作者要感谢博士学生玛塔BISIO培养和维持神经网络，在微电极阵列和滨海南尼博士和Claudia Chiabrera博士NBT-IIT为解剖和分离过程的技术援助。这项研究导致这些结果已收到的资金来自欧盟第七框架计划（ICT-FET FP7 / 2007-2013年FET年轻探险家计划）下的赠款协议N°284772 BRAIN BOW（ <a href="http://www.brainbowproject.eu">www.brainbowproject.eu</a> ）。笔者还想在生产中引入所使用的图形感谢西尔维娅Chiappalone对她的帮助。

1.准备神经文化通过MEA <ol><li>板块神经细胞的MEA芯片，所描述的29。类似的过程的说明，还提供9和讨论部分。 </li><li>转在MEA加热系统上5-10分钟开始记录到经历了由细胞产生的热应力最小化之前:温度控制器的设定温度的目标至37℃，并接通加热板在MEA本身（在大多数商业提供下面MEA系统）和，如果有的话，以显著减少蒸发加热的盖。 </li><li>与湿热（30分钟，130℃）的高压釜中每次使用前消毒气体透过帽。为了避免蒸发，防止改变渗透压，保持盖上瓶盖文化在录制过程中。 </li><li>在实际开始记录将在放大器30分钟的神经元的文化。 注:简单地从INC移动文化ubator的放大器将显著破坏自发活动约半小时。在此期间，培养基的温度将趋于稳定。 </li><li>如果一个carbair电路可用，启动循环carbair的（混合物的5％ 的 CO 2和95％O 2 + N 2），用于实验长于几个小时:培养将需要它，以保持氧和pH水平稳定。 </li></ol> 2.选择录音参数的采集MEA <ol><li>选择软件滤波器带宽检测峰值（ 即多单位活动- MUA）30:在RawDataDisplay形式纪念"300赫兹，3千赫"复选框。 </li><li>开始采集数据:按在RawDataDisplay形式的"开始"按钮。 </li><li>设定的阈值增益尖峰检测的RawDataDisplay 7。 注意:根据所需特异性/选择性折衷和关于通过检测离子算法，此阈值可为6至10倍的计算标准偏差来设置。 </li><li>设置阈值存储到2秒，然后按"锁定"按钮时，没有扣球活动是在显示屏上看到（这两个命令都在RawDataDisplay形式）。标记了"从位数估计SD&#39;以从观测信号31,32的绝对值的中值计算的噪声标准差，如果它是难以提供即使短时间的窗口，没有尖峰活性。从按下"锁定"按钮后，该选项取消勾选标记，作为底层算法是计算密集型的，并可能导致PC滞后。 </li><li>打开尖峰检测程序（"棘波检测"复选框，在RawDataDisplay形式）。如果尖峰检测已经运行（ 即复选框已标记），点击"重置"按钮，在数据记录的形式丢弃检测到这个时候尖峰。 </li></ol> 3.选择MEA电极的刺激神经元文化和响应地图计算<ol><li>在MEA 30分钟的培养神经细胞的记录自发活动:保存数据通过点击"录音"按钮，文件，在DataRecording形式的"尖峰"中，所需的时间量后经过（30分钟，在这种情况下）。 </li><li>找出10个最活跃的通道（ 即 10通道，最高秒杀计数），然后在任意的MEA布局选择那些渠道（无论是在编码，解码和关联图的形式），通过拖动鼠标光标移到需要的地方。一旦通道被选中，右键点击任意位置MEA的布局，并选择"添加到左感觉区"，在弹出的菜单:这些电极将用于在步骤3.5提供电刺激。 </li><li>验证刺激和MEA放大器连接正确:所有配置requIRE每期望刺激通道两根导线，而额外的同轴电缆将需要携带的同步信号（请参阅具体手段在连接图的手册）。继续再打开刺激。 </li><li>定义的关联图的形式刺激参数。所有交付的文化刺激的双相方波电压。设定一半持续时间为300微秒和幅度以1.5V的第 33。 注:刺激足够大，以可靠地唤起的神经反应很可能会折衷活性从同一地点的记录。在本文的其余部分，用于刺激传送电极不应选择记录任何相关信息。 </li><li>记录响应刺激:按在ConnectionMap形式开始按钮。一系列的30刺激，以5秒的间隔，反过来，自动从所选电极中的每一个释放，而响应是记录从剩余的59电极编</li><li>计算一个连接的地图（ 即，观测到来自内从一个不同的电极递送刺激后的时隙的电极的反应的概率），用于通过任何数学软件或SpyCode每个刺激通道，应用程序开发，在过去（和免费提供根据要求）对神经的数据28进行计算。 </li><li>从连接的地图中，选择最好的电极:丢弃所有刺激电极未唤起的响应（ 即 ，在时间窗刺激后，发射率并不比在自发放电显著更高）。 <ol><li>选择，剩下的电极之间，这对有交叠的至少响应。专门为每一个刺激电极，计算出每个记录电极的平均秒杀计数，然后计算相应的电极之间的差异对所有刺激电极对。