Neural modeling and functional neuroimaging

Two research areas that so far have had little interaction with one another are functional neuroimaging and computational neuroscience. The application of computational models and techniques to the inherently rich data sets generated by “standard” neurophysiological methods has proven useful for interpreting these data sets and for providing predictions and hypotheses for further experiments. We suggest that both theory- and data-driven computational modeling of neuronal systems can help to interpret data generated by functional neuroimaging methods, especially those used with human subjects. In this article, we point out four sets of questions, addressable by computational neuroscientists whose answere would be of value and interest to those who perform functional neuroimaging. The first set consist of determining the neurobiological substrate of the signals measured by functional neuroimaging. The second set concerns developing systems-level models of functional neuroimaging data. The third set of questions involves integrating functional neuroimaging data across modalities, with a particular emphasis on relating electromagnetic with hemodynamic data. The last set asks how one can relate systems-level models to those at the neuronal and neural ensemble levels. We feel that there are ample reasons to link functional neuroimaging and neural modeling, and that combining the results from the two disciplines will result in furthering our understanding of the central nervous system. © 1994 Wiley-Liss, Inc.

1 This Article is a US Goverment work and, as such, is in the public domain in the United State of America.

     

Bridging the gap: integrating cellular and functional magnetic resonance imaging studies of the hippocampus

The importance of the medial temporal lobe in memory has been studied extensively at the neuronal, neural ensemble, and systems level. In this report, we discuss recent systems level neuroimaging results in relation to neurophysiological studies of the hippocampus and related structures within the medial temporal lobe. By combining our knowledge across the cellular and systems levels we sought to gain theoretical insight and a better understanding of the function of the hippocampus and related medial temporal lobe structures. The integration of information from studies carried out at the cellular and neural ensemble level with studies at the systems level is difficult because of the vast differences in spatial and temporal resolution of the different research methodologies, differences in neuroanatomy across species, and differences in the types of behavioral and cognitive paradigms used in rat, nonhuman primate, and human studies. Despite these methodological and species-related differences, the neurophysiological studies offer insight into many of the questions raised by recent neuroimaging studies. For instance, there is physiological evidence that suggests that the hippocampal memory system is functionally heterogeneous, which may explain some of the discrepancies in the location and extent of activation reported by different imaging studies of the medial temporal lobe. In addition, we describe recent computational models of the hippocampus which may be useful for bridging the gap between neurophysiological and neuroimaging data.     

A Neuroinformatics of Brain Modeling and its Implementation in the Brain Operation Database BODB

We present principles for an integrated neuroinformatics framework which makes explicit how models are grounded on empirical evidence, explain (or not) existing empirical results and make testable predictions. The new ontological framework makes explicit how models bring together structural, functional, and related empirical observations. We emphasize schematics of the model’s operation linked to summaries of empirical data (SEDs) used in both the design and testing of the model, with tests comparing SEDs to summaries of simulation results (SSRs) from the model. We stress the importance of protocols for models as well as experiments. We complement the structural ontology of nested brain structures with a functional ontology of Brain Operating Principles (BOPs) for observed neural function and an ontological framework for grounding models in empirical data. We present an implementation of this ontological framework in the Brain Operation Database (BODB), an environment in which modelers and experimentalists can work together by making use of their shared empirical data, models and expertise.     

Mirror neurons and imitation: a computationally guided review.

Neurophysiology reveals the properties of individual mirror neurons in the macaque while brain imaging reveals the presence of 'mirror systems' (not individual neurons) in the human. Current conceptual models attribute high level functions such as action understanding, imitation, and language to mirror neurons. However, only the first of these three functions is well-developed in monkeys. We thus distinguish current opinions (conceptual models) on mirror neuron function from more detailed computational models. We assess the strengths and weaknesses of current computational models in addressing the data and speculations on mirror neurons (macaque) and mirror systems (human). In particular, our mirror neuron system (MNS), mental state inference (MSI) and modular selection and identification for control (MOSAIC) models are analyzed in more detail. Conceptual models often overlook the computational requirements for posited functions, while too many computational models adopt the erroneous hypothesis that mirror neurons are interchangeable with imitation ability. Our meta-analysis underlines the gap between conceptual and computational models and points out the research effort required from both sides to reduce this gap.     

