Closed-loop Neuromodulation Research Articles

Challenges and future trends in wearable closed-loop neuromodulation to efficiently treat methamphetamine addiction.

Achieving abstinence from drugs is a long journey and can be particularly challenging in the case of methamphetamine, which has a higher relapse rate than other drugs. Therefore, real-time monitoring of patients' physiological conditions before and when cravings arise to reduce the chance of relapse might help to improve clinical outcomes. Conventional treatments, such as behavior therapy and peer support, often cannot provide timely intervention, reducing the efficiency of these therapies. To more effectively treat methamphetamine addiction in real-time, we propose an intelligent closed-loop transcranial magnetic stimulation (TMS) neuromodulation system based on multimodal electroencephalogram-functional near-infrared spectroscopy (EEG-fNIRS) measurements. This review summarizes the essential modules required for a wearable system to treat addiction efficiently. First, the advantages of neuroimaging over conventional techniques such as analysis of sweat, saliva, or urine for addiction detection are discussed. The knowledge to implement wearable, compact, and user-friendly closed-loop systems with EEG and fNIRS are reviewed. The features of EEG and fNIRS signals in patients with methamphetamine use disorder are summarized. EEG biomarkers are categorized into frequency and time domain and topography-related parameters, whereas for fNIRS, hemoglobin concentration variation and functional connectivity of cortices are described. Following this, the applications of two commonly used neuromodulation technologies, transcranial direct current stimulation and TMS, in patients with methamphetamine use disorder are introduced. The challenges of implementing intelligent closed-loop TMS modulation based on multimodal EEG-fNIRS are summarized, followed by a discussion of potential research directions and the promising future of this approach, including potential applications to other substance use disorders.

A wearable platform for closed-loop stimulation and recording of single-neuron and local field potential activity in freely moving humans

Advances in technologies that can record and stimulate deep brain activity in humans have led to impactful discoveries within the field of neuroscience and contributed to the development of novel therapies for neurological and psychiatric disorders. Further progress, however, has been hindered by device limitations in that recording of single-neuron activity during freely moving behaviors in humans has not been possible. Additionally, implantable neurostimulation devices, currently approved for human use, have limited stimulation programmability and restricted full-duplex bidirectional capability. In this study, we developed a wearable bidirectional closed-loop neuromodulation system (Neuro-stack) and used it to record single-neuron and local field potential activity during stationary and ambulatory behavior in humans. Together with a highly flexible and customizable stimulation capability, the Neuro-stack provides an opportunity to investigate the neurophysiological basis of disease, develop improved responsive neuromodulation therapies, explore brain function during naturalistic behaviors in humans and, consequently, bridge decades of neuroscientific findings across species.

Online Learning Koopman Operator for Closed-Loop Electrical Neurostimulation in Epilepsy.

Electrical neuromodulation as a palliative treatment has been increasingly used in the control of epilepsy. However, current neuromodulations commonly implement predetermined actuation strategies and lack the capability of self-adaptively adjusting stimulation inputs. In this work, rooted in optimal control theory, we propose a Koopman-MPC framework for real-time closed-loop electrical neuromodulation in epilepsy, which integrates i) a deep Koopman operator based dynamical model to predict the temporal evolution of epileptic electroencephalogram (EEG) with an approximate finite-dimensional linear dynamics and ii) a model predictive control (MPC) module to design optimal seizure suppression strategies. The Koopman operator based linear dynamical model is embedded in the latent state space of the autoencoder neural network, in which we can approximate and update the Koopman operator online. The linear dynamical property of the Koopman operator ensures the convexity of the optimization problem for subsequent MPC control. The proposed deep Koopman operator model shows greater predictive capability than the baseline models (e.g., vector autoregressive model, kernel based method and recurrent neural network (RNN)) in both synthetic and real epileptic EEG data. Moreover, compared with the RNN-MPC framework, our Koopman-MPC framework can suppress seizure dynamics with better computational efficiency in both the Jansen-Rit model and the Epileptor model. Koopman-MPC framework opens a new window for model-based closed-loop neuromodulation and sheds light on nonlinear neurodynamics and feedback control policies.

The Emergence of Functional Ultrasound for Noninvasive Brain-Computer Interface.

