Computational Model of Recurrent Subthalamo-Pallidal Circuit for Generation of Parkinsonian Oscillations.

Parkinson's disease is a movement disorder caused by dopamine depletion in the basal ganglia. Abnormally synchronized neuronal oscillations between 8 and 15 Hz in the basal ganglia are implicated in motor symptoms of Parkinson's disease. However, how these abnormal oscillations are generated and maintained in the dopamine-depleted state is unknown. Based on neural recordings in a primate model of Parkinson's disease and other experimental and computational evidence, we hypothesized that the recurrent circuit between the subthalamic nucleus (STN) and the external segment of the globus pallidus (GPe) generates and maintains parkinsonian oscillations, and that the cortical excitatory input to the STN amplifies them. To investigate this hypothesis through computer simulations, we developed a spiking neuron model of the STN-GPe circuit by incorporating electrophysiological properties of neurons and synapses. A systematic parameter search by computer simulation identified regions in the space of the intrinsic excitability of GPe neurons and synaptic strength from the GPe to the STN that reproduce normal and parkinsonian states. In the parkinsonian state, reduced firing of GPe neurons and increased GPe-STN inhibition trigger burst activities of STN neurons with strong post-inhibitory rebound excitation, which is usually subject to short-term depression. STN neuronal bursts are shaped into the 8-15 Hz, synchronous oscillations via recurrent interactions of STN and GPe neurons. Furthermore, we show that cortical excitatory input to the STN can amplify or suppress pathological STN oscillations depending on their phase and strength, predicting conditions of cortical inputs to the STN for suppressing oscillations.

Contribution of nucleus accumbens core (AcbC) to behavior control during a learned resting period: introduction of a novel task and lesion experiments.

In recent years, the study of resting state neural activity has received much attention. To better understand the roles of different brain regions in the regulation of behavioral activity in an arousing or a resting period, we developed a novel behavioral paradigm (8-arm food-foraging task; 8-arm FFT) using the radial 8-arm maze and examined how AcbC lesions affect behavioral execution and learning. Repetitive training on the 8-arm FFT facilitated motivation of normal rats to run quickly to the arm tips and to the center platform before the last-reward collection. Importantly, just after this point and before confirmation of no reward at the next arm traverse, locomotor activity decreased. This indicates that well-trained rats can predict the absence of the reward at the end of food seeking and then start another behavior, namely planned resting. Lesions of the AcbC after training selectively impaired this reduction of locomotor activity after the last-reward collection without changing activity levels before the last-reward collection. Analysis of arm-selection patterns in the lesioned animals suggests little influence of the lesion in the ability to predict the reward absence. AcbC lesions did not change exploratory locomotor activity in an open-field test in which there were no rewards. This suggests that the AcbC controls the activity level of planned resting behavior shaped by the 8-arm FFT. Rats receiving training after AcbC lesioning showed a reduction in motivation for reward seeking. Thus, the AcbC also plays important roles not only in controlling the activity level after the last-reward collection but also in motivational learning for setting the activity level of reward-seeking behavior.

Real-time simulation of a spiking neural network model of the basal ganglia circuitry using general purpose computing on graphics processing units.

Real-time simulation of a biologically realistic spiking neural network is necessary for evaluation of its capacity to interact with real environments. However, the real-time simulation of such a neural network is difficult due to its high computational costs that arise from two factors: (1) vast network size and (2) the complicated dynamics of biologically realistic neurons. In order to address these problems, mainly the latter, we chose to use general purpose computing on graphics processing units (GPGPUs) for simulation of such a neural network, taking advantage of the powerful computational capability of a graphics processing unit (GPU). As a target for real-time simulation, we used a model of the basal ganglia that has been developed according to electrophysiological and anatomical knowledge. The model consists of heterogeneous populations of 370 spiking model neurons, including computationally heavy conductance-based models, connected by 11,002 synapses. Simulation of the model has not yet been performed in real-time using a general computing server. By parallelization of the model on the NVIDIA Geforce GTX 280 GPU in data-parallel and task-parallel fashion, faster-than-real-time simulation was robustly realized with only one-third of the GPU's total computational resources. Furthermore, we used the GPU's full computational resources to perform faster-than-real-time simulation of three instances of the basal ganglia model; these instances consisted of 1100 neurons and 33,006 synapses and were synchronized at each calculation step. Finally, we developed software for simultaneous visualization of faster-than-real-time simulation output. These results suggest the potential power of GPGPU techniques in real-time simulation of realistic neural networks.

[A brain-like computer involving emotion and volition].

This paper outlines the concept of "Associative Interacting Intelligence" with regard to a brain-like computer that evolves by learning the relationship between sensory input and behavioral output through active interaction with the real-world environment. We developed the architecture of this intelligent system by using hypotheses based on the established features of brain function. Our central hypotheses are as follows: (1) intellect, emotion, and volition are processed by interactions between parts of the new brain, the old brain, and the body, (2) the heterogeneous hardware architecture of the brain is ontogenically developed, (3) the anatomically modular hardware structure of the brain, (4) reinforcement learning in the brain is motivation-based, and (5) the firing properties of neurons vary. The 2 different levels of our model, which challenges our concepts of the old brain, were proposed using these hypotheses. The spiking neuron model of the basal ganglia circuitry demonstrated the capability of our model to perform probabilistic selections of behavior. In addition, the model also indicated that the selection probability and execution time could be modulated. The parallel modular neural network model for reinforcement learning illustrated the effectiveness of rich internal state representation and internal rewards for achieving a goal in a reduced number of trials. The results of the models showed their potential toward the "Associative Interacting Intelligence", especially regarding efficiency, flexibility, and adaptability. However, many issues, have yet to be addressed and resolved. Our models, which are yet to be tested in detail, must be morphed to a human body in order to demonstrate the ability of these models to reproduce learned abilities. The future success of our study will depend on multidisciplinary collaborations and advances in allied research areas.