Modeling the Development of Goal-Specificity in Mirror Neurons

Neurophysiological studies have shown that parietal mirror neurons encode not only actions but also the goal of these actions. Although some mirror neurons will fire whenever a certain action is perceived (goal-independently), most will only fire if the motion is perceived as part of an action with a specific goal. This result is important for the action-understanding hypothesis as it provides a potential neurological basis for such a cognitive ability. It is also relevant for the design of artificial cognitive systems, in particular robotic systems that rely on computational models of the mirror system in their interaction with other agents. Yet, to date, no computational model has explicitly addressed the mechanisms that give rise to both goal-specific and goal-independent parietal mirror neurons. In the present paper, we present a computational model based on a self-organizing map, which receives artificial inputs representing information about both the observed or executed actions and the context in which they were executed. We show that the map develops a biologically plausible organization in which goal-specific mirror neurons emerge. We further show that the fundamental cause for both the appearance and the number of goal-specific neurons can be found in geometric relationships between the different inputs to the map. The results are important to the action-understanding hypothesis as they provide a mechanism for the emergence of goal-specific parietal mirror neurons and lead to a number of predictions: (1) Learning of new goals may mostly reassign existing goal-specific neurons rather than recruit new ones; (2) input differences between executed and observed actions can explain observed corresponding differences in the number of goal-specific neurons; and (3) the percentage of goal-specific neurons may differ between motion primitives.     

Computational modelling of social cognition and behaviour—a reinforcement learning primer

Social neuroscience aims to describe the neural systems that underpin social cognition and behaviour. Over the past decade, researchers have begun to combine computational models with neuroimaging to link social computations to the brain. Inspired by approaches from reinforcement learning theory, which describes how decisions are driven by the unexpectedness of outcomes, accounts of the neural basis of prosocial learning, observational learning, mentalizing and impression formation have been developed. Here we provide an introduction for researchers who wish to use these models in their studies. We consider both theoretical and practical issues related to their implementation, with a focus on specific examples from the field.     

Micro-rewiring as a substrate for learning

How does the brain encode life experiences? Recent results derived from vital imaging, computational modeling, cellular physiology and systems neuroscience have pointed to local changes in synaptic connectivity as a powerful substrate, here termed micro-rewiring. To examine this hypothesis, I first review findings on micro-structural dynamics with focus on the extension and retraction of dendritic spines. Although these observations demonstrate a biological mechanism, they do not inform us of the specific changes in circuit configuration that might occur during learning. Here, computational models have made testable predictions for both the neuronal and circuit levels. Integrative approaches in the mammalian neocortex and the barn owl auditory localization pathway provide some of the first direct evidence in support of these 'synaptic-clustering' mechanisms. The implications of these data and the challenges for future research are discussed.     

Frontotemporal neural systems in bipolar disorder

Relatively less research has been performed in the delineation of the neural system abnormalities underlying bipolar disorder (BD) than in their correlates in unipolar depression. However, neuroimaging research has recently provided in vivo evidence to support the involvement of regional brain abnormalities in BD implicated by the localization of lesions associated with secondary mood symptoms. This article reviews (1) neural systems implicated in BD by brain lesions associated with secondary mood changes and impaired neuropsychologic paradigm performance; (2) structural and functional neuroimaging evidence to support the involvement of these neural systems in BD; and (3) potential functional neuroanatomic models of BD symptoms. Because depression is covered in detail elsewhere in this issue, this article focuses primarily on abnormalities associated with the manic state, as well as ones associated with euthymia, and may thus represent trait abnormalities in BD. We suggest that ventral and medial prefrontal and amygdalar abnormalities may play important roles in a subset of BD symptoms and are potential targets for treatments.     

Imaging evidence for disturbances in multiple learning and memory systems in persons with autism spectrum disorders

Aim The aim of this article is to review neuroimaging studies of autism spectrum disorders (ASD) that examine declarative, socio‐emotional, and procedural learning and memory systems. Method We conducted a search of PubMed from 1996 to 2010 using the terms ‘autism,’‘learning,’‘memory,’ and ‘neuroimaging.’ We limited our review to studies correlating learning and memory function with neuroimaging features of the brain. Results The early literature supports the following preliminary hypotheses: (1) abnormalities of hippocampal subregions may contribute to autistic deficits in episodic and relational memory; (2) disturbances to an amygdala‐based network (which may include the fusiform gyrus, superior temporal cortex, and mirror neuron system) may contribute to autistic deficits in socio‐emotional learning and memory; and (3) abnormalities of the striatum may contribute to developmental dyspraxia in individuals with ASD. Interpretation Characterizing the disturbances to learning and memory systems in ASD can inform our understanding of the neural bases of autistic behaviors and the phenotypic heterogeneity of ASD.     