A noninvasive brain-computer interface is a central task in the comprehensive analysis and understanding of the brain and is an important challenge in international brain-science research. Current implanted brain-computer interfaces are cranial and invasive, which considerably limits their applications. The development of new noninvasive reading and writing technologies will advance substantial innovations and breakthroughs in the field of brain-computer interfaces. Here, we review the theory and development of the ultrasound brain functional imaging and its applications. Furthermore, we introduce latest advancements in ultrasound brain modulation and its applications in rodents, primates, and human; its mechanism and closed-loop ultrasound neuromodulation based on electroencephalograph are also presented. Finally, high-frequency acoustic noninvasive brain-computer interface is prospected based on ultrasound super-resolution imaging and acoustic tweezers.

Dynamic Multi-Graph Convolution based Channel-Weighted Transformer Feature Fusion Network for Epileptic Seizure Prediction.

Electroencephalogram (EEG) based seizure prediction plays an important role in the closed-loop neuromodulation system. However, most existing seizure prediction methods based on graph convolution network only focused on constructing the static graph, ignoring multi-domain dynamic changes in deep graph structure. Moreover, the existing feature fusion strategies generally concatenated coarse-grained epileptic EEG features directly, leading to the suboptimal seizure prediction performance. To address these issues, we propose a novel multi-branch dynamic multi-graph convolution based channel-weighted transformer feature fusion network (MB-dMGC-CWTFFNet) for the patient-specific seizure prediction with the superior performance. Specifically, a multi-branch (MB) feature extractor is first applied to capture the temporal, spatial and spectral representations fromthe epileptic EEG jointly. Then, we design a point-wise dynamic multi-graph convolution network (dMGCN) to dynamically learn deep graph structures, which can effectively extract high-level features from the multi-domain graph. Finally, by integrating the local and global channel-weighted strategies with the multi-head self-attention mechanism, a channel-weighted transformer feature fusion network (CWTFFNet) is adopted to efficiently fuse the multi-domain graph features. The proposed MB-dMGC-CWTFFNet is evaluated on the public CHB-MIT EEG and a private intracranial sEEG datasets, and the experimental results demonstrate that our proposed method achieves outstanding prediction performance compared with the state-of-the-art methods, indicating an effective tool for patient-specific seizure warning. Our code will be available at: https://github.com/Rockingsnow/MB-dMGC-CWTFFNet.

Treating Chronic Pain with Deep Brain Stimulation

Deep brain stimulation (DBS) for chronic pain has been controversial. Despite the discouraging outcomes from multicenter clinical trial in the twentieth century, there is sustained interest in optimizing its use to improve patient outcomes. Here we provide a concise overview of DBS for chronic pain as a reference for clinicians. Recently published data lends tentative support for DBS as a means of treating chronic pain. Still, high level-of-evidence data remain elusive. There are a handful of ongoing and prospective clinical trials exploring DBS for pain in the context of closed-loop neuromodulation, invasive electroencephalography monitoring, stimulation parameters, and novel intracranial targets. DBS is a potentially viable method of treating chronic pain. Procedure success is dependent on a number of factors including proper patient and intracranial target selection. Outcomes for ongoing and future clinical trials will help clinicians refine DBS use for this clinical indication.

Responsive neurostimulation device therapy in pediatric patients with complex medically refractory epilepsy.

Pediatric epilepsy is characterized as drug resistant in 20%-30% of patients and defined as persistent seizures despite adequate treatment with two first-line antiepileptic medications. The American Academy of Neurology advocates surgical options earlier in the treatment of epilepsy to provide long-term seizure reduction. The new development of minimally invasive approaches has recently allowed for surgical options to patients not previously deemed surgical candidates. These may include patients with bilateral, deep, eloquent, or poorly localizing epileptogenic foci. To this end, responsive neurostimulation (RNS) is an FDA-approved closed-loop neuromodulation device for adjuvant treatment of adults with medically intractable epilepsy arising from one or multiple foci. In this study, the authors describe their initial institutional experience with the use of RNS in pediatric patients with drug-resistant epilepsy. An IRB-approved retrospective review was conducted of 8 pediatric patients who underwent RNS implantation at Cincinnati Children's Hospital Medical Center between 2019 and 2021. Eight patients met the inclusion criteria for the study. The average age at the time of surgery was 14.7 years (range 8-18 years) with a mean follow-up of 16.5 months. All patients underwent invasive monitoring with stereo-EEG, subdural grid placement, or a combination of both. All patients had either bilateral or eloquent cortex targets. Trajectories were based on noninvasive (phase 1) and invasive (phase 2) seizure onset zone localization data. Four (50%) of the 8 patients underwent surgical intervention for epilepsy prior to RNS placement. RNS electrodes were placed with robot-assisted guidance in a hybrid operating room with intraoperative CT and electrocorticography. The authors demonstrated individualized RNS electrode trajectory and placement with targets in the amygdala/hippocampus, bilateral insula, bilateral parietal and occipital targets, and frontoparietal regions for a total of 14 implanted electrodes. One adverse event occurred, a wound infection requiring return to the operating room for removal of the RNS implant. All patients demonstrated a reduction in seizure frequency. All patients achieved > 50% reduction in seizure frequency at last follow-up. RNS implantation in carefully selected pediatric patients appears safe and efficacious in reducing seizure burden with a low rate of operative complications.

