首页> 外文会议>International Symposium on Advanced Intelligent Systems >Relationship Between Autonomous and Evoked Activities in a Living Neuronal Network of a Neurorobot, 'Vitaenoid'.
【24h】

Relationship Between Autonomous and Evoked Activities in a Living Neuronal Network of a Neurorobot, 'Vitaenoid'.

机译:神经毒素的活性神经元网络中的自主和诱发活动的关系,“百年胸”。

获取原文

摘要

Toward neuroprostheses, establishing a model system for bi-directional communication between a living neuronal circuit and electrical devices is critical. We developed a neu-rorobot with a dissociated hippocampal culture system. Small neuronal circuits on a culture dish with planar microelectrode arrays were used to detect the electrical activity patterns of a neuronal circuit. Responses of neurons against an input fluctuate by internal autonomous activity. If this fluctuation includes certain information, the interface should detect such embedded information. Thus, handling such fluctuation of autonomous activity is critical, which is the focus of this research. In this study, we proposed neurorobot system as a model system for bidirectional communication between a living neuronal circuit and a robot body. We call this system " Vitaenoid, " which means the robot like (-oid) livs (Vitae) with Neuron (N). To illustrate Vitaenoid, we attempted a self-organizing map (SOM) based method for generating robot behavior. SOM was designed so that the distribution of winner nodes reflected the trend of spatiotemporal electrical patterns. We confirmed that the SOM-based " Vitaenoid " performed a collision avoidance behavior, and the distribution of winner node revealed that some parts of spontaneous activity shared the patterns with evoked responses and suggested that the spontaneous activity and evoked responses were not strictly discriminated. In addition, the ratio of winners shared with spontaneous activity tended to decrease during a 60-min-robot-run. Inputs from outer world to a living neuronal circuit are considered not to be tightly coupled to their response patterns, instead of that, they loosely coupled with the specific activity pattern, which is also expressed as spontaneous activity. We propose this hypothesis as "the loose-coupling hypothesis in input-and-output relationship of the neuronal circuit".
机译:朝向神经调节剂,建立活性神经电路和电气设备之间的双向通信模型系统至关重要。我们开发了一种具有解离海马培养系统的Neu-Rorobot。使用平面微电极阵列的培养皿上的小神经元电路用于检测神经元电路的电活动图案。内部自主活动对神经元对输入波动的反应。如果该波动包括某些信息,则接口应检测此类嵌入信息。因此,处理这种自主活动的这种波动至关重要,这是该研究的重点。在这项研究中,我们提出了神经毒素系统作为生活神经电路和机器人体之间的双向通信的模型系统。我们称这种系统“vitaenoid”,这意味着具有神经元(N)的机器人(-NOID)LIV(vitae)。为了说明维护,我们尝试了一种基于自组织的地图(SOM),用于生成机器人行为。设计了SOM,使获胜者节点的分布反映了时空电气图案的趋势。我们确认基于SOM的“vitaenoid”进行了碰撞避免行为,并且获胜者节点的分布透露,自发活动的某些部分与诱发的反应共享这些模式,并建议没有严格歧视自发活动和诱发的反应。此外,在60分钟机器人运行期间,与自发活动共享的获胜者的比率趋于减少。从外面世界到活性神经元电路的输入被认为不紧密地耦合到它们的响应图案,而不是它们,它们松散地与特定的活性模式耦合,其也被表示为自发性活性。我们提出了这一假设作为“神经元电路的输入和输出关系中的松散耦合假设”。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号