SEARCH · Search bioRxiv

Results for “Neuroscience”

Search indexed bioRxiv preprints in genomics, neuroscience, cell biology and bioinformatics. Read source abstracts and check manuscript versions; preprints are not peer reviewed.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 307 records · Page 17Linked to original sources

bioRxiv2020

Development of the Brain Functional Connectome Follows Puberty-Dependent Nonlinear Trajectories

Adolescence is a developmental period that dramatically impacts body and behavior, with pubertal hormones playing an important role not only in the morphological changes in the body but also in brain structure and function. Understanding brain development during adolescence has become a priority in neuroscience because it coincides with the onset of many psychiatric and behavioral disorders. However, little is known about how puberty influences the brain functional connectome. In this study, taking a longitudinal human sample of typically developing children and adolescents (of both sexes), we demonstrate that the development of the brain functional connectome better fits pubertal status than chronological age. In particular, centrality, segregation, efficiency, and integration of the brain functional connectome increase after the onset of the pubertal markers. We found that these effects are stronger in attention and task control networks. Lastly, after controlling for this effect, we showed that functional connectivity between these networks is related to better performance in cognitive flexibility. This study points out the importance of considering longitudinal nonlinear trends when exploring developmental trajectories, and emphasizes the impact of puberty on the functional organization of the brain in adolescence. Significance StatementUnderstanding the brain organization along development is a crucial challenge for Neuroscience. In particular, during adolescence there is a great impact in body and cognitive functions as well as substantial incidence of mental health disruptions. Here, we tested how brain organization changes along this period based on the properties of the functional connectome in a longitudinal pediatric sample. We found a nonlinear increase in the connectivity and the brain network efficiency, particularly after the onset of puberty. These effects were more prominent in association networks. In addition, higher connectivity in those areas was associated with better performance in cognitive flexibility. These results demonstrate the importance of considering pubertal assessment as well as nonlinear trends in developmental studies.

neuroscience↗

bioRxiv2020

Brain Activation-based Filtering: a Scholarly Journal Recommendation Model for Human Neuroimaging Studies

The development of noninvasive neuroimaging techniques such as functional magnetic resonance imaging was followed by a large volume of human neuroimaging studies of mental processes, mechanisms, and diseases. Due to the high volume of studies and the large number of journals, it is increasingly challenging for neuroscientists to review existing scholarly journals and find the most suitable journal to publish their studies. Therefore, this paper proposes a scholarly journal recommendation model for human neuroimaging studies called brain activation-based filtering (BAF). Based on the collective matrix factorization technique, BAF recommends journals relevant to the activated brain regions that are described in a given neuroimaging study. For instance, if social brain regions such as the dorsomedial prefrontal cortex, precuneus, and temporoparietal junction are activated in a study, BAF recommends relevant social neuroscience journals (e.g., Social Cognitive and Affective Neuroscience). Five-fold cross-validation shows that BAF predicts journals with a reliable area under the curve score of 0.855. Furthermore, an interactive Google Colab notebook is offered to recommend relevant journals for a novel human neuroimaging study (https://github.com/JunsolKim/brain-activation-based-filtering).

neuroscience↗

bioRxiv2020

Learning sparse codes from compressed representations with biologically plausible local wiring constraints

Sparse coding is an important method for unsupervised learning of task-independent features in theoretical neuroscience models of neural coding. While a number of algorithms exist to learn these representations from the statistics of a dataset, they largely ignore the information bottlenecks present in fiber pathways connecting cortical areas. For example, the visual pathway has many fewer neurons transmitting visual information to cortex than the number of photoreceptors. Both empirical and analytic results have recently shown that sparse representations can be learned effectively after performing dimensionality reduction with randomized linear operators, producing latent coefficients that preserve information. Unfortunately, current proposals for sparse coding in the compressed space require a centralized compression process (i.e., dense random matrix) that is biologically unrealistic due to local wiring constraints observed in neural circuits. The main contribution of this paper is to leverage recent results on structured random matrices to propose a theoretical neuroscience model of randomized projections for communication between cortical areas that is consistent with the local wiring constraints observed in neuroanatomy. We show analytically and empirically that unsupervised learning of sparse representations can be performed in the compressed space despite significant local wiring constraints in compression matrices of varying forms (corresponding to different local wiring patterns). Our analysis verifies that even with significant local wiring constraints, the learned representations remain qualitatively similar, have similar quantitative performance in both training and generalization error, and are consistent across many measures with measured macaque V1 receptive fields.