NeuralTree: A 256-Channel 0.227-μJ/Class Versatile Neural Activity Classification and Closed-Loop Neuromodulation SoC.

Closed-loop neural interfaces with on-chip machine learning can detect and suppress disease symptoms in neurological disorders or restore lost functions in paralyzed patients. While high-density neural recording can provide rich neural activity information for accurate disease-state detection, existing systems have low channel counts and poor scalability, which could limit their therapeutic efficacy. This work presents a highly scalable and versatile closed-loop neural interface SoC that can overcome these limitations. A 256-channel time-division multiplexed (TDM) front-end with a two-step fast-settling mixed-signal DC servo loop (DSL) is proposed to record high-spatial-resolution neural activity and perform channel-selective brain-state inference. A tree-structured neural network (NeuralTree) classification processor extracts a rich set of neural biomarkers in a patient- and disease-specific manner. Trained with an energy-aware learning algorithm, the NeuralTree classifier detects the symptoms of underlying disorders (e.g., epilepsy and movement disorders) at an optimal energy-accuracy trade-off. A 16-channel high-voltage (HV) compliant neurostimulator closes the therapeutic loop by delivering charge-balanced biphasic current pulses to the brain. The proposed SoC was fabricated in 65nm CMOS and achieved a 0.227μJ/class energy efficiency in a compact area of 0.014mm2/channel. The SoC was extensively verified on human electroencephalography (EEG) and intracranial EEG (iEEG) epilepsy datasets, obtaining 95.6%/94% sensitivity and 96.8%/96.9% specificity, respectively. In-vivo neural recordings using soft μECoG arrays and multi-domain biomarker extraction were further performed on a rat model of epilepsy. In addition, for the first time in literature, on-chip classification of rest-state tremor in Parkinson's disease (PD) from human local field potentials (LFPs) was demonstrated.

PC001 / #1148 A MACHINE LEARNING BASED CLOSED-LOOP NEUROMODULATION FRAMEWORK: E-POSTER VIEWING

Closed-loop neuromodulation enables the utilization of dynamic and ongoing neural or physiological activity to guide neuromodulatory interventions for behavioral, cognitive, or performance enhancement. Many non-invasive neuromodulation methods like transcranial electrical stimulation (tES; including tDCS, tACS, tRNS, tPCS) typically implement a one-size-fits-all approach to stimulation dosing and application. Closed-loop stimulation approaches have typically optimized for one stimulation modality, and for error-based signals.

EEG-fNIRS-based hybrid image construction and classification using CNN-LSTM.

The constantly evolving human–machine interaction and advancement in sociotechnical systems have made it essential to analyze vital human factors such as mental workload, vigilance, fatigue, and stress by monitoring brain states for optimum performance and human safety. Similarly, brain signals have become paramount for rehabilitation and assistive purposes in fields such as brain–computer interface (BCI) and closed-loop neuromodulation for neurological disorders and motor disabilities. The complexity, non-stationary nature, and low signal-to-noise ratio of brain signals pose significant challenges for researchers to design robust and reliable BCI systems to accurately detect meaningful changes in brain states outside the laboratory environment. Different neuroimaging modalities are used in hybrid settings to enhance accuracy, increase control commands, and decrease the time required for brain activity detection. Functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG) measure the hemodynamic and electrical activity of the brain with a good spatial and temporal resolution, respectively. However, in hybrid settings, where both modalities enhance the output performance of BCI, their data compatibility due to the huge discrepancy between their sampling rate and the number of channels remains a challenge for real-time BCI applications. Traditional methods, such as downsampling and channel selection, result in important information loss while making both modalities compatible. In this study, we present a novel recurrence plot (RP)-based time-distributed convolutional neural network and long short-term memory (CNN-LSTM) algorithm for the integrated classification of fNIRS EEG for hybrid BCI applications. The acquired brain signals are first projected into a non-linear dimension with RPs and fed into the CNN to extract essential features without performing any downsampling. Then, LSTM is used to learn the chronological features and time-dependence relation to detect brain activity. The average accuracies achieved with the proposed model were 78.44% for fNIRS, 86.24% for EEG, and 88.41% for hybrid EEG-fNIRS BCI. Moreover, the maximum accuracies achieved were 85.9, 88.1, and 92.4%, respectively. The results confirm the viability of the RP-based deep-learning algorithm for successful BCI systems.