Model-Based Optimization of Spinal Cord Stimulation for Inspiratory Muscle Activation

ObjectiveHigh-frequency spinal cord stimulation (HF-SCS) is a potential method to provide natural and effective inspiratory muscle pacing in patients with ventilator-dependent spinal cord injuries. Experimental data have demonstrated that HF-SCS elicits physiological activation of the diaphragm and inspiratory intercostal muscles via spinal cord pathways. However, the activation thresholds, extent of activation, and optimal electrode configurations (i.e., lead separation, contact spacing, and contact length) to activate these neural elements remain unknown. Therefore, the goal of this study was to use a computational modeling approach to investigate the direct effects of HF-SCS on the spinal cord and to optimize electrode design and stimulation parameters.Materials and MethodsWe developed a computer model of HF-SCS that consisted of two main components: 1) finite element models of the electric field generated during HF-SCS, and 2) multicompartment cable models of axons and motoneurons within the spinal cord. We systematically evaluated the neural recruitment during HF-SCS for several unique electrode designs and stimulation configurations to optimize activation of these neural elements. We then evaluated our predictions by testing two of these lead designs with in vivo canine experiments.ResultsOur model results suggested that within physiological stimulation amplitudes, HF-SCS activates both axons in the ventrolateral funiculi (VLF) and inspiratory intercostal motoneurons. We used our model to predict a lead design to maximize HF-SCS activation of these neural targets. We evaluated this lead design via in vivo experiments, and our computational model predictions demonstrated excellent agreement with our experimental testing.ConclusionsOur computational modeling and experimental results support the potential advantages of a lead design with longer contacts and larger edge-to-edge contact spacing to maximize inspiratory muscle activation during HF-SCS at the T2 spinal level. While these results need to be further validated in future studies, we believe that the results of this study will help improve the efficacy of HF-SCS technologies for inspiratory muscle pacing.     

Predictive, interactive multiple memory systems

Most lesion studies in animals, and neuropsychological and functional neuroimaging studies in humans, have focused on finding dissociations between the functions of different brain regions, for example in relation to different types of memory. While some of these dissociations can be questioned, particularly in the case of neuroimaging data, we start by assuming a "modal model" in which at least three different memory systems are distinguished: an episodic system (which stores associations between items and spatial/temporal contexts, and which is supported primarily by the hippocampus); a semantic system (which extracts combinations of perceptual features that define items, and which is supported primarily by anterior temporal cortex); and modality-specific perceptual systems (which represent the sensory features extracted from a stimulus, and which are supported by higher sensory cortices). In most situations however, behavior is determined by interactions between these systems. These interactions reflect the flow of information in both "forward" and "backward" directions between memory systems, where backward connections transmit predictions about the current item/features based on the current context/item. Importantly, it is the resulting "prediction error"--the difference between these predictions and the forward transmission of sensory evidence--that drives memory encoding and retrieval. We describe how this "predictive interactive multiple memory systems" (PIMMS) framework can be applied to human neuroimaging data acquired during encoding or retrieval phases of the recognition memory paradigm. Our novel emphasis is thus on associations rather than dissociations between activity measured in key brain regions; in particular, we propose that measuring the functional coupling between brain regions will help understand how these memory systems interact to guide behavior.     

The Functional Organization of Working Memory Processes Within Human Lateral Frontal Cortex: The Contribution of Functional Neuroimaging

Recent functional neuroimaging studies have provided a wealth of new information about the likely organization of working memory processes within the human lateral frontal cprtex. This article seeks to evaluate the results of these studies in the context of two contrasting theoretical models of lateral frontal-lobe function, developed through lesion and electrophysiological recording work in non-human primates (Goldman-Rakic, 1994, 1995; Petrides, 1994, 1995). Both models focus on a broadly similar distinction between anatomically and cytoarchitectonically distinct dorsolateral and ventrolateral frontal cortical areas, but differ in the precise functions ascribed to those regions. Following a review of the relevant anatomical data, the origins of these two theoretical positions are considered in some detail and the main predictions arising from each are identified. Recent functional neuroimaging studies of working memory processes are then critically reviewed in order to assess the extent to which they support either, or both, sets of predictions. The results of this meta-analysis suggest that lateral regions of the frontal lobe are not functionally organized according to stimulus modality, as has been widely assumed, but that specific regions within the dorsolateral or ventrolateral frontal cortex make identical functional contributions to both spatial and non-spatial working memory.     

Synthetic brain imaging: grasping, mirror neurons and imitation.