neuroscience↗

bioRxiv2020

Head-free eye tracking, and efficient receptive field mapping in the marmoset

The marmoset has emerged as a promising primate model system, in particular for visual neuroscience. Many common experimental paradigms rely on head fixation and an extended period of eye fixation during the presentation of salient visual stimuli. Both of these behavioral requirements can be challenging for marmosets. Here, we present two methodological developments, each addressing one of these difficulties. First, we show that it is possible to use a standard eye tracking system without head fixation to assess visual behavior in the marmoset. Eye tracking quality from head-free animals is sufficient to obtain precise psychometric functions from a visual acuity task. Secondly, we introduce a novel method for efficient receptive field mapping that does not rely on moving stimuli but uses fast flashing annuli and wedges. We present data recorded during head-fixation in areas V1 and V6 and show that receptive field locations are readily obtained within a short period of recording time. Thus, the methodological advancements presented in this work will contribute to establish the marmoset as a valuable model in neuroscience.

neuroscience↗

bioRxiv2020

Learning to live with Dale's principle: ANNs with separate excitatory and inhibitory units

AO_SCPLOWBSTRACTC_SCPLOWThe units in artificial neural networks (ANNs) can be thought of as abstractions of biological neurons, and ANNs are increasingly used in neuroscience research. However, there are many important differences between ANN units and real neurons. One of the most notable is the absence of Dales principle, which ensures that biological neurons are either exclusively excitatory or inhibitory. Dales principle is typically left out of ANNs because its inclusion impairs learning. This is problematic, because one of the great advantages of ANNs for neuroscience research is their ability to learn complicated, realistic tasks. Here, by taking inspiration from feedforward inhibitory interneurons in the brain we show that we can develop ANNs with separate populations of excitatory and inhibitory units that learn just as well as standard ANNs. We call these networks Dales ANNs (DANNs). We present two insights that enable DANNs to learn well: (1) DANNs are related to normalization schemes, and can be initialized such that the inhibition centres and standardizes the excitatory activity, (2) updates to inhibitory neuron parameters should be scaled using corrections based on the Fisher Information matrix. These results demonstrate how ANNs that respect Dales principle can be built without sacrificing learning performance, which is important for future work using ANNs as models of the brain. The results may also have interesting implications for how inhibitory plasticity in the real brain operates.

neuroscience↗

bioRxiv2020

ATP signaling in the integrative neural center of Aplysia californica

ATP and its ionotropic P2X receptors are components of one of the most ancient signaling systems. However, little is known about the distribution and function of purinergic transmission in invertebrates. Here, we cloned, expressed, and pharmacologically characterized P2X receptors in the sea slug Aplysia californica - the prominent model in cellular and system neuroscience. These functional P2X receptors were successfully expressed in Xenopus oocytes and displayed activation by ATP (EC50=306 M) with two-phased kinetics as well as Na+-dependence. The ATP analog, Bz-ATP, was a less effective agonist (~20%) than ATP, and PPADS was a more potent inhibitor of the P2X receptors than the suramin. We showed that P2X receptors are uniquely expressed within Aplysias cerebral bioenergetic center (also known as F-cluster). Using RNA-seq, we found that the F-cluster contains more than a dozen unique secretory peptides, including three insulins, interleukins, and potential toxins, as well as ecdysone-type receptors and a district subset of ion channels. This structure is one of the most prominent integrative centers in the entire CNS and remarkably different from the morphologically similar neurosecretory center (bag cluster) involved in egg-laying behavior. Using RNA-seq, we also characterized the expression of P2X receptors across more than a dozen Aplysia peripheral tissues and developmental stages. We showed that P2X receptors are predominantly expressed in chemosensory structures and during early cleavage stages. The localization and pharmacology of P2X receptors in Aplysia highlight the evolutionary conservation of bioenergetic sensors and chemosensory purinergic transmission across animals. This study also provides a foundation to decipher homeostatic mechanisms in development and neuroendocrine systems. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=127 SRC="FIGDIR/small/392001v1_ufig1.gif" ALT="Figure 1"> View larger version (47K): org.highwire.dtl.DTLVardef@1ea3ab5org.highwire.dtl.DTLVardef@b45aadorg.highwire.dtl.DTLVardef@ad1851org.highwire.dtl.DTLVardef@17abbc_HPS_FORMAT_FIGEXP M_FIG We show that ATP and its ligand-gated P2X receptors are essential signaling components within both the chemosensory systems and the unique integrative neurosecretory center, present in the CNS of the sea slug Aplysia - a prominent model in neuroscience. Expression and pharmacology of P2X receptors in Aplysia confirms the preservation of evolutionary conserved bioenergetic sensors across animals and provide new tools to decipher homeostatic mechanisms in neuro-endocrine systems in general. C_FIG