Closed-loop seizure modulation via extreme learning machine based extended state observer.

Neuromodulation is a promising way in clinical treatment of epilepsy, but the existing methods cannot adjust stimulations according to patients' real-time reactions. Therefore, it is necessary to acquire a systematic and a scientific regulation method based on patients' real-time reactions. The linear active disturbance rejection control can adapt to complex epileptic dynamics and improve the epilepsy regulation, even if little model information is available, and various uncertainties and external disturbances exist. However, a linear extended state observer estimates the time-varying total disturbance with a steady-state error. To improve regulation, it is crucial to estimate the total disturbance in a more accurate manner. An extreme learning machine is capable of approximating any nonlinear function. Its initial parameter generation is more convenient, adjustable parameters are fewer, and learning speed is faster. Thus, a nonlinear time-varying function can be estimated more timely and accurately. Then, an extreme learning machine based extended state observer is proposed to get a more satisfactory total disturbance estimation and more desired closed-loop regulation. The convergence of the extreme learning machine based extended state observer is verified and the stability of the closed-loop system is analyzed. Numerical results show that the proposed extended state observer is much better than a linear extended state observer in estimating the total disturbance. It guarantees a more satisfied closed-loop neuromodulation.

A lightweight learning-based decoding algorithm for intraneural vagus nerve activity classification in pigs

Objective. Bioelectronic medicine is an emerging field that aims at developing closed-loop neuromodulation protocols for the autonomic nervous system (ANS) to treat a wide range of disorders. When designing a closed-loop protocol for real time modulation of the ANS, the computational execution time and the memory and power demands of the decoding step are important factors to consider. In the context of cardiovascular and respiratory diseases, these requirements may partially explain why closed-loop clinical neuromodulation protocols that adapt stimulation parameters on patient’s clinical characteristics are currently missing. Approach. Here, we developed a lightweight learning-based decoder for the classification of cardiovascular and respiratory functional challenges from neural signals acquired through intraneural electrodes implanted in the cervical vagus nerve (VN) of five anaesthetized pigs. Our algorithm is based on signal temporal windowing, nine handcrafted features, and random forest (RF) model for classification. Temporal windowing ranging from 50 ms to 1 s, compatible in duration with cardio-respiratory dynamics, was applied to the data in order to mimic a pseudo real-time scenario. Main results. We were able to achieve high balanced accuracy (BA) values over the whole range of temporal windowing duration. We identified 500 ms as the optimal temporal windowing duration for both BA values and computational execution time processing, achieving more than 86% for BA and a computational execution time of only ∼6.8 ms. Our algorithm outperformed in terms of BA and computational execution time a state of the art decoding algorithm tested on the same dataset (Vallone et al 2021 J. Neural Eng. 18 0460a2). We found that RF outperformed other machine learning models such as support vector machines, K-nearest neighbors, and multi-layer perceptrons. Significance. Our approach could represent an important step towards the implementation of a closed-loop neuromodulation protocol relying on a single intraneural interface able to perform real-time decoding tasks and selective modulation of the VN.

Automated Tuning of Closed-loop Neuromodulation Control Systems using Bayesian Optimization.