The article contributes to the quest to relate global data on brain and behavior (e.g. from PET, Positron Emission Tomography, and fMRI. functional Magnetic Resonance Imaging) to the underpinning neural networks. Models tied to human brain imaging data often focus on a few "boxes" based on brain regions associated with exceptionally high blood flow, rather than analyzing the cooperative computation of multiple brain regions. For analysis directly at the level of such data, a schema-based model may be most appropriate. To further address neurophysiological data, the Synthetic PET imaging method uses computational models of biological neural circuitry based on animal data to predict and analyze the results of human PET studies. This technique makes use of the hypothesis that rCBF (regional cerebral blood flow) is correlated with the integrated synaptic activity in a localized brain region. We also describe the possible extension of the Synthetic PET method to fMRI. The second half of the paper then exemplifies this general research program with two case studies, one on visuo-motor processing for control of grasping (Section 3 in which the focus is on Synthetic PET) and the imitation of motor skills (Sections 4 and 5, with a focus on Synthetic fMRI). Our discussion of imitation pays particular attention to data on the mirror system in monkey (neural circuitry which allows the brain to recognize actions as well as execute them). Finally, Section 6 outlines the immense challenges in integrating models of different portions of the nervous system which address detailed neurophysiological data from studies of primates and other species; summarizes key issues for developing the methodology of Synthetic Brain Imaging; and shows how comparative neuroscience and evolutionary arguments will allow us to extend Synthetic Brain Imaging even to language and other cognitive functions for which few or no animal data are available.     

Emotional memory function, personality structure and psychopathology: a neural system approach to the identification of vulnerability markers

It is well established that emotional events are ingrained stronger into memory relative to neutral events. Facilitated emotional memory is highly variable between individuals within the normal population and is particularly exacerbated in those diagnosed with mood and anxiety disorders. In order to elucidate how variation of enhanced emotional memory within the normal population may manifest into psychopathological states, we explored the convergence between studies investigating the neural systems engaged in emotional memory facilitation and studies investigating how these systems differ from person to person. Converging evidence highlights the roles of three neural systems (1. Amygdala function and attention, 2. Neuroendocrine function, 3. Interactive effects with mood) that all govern emotional memory facilitation and are highly variable between individuals as a function of personality. We applied this neural system approach to models of vulnerability of three forms of psychopathology that are particularly characterized by atypical emotional memory function (depression, generalized anxiety disorder and post-traumatic stress disorder). This application suggests that the incorporation of known vulnerability markers across psychological, neuroimaging and neuroendocrinological domains is cardinal to how susceptibility is conceptualized and assessed in these disorders.     

Revisiting the Neural Architecture of Adolescent Decision-Making: Univariate and Multivariate Evidence for System-Based Models

Understanding adolescent decision-making is significant for informing basic models of neurodevelopment as well as for the domains of public health and criminal justice. System-based theories posit that adolescent decision-making is guided by activity related to reward and control processes. While successful at explaining behavior, system-based theories have received inconsistent support at the neural level, perhaps because of methodological limitations. Here, we used two complementary approaches to overcome said limitations and rigorously evaluate system-based models. Using decision-level modeling of fMRI data from a risk-taking task in a sample of 2000+ decisions across 51 human adolescents (25 females, mean age = 15.00 years), we find support for system-based theories of decision-making. Neural activity in lateral PFC and a multivariate pattern of cognitive control both predicted a reduced likelihood of risk-taking, whereas increased activity in the NAcc predicted a greater likelihood of risk-taking. Interactions between decision-level brain activity and age were not observed. These results garner support for system-based accounts of adolescent decision-making behavior.SIGNIFICANCE STATEMENT Adolescent decision-making behavior is of great import for basic science, and carries equally consequential implications for public health and criminal justice. While dominant psychological theories seeking to explain adolescent decision-making have found empirical support, their neuroscientific implementations have received inconsistent support. This may be partly because of statistical approaches used by prior neuroimaging studies of system-based theories. We used brain modeling, an approach that predicts behavior from brain activity, of univariate and multivariate neural activity metrics to better understand how neural components of psychological systems guide decision behavior in adolescents. We found broad support for system-based theories such that neural systems involved in cognitive control predicted a reduced likelihood to make risky decisions, whereas value-based systems predicted greater risk-taking propensity.     