neuroscience↗

bioRxiv2020

Neural Flip-Flops III: Stomatogastric Ganglion

The stomatogastric ganglion (STG) is a group of about 30 neurons that resides on the stomach in decapod crustaceans. Its two central pattern generators (CPGs) control the chewing action of the gastric mill and the peristaltic movement of food through the pylorus to the gut. The STG has been studied extensively because it has properties that are common to all nervous systems and because of the small number of neurons and other features that make it convenient to study. So many details are known that the STG is considered a classic test case in neuroscience for the reductionist strategy of explaining the emergence of macro-level phenomena from micro-level data. In spite of the intense scrutiny the STG has received, how it generates its rhythmic patterns of bursts remains unknown. The explicit neural networks proposed here model the pyloric CPG of the American lobster (Homarus americanus). The models share enough significant features with the lobsters CPG that they may be considered first approximations, or perhaps simplified versions, of STG architecture. The similarities include 1) mostly inhibitory synapses; 2) pairs of cells with reciprocal inhibitory inputs, complementary outputs that are approximately 180 degrees out of phase, and state changes occurring with the high output changing first; 3) cells that have reciprocal, inhibitory inputs with more than one other cell; and 4) six cells that produce coordinated oscillations with the same period, four phases distributed approximately uniformly over the period, and half of the burst durations approximately 1/4 of the period and the other half 3/8. Each models connectivity is explicit, and its operation depends only on minimal neuron capabilities of excitation and inhibition. One model performs a function that fills a gap in standard ring oscillators. It is apparently new to engineering, making it an example of neuroscience and logic circuit design informing each other. Some models are derived from standard circuit designs by moving each negation symbol from one end of a connection to the other. This does not change the logic of the network, but it changes each logic gate to one that can be implemented with a single neuron.

neuroscience↗

bioRxiv2021

Online learning for orientation estimation during translation in an insect ring attractor network

Insect neural systems are a promising source of inspiration for new algorithms for navigation, especially on low size, weight, and power platforms. There have been unprecedented recent neuroscience breakthroughs with Drosophila in behavioral and neural imaging experiments as well as the mapping of detailed connectivity of neural structures. General mechanisms for learning orientation in the central complex (CX) of Drosophila have been investigated previously; however, it is unclear how these underlying mechanisms extend to cases where there is translation through an environment (beyond only rotation), which is critical for navigation in robotic systems. Here, we develop a CX neural connectivity-constrained model that performs sensor fusion, as well as unsupervised learning of visual features for path integration; we demonstrate the viability of this circuit for use in robotic systems in simulated and physical environments. Furthermore, we propose a theoretical understanding of how distributed online unsupervised network weight modification can be leveraged for learning in a trajectory through an environment by minimizing of orientation estimation error. Overall, our results here may enable a new class of CX-derived low power robotic navigation algorithms and lead to testable predictions to inform future neuroscience experiments. SummaryAn insect neural connectivity-constrained model performs sensor fusion and online learning for orientation estimation.

neuroscience↗

bioRxiv2021

A Simple Parametric Representation of the Hodgkin-Huxley Model

The Hodgkin-Huxley model, decades after its first presentation, is still a reference model in neuroscience as it has successfully reproduced the electrophysiological activity of many organisms. The primary signal in the model represents the membrane potential of a neuron. A parametric and simple representation of this signal is presented in this paper. The new proposal is an adapted Frequency Modulated Mobius multicomponent model defined as a flexible decomposition in waves that describe the signal morphology. A specific feature of the new model is that the parameters are subject to interpretable restrictions. A broad simulation experiment is conducted to show the new model accurately represents the simulated Hodgkin-Huxley signal. Moreover, the model potential to predict the neurons relevant characteristics, described with parameters of the Hodgkin Huxley model, is shown using different Machine Learning methods. The proposed model is also validated with real data from Squid Giant Axons. The comparison of the parameter configuration between the simulated and real data demonstrated the flexibility of the model as well as interesting differences. Author summaryAlejandro Rodriguez-Collado. I received the double degree in Statistics and Computer Engineering and the Masters degree in Business Intelligence and Big Data from the Universidad de Valladolid in 2019 and 2020, respectively. I work as researcher and Professor for the Department of Statistics and Operational Research at the Universidad de Valladolid. My main research interests include oscillatory signal processing, neuroscience, multivariate data analysis and supervised learning. Cristina Rueda. I received the BS degree in mathematics from the Universidad de Valladolid in 1987 and the PhD degree in statistical science from the Universidad de Valladolid in 1989. I am currently Professor in the Department of Statistics and Operational Research at the Universidad de Valladolid. My main research interests include statistical inference methods under restrictions, circular data, computational biology, and statistical methods for signal analysis.