选择夫妇WHICħ的响应差的绝对值在所有的记录信道的总和是最高的。 </li></ol></li><li>选择这些电极之一，从机器人的左侧，并从右侧代码读数其他代码感官信息:为了做到这一点，拖动鼠标光标放在一个电极，用鼠标右键单击该MEA的布局，然后选择"添加到左侧感觉区"（或"添加到右侧感觉区"）。 </li></ol> 4.接口的神经元文化与机器人:选择编码和解码方案<ol><li>坐落在编码形式为线性"编码类型"。 </li><li>定义最小和最大的刺激率的编码形式。使用0.5-2赫兹的默认范围。 </li><li>在编码形式的"抖动"参数设置为0。 </li><li>坐落在解码格式的解码算法的参数（重量，消光系数）为1，适度活跃的文化（约1穗/每陈荫罴秒EL）。定义一个新的参数对，如果燃烧率强烈和不断地从这个值偏离。对于解码参数的确切功能见讨论。 </li><li>设置在解码算法中的解码表突发参数。重量设定为0（衰减时间则无关紧要），除非尖峰和脉冲之间的区别是研究的课题。 注:在步骤4.4所建议的值会产生平滑的机器人动作和速度与机器人的反应时间为适度活跃的大鼠原代培养兼容。突发参数具有完全相同的功能中的那些步骤4.4中描述的，但该触发事件是突发的，而不是一个尖峰的检测:每个车轮的实际速度是一个简单的检测尖峰和突发的贡献的总和。 </li></ol>导航竞技场的机器人5.设计<ol><li>在虚拟竞技场设计形式，界（竞技场的边界是可见的机器人，我的选择mpassable）或无边（如果从舞台的一侧机器人退出时，它会立即从对面1）舞台上重新输入，并设置在像素竞技场大小。 <ol><li>不使用有界领域超过100×100像素更小，以便允许显著运动。不使用的障碍，其半径小于5个像素，因为它们可以简单地视线机器人的线路之间。 </li><li>请记住，非常大的舞台上的大小可能会导致计算机性能下降:如果需要一个大的舞台上，用在开始难以重复实验前所需的舞台上测试软件的性能。 </li></ol></li><li>设置手动机器人起始位置是（请在虚拟竞技场设计形式，那么所需的位置了"手动选择"按钮），或通过指定机器人在实验坐标开始在"机器人起始位置"字段。 </li><li>添加任何数量的竞技场内不可逾越的障碍。要么把他们的人ually中点击"添加障碍手动"按钮或设置数量和大小的范围后，舞台上。 </li><li>点击"生成竞技场"按钮，生成与所选功能的舞台。也不会发生变化，直到该按钮被按下。 </li><li>保存该设计的舞台，并加载相对文件，在使用前在一项实验中，用命令的按钮在虚拟竞技场设计形式的下部。 </li></ol> 6.选择MEA电极记录神经元电活动的文化<ol><li>选择在步骤3.8临时记录电极那些对不同的反应，已经观察到:这表明刺激后从&#39;左&#39;（或&#39;右&#39;）的电极将构成"左"更大的反应电极（或"右"）记录电极。不使用电极未显示显著反应要么刺激电极来控制机器人。 ˚FROM的MEA布局弹出菜单中选择"添加到左运动区"（或"添加到右侧运动区"）来定义记录电极。 </li><li>选择功能的实验管理的形式来记录。在这个阶段，尖峰和刺激时间标记是唯一的相关信息。 </li><li>进行10分钟的测试驱动器:开始通过单击实验管理形式开始实验按钮运行的机器人，再次单击它10分钟已经过去之后。 注意:一旦所有的参数都被设置（编码和解码，机器人竞技场，刺激和记录电极，拥有创纪录），文件名选择将询问记录文件名称和目的地，然后机器人将开始移动在舞台上，按照定义的规则。所有选定的功能会自动保存在实时。 </li><li>重复步骤3.6和6.1在步骤6.3获得的数据，以便选择与实际的机器人运行期间采集的数据记录的电极（见讨论内容n对于这两个步骤的方法的原理）。 </li></ol> 7.执行神经机器人实验<ol><li>选择在实验管理形成要记录的数据:纪念钉，机器人和刺激数据复选框。 </li><li>启动前学习机器人运行:单击实验管理形式"开始实验"按钮。提示时选择数据文件的新文件名。时30分过去了，再次点击"开始实验"按钮停止机器人运行。 </li><li>接通学习协议（标记的实验管理形式"交付强直刺激命中后"复选框），并作为预培训阶段（ 即 30分）执行相同长度的训练机器人的运行。请记住，输入不同的文件名出现提示时，以避免步7.2覆盖数据。 </li><li>再次关闭学习协议（取消了"交付强直刺激命中后"复选框）并执行后学习机器人的运行。再次，记得更改文件名以防止覆盖。 </li></ol> 8.第二反应地图计算<ol><li>重复步骤3.1-3.6。从这些记录使用数据来检验，如果有任何改变引起的无论是在自发或诱发活动模式的学习方案。 </li></ol>

<ol>
