A goal-driven modular neural network predicts parietofrontal neural dynamics during grasping.

One of the primary ways we interact with the world is using our hands. In macaques, the circuit spanning the anterior intraparietal area, the hand area of the ventral premotor cortex, and the primary motor cortex is necessary for transforming visual information into grasping movements. However, no comprehensive model exists that links all steps of processing from vision to action. We hypothesized that a recurrent neural network mimicking the modular structure of the anatomical circuit and trained to use visual features of objects to generate the required muscle dynamics used by primates to grasp objects would give insight into the computations of the grasping circuit. Internal activity of modular networks trained with these constraints strongly resembled neural activity recorded from the grasping circuit during grasping and paralleled the similarities between brain regions. Network activity during the different phases of the task could be explained by linear dynamics for maintaining a distributed movement plan across the network in the absence of visual stimulus and then generating the required muscle kinematics based on these initial conditions in a module-specific way. These modular models also outperformed alternative models at explaining neural data, despite the absence of neural data during training, suggesting that the inputs, outputs, and architectural constraints imposed were sufficient for recapitulating processing in the grasping circuit. Finally, targeted lesioning of modules produced deficits similar to those observed in lesion studies of the grasping circuit, providing a potential model for how brain regions may coordinate during the visually guided grasping of objects.

Chronaxie Measurements in Patterned Neuronal Cultures from Rat Hippocampus.

Excitation of neurons by an externally induced electric field is a long standing question that has recently attracted attention due to its relevance in novel clinical intervention systems for the brain. Here we use patterned quasi one-dimensional neuronal cultures from rat hippocampus, exploiting the alignment of axons along the linear patterned culture to separate the contribution of dendrites to the excitation of the neuron from that of axons. Network disconnection by channel blockers, along with rotation of the electric field direction, allows the derivation of strength-duration (SD) curves that characterize the statistical ensemble of a population of cells. SD curves with the electric field aligned either parallel or perpendicular to the axons yield the chronaxie and rheobase of axons and dendrites respectively, and these differ considerably. Dendritic chronaxie is measured to be about 1 ms, while that of axons is on the order of 0.1 ms. Axons are thus more excitable at short time scales, but at longer time scales dendrites are more easily excited. We complement these studies with experiments on fully connected cultures. An explanation for the chronaxie of dendrites is found in the numerical simulations of passive, realistically structured dendritic trees under external stimulation. The much shorter chronaxie of axons is not captured in the passive model and may be related to active processes. The lower rheobase of dendrites at longer durations can improve brain stimulation protocols, since in the brain dendrites are less specifically oriented than axonal bundles, and the requirement for precise directional stimulation may be circumvented by using longer duration fields.

Decoding a wide range of hand configurations from macaque motor, premotor, and parietal cortices.

Despite recent advances in decoding cortical activity for motor control, the development of hand prosthetics remains a major challenge. To reduce the complexity of such applications, higher cortical areas that also represent motor plans rather than just the individual movements might be advantageous. We investigated the decoding of many grip types using spiking activity from the anterior intraparietal (AIP), ventral premotor (F5), and primary motor (M1) cortices. Two rhesus monkeys were trained to grasp 50 objects in a delayed task while hand kinematics and spiking activity from six implanted electrode arrays (total of 192 electrodes) were recorded. Offline, we determined 20 grip types from the kinematic data and decoded these hand configurations and the grasped objects with a simple Bayesian classifier. When decoding from AIP, F5, and M1 combined, the mean accuracy was 50% (using planning activity) and 62% (during motor execution) for predicting the 50 objects (chance level, 2%) and substantially larger when predicting the 20 grip types (planning, 74%; execution, 86%; chance level, 5%). When decoding from individual arrays, objects and grip types could be predicted well during movement planning from AIP (medial array) and F5 (lateral array), whereas M1 predictions were poor. In contrast, predictions during movement execution were best from M1, whereas F5 performed only slightly worse. These results demonstrate for the first time that a large number of grip types can be decoded from higher cortical areas during movement preparation and execution, which could be relevant for future neuroprosthetic devices that decode motor plans.

Solving the orientation specific constraints in transcranial magnetic stimulation by rotating fields.

Transcranial Magnetic Stimulation (TMS) is a promising technology for both neurology and psychiatry. Positive treatment outcome has been reported, for instance in double blind, multi-center studies on depression. Nonetheless, the application of TMS towards studying and treating brain disorders is still limited by inter-subject variability and lack of model systems accessible to TMS. The latter are required to obtain a deeper understanding of the biophysical foundations of TMS so that the stimulus protocol can be optimized for maximal brain response, while inter-subject variability hinders precise and reliable delivery of stimuli across subjects. Recent studies showed that both of these limitations are in part due to the angular sensitivity of TMS. Thus, a technique that would eradicate the need for precise angular orientation of the coil would improve both the inter-subject reliability of TMS and its effectiveness in model systems. We show here how rotation of the stimulating field relieves the angular sensitivity of TMS and provides improvements in both issues. Field rotation is attained by superposing the fields of two coils positioned orthogonal to each other and operated with a relative phase shift in time. Rotating field TMS (rfTMS) efficiently stimulates both cultured hippocampal networks and rat motor cortex, two neuronal systems that are notoriously difficult to excite magnetically. This opens the possibility of pharmacological and invasive TMS experiments in these model systems. Application of rfTMS to human subjects overcomes the orientation dependence of standard TMS. Thus, rfTMS yields optimal targeting of brain regions where correct orientation cannot be determined (e.g., via motor feedback) and will enable stimulation in brain regions where a preferred axonal orientation does not exist.

Computationally efficient simulation of electrical activity at cell membranes interacting with self-generated and externally imposed electric fields.

OBJECTIVE: We present a computational method that implements a reduced set of Maxwell's equations to allow simulation of cells under realistic conditions: sub-micron cell morphology, a conductive non-homogeneous space and various ion channel properties and distributions. APPROACH: While a reduced set of Maxwell's equations can be used to couple membrane currents to extra- and intracellular potentials, this approach is rarely taken, most likely because adequate computational tools are missing. By using these equations, and introducing an implicit solver, numerical stability is attained even with large time steps. The time steps are limited only by the time development of the membrane potentials. MAIN RESULTS: This method allows simulation times of tens of minutes instead of weeks, even for complex problems. The extracellular fields are accurately represented, including secondary fields, which originate at inhomogeneities of the extracellular space and can reach several millivolts. We present a set of instructive examples that show how this method can be used to obtain reference solutions for problems, which might not be accurately captured by the traditional approaches. This includes the simulation of realistic magnitudes of extracellular action potential signals in restricted extracellular space. SIGNIFICANCE: The electric activity of neurons creates extracellular potentials. Recent findings show that these endogenous fields act back onto the neurons, contributing to the synchronization of population activity. The influence of endogenous fields is also relevant for understanding therapeutic approaches such as transcranial direct current, transcranial magnetic and deep brain stimulation. The mutual interaction between fields and membrane currents is not captured by today's concepts of cellular electrophysiology, including the commonly used activation function, as those concepts are based on isolated membranes in an infinite, isopotential extracellular space. The presented tool makes simulations with detailed morphology and implicit interactions of currents and fields available to the electrophysiology community.