Tuning the parameters of controllers to attain the best performance is a challenging task in designing effective closed-loop neuromodulation systems. In this paper, we present a distributed architecture for automated tuning and adaptation of closed-loop neuromodulation control systems. We use this approach for the automated parameter tuning of a Proportional-Integral (PI) neuromodulation controller using Bayesian optimization. We use a biophysically-grounded mean-field model of neural populations under electrical stimulation as a simulation environment for testing and prototyping the proposed framework and characterizing its performance. Our results demonstrate the feasibility of using Bayesian optimization for performance-based automated tuning of a PI controller in closed-loop set-point neuromodulation control tasks.

An in-vitro system for closed loop neuromodulation of peripheral nerves.

Current neuromodulation research relies heavily on in-vivo animal experiments for developing novel devices and paradigms, which can be costly, time-consuming, and ethically contentious. As an alternative to this, in-vitro systems are being developed for examining explanted tissue in a controlled environment. However, these systems are typically tailored for cellular studies. Thus, this paper describes the development of an in-vitro system for electrically recording and stimulating large animal nerves. This is demonstrated experimentally using explanted pig ulnar nerves, which show evoked compound action potentials (eCAPs) when stimulated. These eCAPs were examined both in the time and velocity domain at a baseline temperature of 20° C, and at temperatures increasing up to those seen in-vivo (37°C). The results highlight that as the temperature is increased within the in-vitro system, faster conduction velocities (CVs) similar to those present in-vivo can be observed. To our knowledge, this is the first time an in-vitro peripheral nerve system has been validated against in-vivo data, which is crucial for promoting more widespread adoption of such systems for the optimisation of neural interfaces.

Computationally efficient neural network classifiers for next generation closed loop neuromodulation therapy - a case study in epilepsy.

This work explores the potential utility of neural network classifiers for real-time classification of field-potential based biomarkers in next-generation responsive neuromodulation systems. Compared to classical filter-based classifiers, neural networks offer an ease of patient-specific parameter tuning, promising to reduce the burden of programming on clinicians. The paper explores a compact, feed-forward neural network architecture of only dozens of units for seizure-state classification in refractory epilepsy. The proposed classifier offers comparable accuracy to filterclassifiers on clinician-labeled data, while reducing detection latency. As a trade-off to classical methods, the paper focuses on keeping the complexity of the architecture minimal, to accommodate the on-board computational constraints of implantable pulse generator systems.Clinical relevance—A neural network-based classifier is presented for responsive neurostimulation, with comparable accuracy to classical methods at reduced latency.

A fast and accurate learning-based decoding algorithm for the classification of cardiovascular and respiratory challenges using intraneural electrodes in the pig vagus nerve.

Bioelectronic medicine is a new approach for developing closed-loop neuromodulation protocols on the peripheral nervous system (PNS) to treat a wide range of disorders currently treated with pharmacological approaches. Algorithms need to have low computational cost in order to acquire, process and model data for the modulation of the PNS in real time. Here, we present a fast learning-based decoding algorithm for the classification of cardiovascular and respiratory functional alterations (i.e., challenges) by using neural signals recorded from intraneural electrodes implanted in the vagus nerve of 5 pigs. Our algorithm relies on 9 handcrafted features, extracted following signal temporal windowing, and a multi-layer perceptron (MLP) for feature classification. We achieved fast and accurate classification of the challenges, with a computational time for feature extraction and prediction lower than 1.5 ms. The MLP achieved a balanced accuracy higher than 80 % for all recordings. Our algorithm could represent a step towards the development of a closed-loop system based on a single intraneural interface with both the potential of real time classification and selective modulation of the PNS.

Adaptive closed-loop control strategy inhibiting pathological basal ganglia oscillations

The presence of pathological basal ganglia oscillations in the beta (12–35 Hz) frequency band is associated with Parkinson’s disease (PD). Suppressing the abnormal beta rhythm can effectively alleviate prominent PD movement disorders such as bradykinesia and rigidity. Brain stimulations, such as deep brain stimulation or transcranial stimulation, are effective therapeutic methods in managing the beta rhythm. However, electrostimulation using the current open-loop paradigm for stimulation is not optimal, especially when the controlled system experiences a substantial unknown disturbance. In this work we propose an adaptive radial basis function (ARBF) neural network strategy to achieve closed-loop brain stimulation based on real-time observed neural oscillation feedback. The underlying system is assumed to be an unknown nonlinear system, and the closed-loop strategy adaptively modulates the stimulation signal to cope with the abnormal neuronal discharge fluctuations, so as to eliminate the beta rhythm of the STN-GPe network. The proposed ARBF neural network closed-loop scheme is tested in a neural mass model composed of the subthalamic nucleus and external globus pallidus. It is shown that the performance of the ARBF controller is robust, including when internal synaptic connections within the basal ganglia network are enhanced to endogenously impact pathological conditions, and also when pathological oscillations are induced by exogenous cortical inputs. Simulation results demonstrate the effectiveness of the proposed closed-loop neuromodulation pattern based on an ARBF neural network. This work may help to develop DBS control systems with adaptive optimization and less network complexity.