Spatially Specific Working Memory Activity in the Human Superior Colliculus

Theoretically, working memory (WM) representations are encoded by population activity of neurons with distributed tuning across the stored feature. Here, we leverage computational neuroimaging approaches to map the topographic organization of human superior colliculus (SC) and model how population activity in SC encodes WM representations. We first modeled receptive field properties of voxels in SC, deriving a detailed topographic organization resembling that of the primate SC. Neural activity within human (5 male and 1 female) SC persisted throughout a retention interval of several types of modified memory-guided saccade tasks. Assuming an underlying neural architecture of the SC based on its retinotopic organization, we used an encoding model to show that the pattern of activity in human SC represents locations stored in WM. Our tasks and models allowed us to dissociate the locations of visual targets and the motor metrics of memory-guided saccades from the spatial locations stored in WM, thus confirming that human SC represents true WM information. These data have several important implications. They add the SC to a growing number of cortical and subcortical brain areas that form distributed networks supporting WM functions. Moreover, they specify a clear neural mechanism by which topographically organized SC encodes WM representations.SIGNIFICANCE STATEMENT Using computational neuroimaging approaches, we mapped the topographic organization of human superior colliculus (SC) and modeled how population activity in SC encodes working memory (WM) representations, rather than simpler visual or motor properties that have been traditionally associated with the laminar maps in the primate SC. Together, these data both position the human SC into a distributed network of brain areas supporting WM and elucidate the neural mechanisms by which the SC supports WM.     

Interpretation of Neuroimaging Data Based on Network Concepts

By capturing the actions of distributed brain regions, neuroimaging can give unique insights into the networks underlying complex behavioral and cognitive functions. An approach to interpreting neuroimaging data grounded in emerging ideas in brain network theory is needed to better characterize these large-scale network dynamics. This paper focuses on three concepts germane to this approach to interpretation: “connectivity”, “neural context”, and “small-world properties”. Measures of brain connectivity emphasize the combined action of areas. Functional connectivity analyses focus on interacting neural patterns, whereas effective connectivity analyses uncover directional influences between brain areas. The second concept, neural context, purports that a region’s contribution to a function is more fully appreciated in relation to other coactive brain areas. The final concept is the extension of graph theory measures to the estimation of small-world properties. Measures such as clustering and path length can be used to infer the computational capacity of functional networks. These three constructs are central to the interpretation of neuroimaging data that will further unravel how brain network dynamics guide mental function, and are beginning to be applied to the study of neural disorders.     

Addiction,compulsive drug seeking,and the role of frontostriatal mechanisms in regulating inhibitory control

A principal feature of drug addiction is a reduced ability to regulate control over the desire to procure drugs regardless of the risks involved. Traditional models implicated the neural ‘reward’ system in providing a neurobiological model of addiction. Newer models however, have expanded on this circuitry to include two separate, but interconnecting systems, the limbic system in the incentive sensitization of drugs, and the prefrontal cortex (PFC) in regulating inhibitory control over drug use. Until the recent developments in neuroimaging and brain stimulation techniques, it has been extremely difficult to assess the involvement of the PFC in addiction. In the current review, we explore the involvement of the frontostriatal circuitry in regulating inhibitory control, and suggest how dysregulation of these circuits could be involved in an increased difficulty in ceasing drug use. Following this, we investigate the recent neuropsychological, neuroimaging and brain stimulation studies that explore the presence of these inhibitory deficits, and frontostriatal dysfunctions, across various different substance groups. Further insight into these deficits could contribute to the development of treatment strategies which target these cognitive impairments, and frontostriatal dysfunction, in reducing drug-seeking behaviors.     

Probabilistic language models in cognitive neuroscience: Promises and pitfalls

Cognitive neuroscientists of language comprehension study how neural computations relate to cognitive computations during comprehension. On the cognitive part of the equation, it is important that the computations and processing complexity are explicitly defined. Probabilistic language models can be used to give a computationally explicit account of language complexity during comprehension. Whereas such models have so far predominantly been evaluated against behavioral data, only recently have the models been used to explain neurobiological signals. Measures obtained from these models emphasize the probabilistic, information-processing view of language understanding and provide a set of tools that can be used for testing neural hypotheses about language comprehension. Here, we provide a cursory review of the theoretical foundations and example neuroimaging studies employing probabilistic language models. We highlight the advantages and potential pitfalls of this approach and indicate avenues for future research.     

Functional consequences of model complexity in rhythmic systems: II. Systems performance of model and hybrid oscillators

Neural models are increasingly being used as design components of physical systems. In order to best use models in these novel contexts, we must develop design rules that describe how decisions in model construction relate to the functional performance of the resulting system. In the accompanying paper, we described a series of related neuron models of varying complexity. Here, we use these models to build several half-center oscillators, and investigate how model complexity influences the robustness and flexibility of these oscillators. Our results indicate that model complexity has a significant effect on the robustness and flexibility of systems that incorporate neural models.