neuroscience↗

bioRxiv2021

Getting Blood from a Stone: Improving Neural Inferences without Additional Neural Data

In recent years, the cognitive neuroscience literature has come under criticism for containing many low-powered studies, limiting the ability to make reliable statistical inferences. Typically, the suggestion for increasing power is to collect more data with neural signals. However, many studies in cognitive neuroscience use parameters estimated from behavioral data in order to make inferences about neural signals (such as fMRI BOLD signal). In this paper, we explore how cognitive neuroscientists can learn more about their neuroimaging signal by collecting data on behavior alone and using alternative estimators designed to leverage this information. We demonstrate through simulation and mathematical derivations that knowing more about the marginal distribution of behavior can improve inferences about the mapping between cognitive processes and neural data. We analyze the magnitude of this benefit, finding that it depends on the desired estimand and several underlying study parameters. While in many cases the absolute gains in precision can be modest, our results demonstrate that, in realistic settings, additional behavioral data can lead to the same improvement in the precision of inferences more cheaply and easily than collecting additional data from subjects in a neuroimaging study. This means that when conducting a neuroimaging study, researchers now have another knob to turn in a design analysis: the number of subjects collected in the scanner and the number of behavioral subjects collected outside the scanner (in the lab or online).

neuroscience↗

bioRxiv2021

MEG, myself, and I: individual identification from neurophysiological brain activity

Large, openly available datasets and current analytic tools promise the emergence of population neuroscience. The considerable diversity in personality traits and behaviour between individuals is reflected in the statistical variability of neural data collected in such repositories. This amount of variability challenges the sensitivity and specificity of analysis methods to capture the personal characteristics of a putative neural portrait. Recent studies with functional magnetic resonance imaging (fMRI) have concluded that patterns of resting-state functional connectivity can both successfully identify individuals within a cohort and predict some individual traits, yielding the notion of a neural fingerprint. Here, we aimed to clarify the neurophysiological foundations of individual differentiation from features of the rich and complex dynamics of resting-state brain activity using magnetoencephalography (MEG) in 158 participants. Akin to fMRI approaches, neurophysiological functional connectomes enabled the identification of individuals, with identifiability rates similar to fMRIs. We also show that individual identification was equally successful from simpler measures of the spatial distribution of neurophysiological spectral signal power. Our data further indicate that identifiability can be achieved from brain recordings as short as 30 seconds, and that it is robust over time: individuals remain identifiable from recordings performed weeks after their baseline reference data was collected. Based on these results, we can anticipate a vast range of further research and practical applications of individual differentiation from neural electrophysiology in personalized, clinical, and basic neuroscience.

neuroscience↗

bioRxiv2021

Diversify your workflow! - An inconvenient suggestion to analyze spike data from intracranial recordings

An important challenge of neuroscience research and future brain machine interfacing is the reliable assignment of spikes to putative neurons. By means of extracellular recordings, researchers try to match different types action potentials with their putative neuronal source and timing. Unfortunately, this procedure is by far not standardized and reliable, leading to many different suggestions and as many differing results. It appears that sharing of data is thus hampered by different processing pipelines in different labs, thus playing along the reproducibility crisis in neurosciences. To systematically shed light on this issue, we present preliminary results of several easy event detection schemes on one data set, meant to illustrate the inconsistencies arising from different processing pipelines already in its initial step. The results indicate that thresholding choices alter findings due to a lack of a ground truth for spike sorting. We suggest to increase reliability in findings by only accepting and further processing events accepted by more than one processing pipeline.

neuroscience↗

bioRxiv2021

A novel theoretical framework for simultaneous measurement of excitatory and inhibitory conductances