Dopaminergic dysfunction and excitatory/inhibitory imbalance in treatment-resistant schizophrenia and novel neuromodulatory treatment.

Antipsychotic drugs are the mainstay in the treatment of schizophrenia. However, one-third of patients do not show adequate improvement in positive symptoms with non-clozapine antipsychotics. Additionally, approximately half of them show poor response to clozapine, electroconvulsive therapy, or other augmentation strategies. However, the development of novel treatment for these conditions is difficult due to the complex and heterogenous pathophysiology of treatment-resistant schizophrenia (TRS). Therefore, this review provides key findings, potential treatments, and a roadmap for future research in this area. First, we review the neurobiological pathophysiology of TRS, particularly the dopaminergic, glutamatergic, and GABAergic pathways. Next, the limitations of existing and promising treatments are presented. Specifically, this article focuses on the therapeutic potential of neuromodulation, including electroconvulsive therapy, repetitive transcranial magnetic stimulation, transcranial direct current stimulation, and deep brain stimulation. Finally, we propose multivariate analyses that integrate various perspectives of the pathogenesis, such as dopaminergic dysfunction and excitatory/inhibitory imbalance, thereby elucidating the heterogeneity of TRS that could not be obtained by conventional statistics. These analyses can in turn lead to a precision medicine approach with closed-loop neuromodulation targeting the detected pathophysiology of TRS.

Activity dependent stimulation increases synaptic efficacy in spared pathways in an anesthetized rat model of spinal cord contusion injury.

Background:Closed-loop neuromodulation systems have received increased attention in recent years as potential therapeutic approaches for treating neurological injury and disease.Objective:The purpose of this study was to assess the ability of intraspinal microstimulation (ISMS), triggered by action potentials (spikes) recorded in motor cortex, to alter synaptic efficacy in descending motor pathways in an anesthetized rat model of spinal cord injury (SCI).Methods:Experiments were carried out in adult, male, Sprague Dawley rats with a moderate contusion injury at T8. For activity-dependent stimulation (ADS) sessions, a recording microelectrode was used to detect neuronal spikes in motor cortex that triggered ISMS in the spinal cord grey matter. SCI rats were randomly assigned to one of four experimental groups differing by: a) cortical spike-ISMS stimulus delay (10 or 25 ms) and b) number of ISMS pulses (1 or 3). Four weeks after SCI, ADS sessions were conducted in three consecutive 1-hour conditioning bouts for a total of 3 hours. At the end of each conditioning bout, changes in synaptic efficacy were assessed using intracortical microstimulation (ICMS) to examine the number of spikes evoked in spinal cord neurons during 5-minute test bouts. A multichannel microelectrode recording array was used to record cortically-evoked spike activity from multiple layers of the spinal cord.Results:The results showed that ADS resulted in an increase in cortically-evoked spikes in spinal cord neurons at specific combinations of spike-ISMS delays and numbers of pulses. Efficacy in descending motor pathways was increased throughout all dorsoventral depths of the hindlimb spinal cord.Conclusions:These results show that after an SCI, ADS can increase synaptic efficacy in spared pathways between motor cortex and spinal cord. This study provides further support for the potential of ADS therapy as an effective method for enhancing descending motor control after SCI.

Closed-loop neuromodulation for studying spontaneous activity and causality.

Having established that spontaneous brain activity follows meaningful coactivation patterns and correlates with behavior, researchers have turned their attention to understanding its function and behavioral significance. We suggest closed-loop neuromodulation as a neural perturbation tool uniquely well suited for this task. Closed-loop neuromodulation has primarily been viewed as an interventionist tool to teach subjects to directly control their own brain activity. We examine an alternative operant conditioning model of closed-loop neuromodulation which, through implicit feedback, can manipulate spontaneous activity at the network level, without violating the spontaneous or endogenous nature of the signal, thereby providing a direct test of network causality.