual modulation can be parametrized with an area summation function (ASF): increasing stimulus size causes first an increase and then a decrease of firing rate before reaching an asymptote. Earlier work has reported increase of sparseness when CRF stimulation is extended to its surroundings. However, there has been no clear connection between the ASF and network efficiency. Here we aimed to investigate possible link between ASF and network efficiency. In this study, we simulated the responses of a biomimetic spiking neural network model of the visual cortex to a set of natural images. We varied the network parameters, and compared the V1 excitatory neuron spike responses to the corresponding responses predicted from earlier single neuron data from primate visual cortex. The network efficiency was quantified with firing rate (which has direct association to neural energy consumption), entropy per spike and population sparseness. All three measures together provided a clear association between the network efficiency and the ASF. The association was clear when varying the horizontal connectivity within V1, which influenced both the efficiency and the distance to ASF, DAS. Given the limitations of our biophysical model, this association is qualitative, but nevertheless suggests that an ASF-like receptive field structure can cause efficient population response.

Cumulative psychophysical evidence suggests that the shape of closed contours is analysed by means of their radial frequency components (RFC). However, neurophysiological evidence for RFC-based representations is still missing. We investigated the representation of radial frequency in the human visual cortex with functional magnetic resonance imaging. We parametrically varied the radial frequency, amplitude and local curvature of contour shapes. The stimuli evoked clear responses across visual areas in the univariate analysis, but the response magnitude did not depend on radial frequency or local curvature. Searchlight-based, multivariate representational similarity analysis revealed RFC specific response patterns in areas V2d, V3d, V3AB, and IPS0. Interestingly, RFC-specific representations were not found in hV4 or LO, traditionally associated with visual shape analysis. The modulation amplitude of the shapes did not affect the responses in any visual area. Local curvature, SF-spectrum and contrast energy related representations were found across visual areas but without similar specificity for visual area that was found for RFC. The results suggest that the radial frequency of a closed contour is one of the cortical shape analysis dimensions, represented in the early and mid-level visual areas.

Every stimulus or task activates multiple areas in the mammalian cortex. These distributed activations can be measured with functional magnetic resonance imaging (fMRI), which has the best spatial resolution among the noninvasive brain imaging methods. Unfortunately, the relationship between the fMRI activations and distributed cortical processing has remained unclear, both because the coupling between neural and fMRI activations has remained poorly understood and because fMRI voxels are too large to directly sense the local neural events. To get an idea of the local processing given the macroscopic data, we need models to simulate the neural activity and to provide output that can be compared with fMRI data. Such models can describe neural mechanisms as mathematical functions between input and output in a specific system, with little correspondence to physiological mechanisms. Alternatively, models can be biomimetic, including biological details with straightforward correspondence to experimental data. After careful balancing between complexity, computational efficiency, and realism, a biomimetic simulation should be able to provide insight into how biological structures or functions contribute to actual data processing as well as to promote theory-driven neuroscience experiments. This review analyzes the requirements for validating system-level computational models with fMRI. In particular, we study mesoscopic biomimetic models, which include a limited set of details from real-life networks and enable system-level simulations of neural mass action. In addition, we discuss how recent developments in neurophysiology and biophysics may significantly advance the modelling of fMRI signals.

The blood oxygenation level-dependent (BOLD) response has been strongly associated with neuronal activity in the brain. However, some neuronal tuning properties are consistently different from the BOLD response. We studied the spatial extent of neural and hemodynamic responses in the primary visual cortex, where the BOLD responses spread and interact over much longer distances than the small receptive fields of individual neurons would predict. Our model shows that a feedforward-feedback loop between V1 and a higher visual area can account for the observed spread of the BOLD response. In particular, anisotropic landing of inputs to compartmental neurons were necessary to account for the BOLD signal spread, while retaining realistic spiking responses. Our work shows that simple dendrites can separate tuning at the synapses and at the action potential output, thus bridging the BOLD signal to the neural receptive fields with high fidelity.

In the primary visual cortex (V1), increasing stimulus size first increases and then decreases the neural firing rate before reaching an asymptote [1,2]. This typical response curve, area summation function (ASF), defines canonical contextual modulation in typical V1 neuron. Previous studies suggest that contextual modulation associates with efficiency of the neural network [3,4]. Here, we studied the relationship between ASF and efficiency of system level neural activation patterns. Cavanaugh et al. [1] provided quantitative data from macaque cortex, as well as a mathematical description for the ASF. Therefore, we created a theoretical expected cortical response for an arbitrary gray-level image input stimulus based on ASF. Next, we used a biophysically meaningful (biomimetic) network of exponential integrate-and-fire neurons to stimulate V1 response (Heikkinen, Sharifian, Vigario and Vanni, unpublished observations). As an outcome, we compared the distance between the biomimetic simulator outputs to the expected ASF function (Distance to Area Summation, DAS). Finally, we studied entropy per spike, energy consumption and neural population sparseness as a function of the DAS for 20 natural grayscale images. In the biomimetic simulation we left V1-extrastriate and V1 excitatory-inhibitory connection strengths as free parameters in the simulations. A square search with 626 combinations of values was run for all the 20 natural stimuli resulting in 20 * 626 = 12520 simulated response patterns. Our results show that: i) there is a clear association between the entropy per spike and DAS (distance correlation = 0.66, Fig 1.a). ii) A negative link exists between the level of energy consumption (linearly associated with spike frequency) in modeled output patterns and DAS (r = 0.62, p < 0.001, Fig 1.b). iii) Population sparseness of the modeled output patterns is positively related to the DAS (r = 0.60, p < 0.001, Fig 1.c). Figure 1 In conclusion, the results of our study suggest that there is a link between area summation function and efficient information coding in the cortical visual system.

Although a great deal of research has been performed on special abilities occurring in connection with developmental disorders and injuries, their biological background remains unknown. It is tempting to think that understanding of the mechanism of generation of special ability would help each of us to liberate our brain capacity and direct it in a desired manner. Poor knowledge of the general functioning principle of the brain remains an essential restriction against establishing whether there is extra capacity in the brain and whether its liberation is possible now or in the near future. Model-based brain research is rising to the central position in understanding the functional principle.