Firing of neurons throughout the brain is determined by the precise relations between excitatory and inhibitory inputs and disruption of their balance underlies many psychiatric diseases. Whether or not these inputs covary over time or between repeated stimuli remains unclear due to the lack of experimental methods for measuring both inputs simultaneously. We developed a new analytical framework for instantaneous and simultaneous measurements of both the excitatory and inhibitory neuronal inputs during a single trial under current clamp recording. This can be achieved by injecting a current composed of two high frequency sinusoidal components followed by analytical extraction of the conductances. We demonstrate the ability of this method to measure both inputs in a single trial under realistic recording constraints and from morphologically realistic CA1 pyramidal model cells. Experimental implementation of our new method will facilitate the understanding of fundamental questions about the health and disease of the nervous system. ClassificationSystem Neuroscience, Cellular and Molecular Neuroscience

neuroscience↗

bioRxiv2021

Pigs as a new behavioral model for studying Pavlovian eyeblink conditioning

IntroPigs have been an increasingly popular preclinical model in nutritional neuroscience, as their anatomy, physiology, and nutrition requirements are highly comparable to those of humans. Eyeblink conditioning is one of the most well-validated behavioral paradigms in neuroscience to study underlying mechanisms of learning and memory formation in the cerebellum. Eyeblink conditioning has been performed in many species but has never been done on young pigs. Therefore, our aim here was to develop and validate an eyeblink conditioning paradigm in young pigs. MethodEighteen intact male pigs were artificially reared from postnatal day 2 to 30. The eyeblink conditioning setup consisted of a sound-damping box with a hammock that pigs were placed in, which allowed the pig to remain comfortable yet maintain a typical range of head motion. In a delay conditioning paradigm, the conditional stimulus (CS) was a 550 ms blue light-emitting diode (LED), the unconditional stimulus (US) was a 50 ms eye air-puff, the CS-US interval was 500 ms. Starting at postnatal day 14, pigs were habituated for five days to the eyeblink conditioning setup, followed by 5 daily sessions of acquisition training (40 paired CS-US trials each day). ResultsThe group-averaged amplitude of eyelid responses gradually increased over the course of the five days of training, indicating that pigs learned to make the association between the LED light CS and the air-puff US. A similar increase was found for the conditioned response (CR) probability: the group-averaged CR probability on session 1 was about 12% and reached a CR probability of 55% on day 5. The latency to CR peak time lacked a temporal preference in the first session, but clearly showed preference from the moment that animals started to show more CRs in session 2 and onwards whereby the eyelid was maximally closed exactly at the moment that the US would be delivered. ConclusionWe concluded that 4-week-old pigs have the capability of performing in a cerebellar classical association learning task, demonstrating for the first time that eyeblink conditioning in young pigs has the potential to be a valuable behavioral tool to measure neurodevelopment.

neuroscience↗

bioRxiv2021

Bipartite functional fractionation within the neural system for social cognition supports the psychological continuity of self vs. other

Research of social neuroscience establishes that regions in the brains default network (DN) and semantic network (SN) are engaged by socio-cognitive tasks. Research of the human connectome shows that DN and SN regions are both situated at the high-order end of cortical gradient but differ in their positions on this gradient. In the present study, we integrated these two bodies of research, used the psychological continuity of self vs. other as a test-case, and used fMRI to investigate whether these networks would encode social concepts differently. We found a robust dissociation between the DN and SN - while both networks contained sufficient information for decoding broad-stroke distinction of social categories, the DN carried more generalisable information for cross-classifying across social distance and emotive valence than did the SN. We also found that the overarching distinction of self vs. other was a principal divider of the representational space while social distance was an auxiliary factor (subdivision, nested within the principal dimension), and this representational landscape was more manifest in the DN than in the SN. Taken together, our findings demonstrate how insights from connectome research can benefit social neuroscience, and have implications for clarifying the two networks differential contributions to social cognition.

neuroscience↗

bioRxiv2021

Babytwins Study Sweden (BATSS): A multi-method infant twin study of genetic and environmental factors influencing infant brain and behavioral development

Twin studies can help us understand the relative contributions of genes and environment to phenotypic trait variation including attentional and brain activation measures. In terms of applying methodologies like electroencephalography (EEG) and eye tracking, which are key methods in developmental neuroscience, infant twin studies are almost non-existent. Here we describe the Babytwins Study Sweden (BATSS), a multi-method longitudinal twin study of 177 MZ and 134 DZ twin pairs (i.e. 622 individual infants) covering the 5 - 36 month time period. The study includes EEG, eye tracking and genetics, together with more traditional measures based on in-person testing, direct observation and questionnaires. The results show that interest in participation in research among twin parents is high, despite the comprehensive protocol. DNA analysis from saliva samples was possible in virtually all participants, allowing for both zygosity confirmation and polygenic score analyses. Combining a longitudinal twin design with advanced technologies in developmental cognitive neuroscience and genomics, BATSS represents a new approach in infancy research, which we hope to have impact across multiple disciplines in the coming years.

neuroscience↗

bioRxiv2021

How spatial attention affects the decision process: looking through the lens of Bayesian hierarchical diffusion model & EEG analysis

Model-based cognitive neuroscience elucidates the cognitive processes and neurophysiological oscillations that lead to behavioral performance in cognitive tasks (e.g., response times and accuracy). In this paper we explore the underlying latent process of spatial prioritization in perceptual decision processes, based on one well-known sequential sampling model (SSM), the drift-diffusion model (DDM), and subsequent nested model comparison. Neural components of spatial attention which contributed to the latent process and behavioral performance in a visual face-car perceptual decision were detected based on both time-frequency decomposition and event-related potential (ERP) analysis. For estimating DDM parameters (i.e. the drift rate, the boundary separation, and the non-decision time), a Bayesian hierarchical approach is considered, which allows inferences to be performed simultaneously on the group and individual level. Our cognitive modeling analysis revealed that spatial attention changed the non-decision time parameter across experimental conditions, such that a model with a changing non-decision time parameter provides a better fit to the data than other model parameters, quantified using the deviance information criterion (DIC) score and R-squared. Using multiple regression analysis on the contralateral minus neutral N2 sub-component (N2nc) at central electrodes, it can be concluded that poststimulus N2nc can predict mean response times (RTs) and non-decision time parameters related to spatial prioritization. However the contralateral minus neutral alpha power (Anc) at parieto-occipital electrodes can only predict the mean RTs and not the non-decision time relating to spatial prioritization. It was also found that the difference of contralateral minus neutral neural oscillations were more reflective of the modulation of the top-down spatial attention in comparison to the difference of ipsilateral minus neutral neural oscillations. These results suggest that individual differences in spatial attention are encoded by contralateral (and not ipsilateral) N2 oscillations and non-decision times. This work highlights how model-based Cognitive Neuroscience can further reveal the role of EEG in spatial attention during perceptual decision making.

neuroscience↗

bioRxiv2021

A connectivity-constrained computational account of topographic organization in high-level visual cortex

Inferotemporal cortex (IT) in humans and other primates is topo-graphically organized, containing multiple hierarchically-organized areas selective for particular domains, such as faces and scenes. This organization is commonly viewed in terms of evolved domain-specific visual mechanisms. Here, we develop an alternative, domain-general and developmental account of IT cortical organization. The account is instantiated as an Interactive Topographic Network (ITN), a form of computational model in which a hierarchy of model IT areas, subject to connectivity-based constraints, learns high-level visual representations optimized for multiple domains. We find that minimizing a wiring cost on spatially organized feedforward and lateral connections within IT, combined with constraining the feedforward processing to be strictly excitatory, results in a hierarchical, topographic organization. This organization replicates a number of key properties of primate IT cortex, including the presence of domain-selective spatial clusters preferentially involved in the representation of faces, objects, and scenes, columnar responses across separate excitatory and inhibitory units, and generic spatial organization whereby the response correlation of pairs of units falls off with their distance. We thus argue that domain-selectivity is an emergent property of a visual system optimized to maximize behavioral performance while minimizing wiring costs. Significance StatementWe introduce the Interactive Topographic Network, a framework for modeling high-level vision, to demonstrate in computational simulations that the spatial clustering of domains in late stages of the primate visual system may arise from the demands of visual recognition under the constraints of minimal wiring costs and excitatory between-area neuronal communication. The learned organization of the model is highly specialized but not fully modular, capturing many of the properties of organization in primates. Our work is significant for cognitive neuroscience, by providing a domain-general developmental account of topo-graphic functional specialization, and for computational neuroscience, by demonstrating how well-known biological details can be successfully incorporated into neural network models in order to account for critical empirical findings.

neuroscience↗