	<li>Fetz, E. E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Volitional+control+of+neural+activity:+implications+for+brain&#8211;computer+interfaces.">Volitional control of neural activity: implications for brain&#8211;computer interfaces.</a> The Journal of Physiology. 579 (3), 571-579 (2007).</li><li>Carmena, J. M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Learning+to+control+a+brain&#8211;machine+interface+for+reaching+and+grasping+by+primates.">Learning to control a brain&#8211;machine interface for reaching and grasping by primates.</a> PLoS Biology. 1 (2), e42 (2003).</li><li>Azin, M., Guggenmos, D. J., Barbay, S., Nudo, R. J., Mohseni, P. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+miniaturized+system+for+spike-triggered+intracortical+microstimulation+in+an+ambulatory+rat.">A miniaturized system for spike-triggered intracortical microstimulation in an ambulatory rat.</a> Biomedical Engineering, IEEE Transactions on. 58 (9), 2589-2597 (2011).</li><li>Angotzi, G. N., Boi, F., Zordan, S., Bonfanti, A., Vato, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+programmable+closed-loop+recording+and+stimulating+wireless+system+for+behaving+small+laboratory+animals.">A programmable closed-loop recording and stimulating wireless system for behaving small laboratory animals.</a> Scientific Reports. 4, (2014).</li><li>Maher, M., Pine, J., Wright, J., Tai, Y. -. C. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+neurochip:+a+new+multielectrode+device+for+stimulating+and+recording+from+cultured+neurons.">The neurochip: a new multielectrode device for stimulating and recording from cultured neurons.</a> Journal of Neuroscience. 87 (1), 45-56 (1999).</li><li>Olsson, R. H., Buhl, D. L., Sirota, A. M., Buzsaki, G., Wise, K. D. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Band-tunable+and+multiplexed+integrated+circuits+for+simultaneous+recording+and+stimulation+with+microelectrode+arrays.">Band-tunable and multiplexed integrated circuits for simultaneous recording and stimulation with microelectrode arrays.</a> Biomedical Engineering, IEEE Transactions on. 52 (7), 1303-1311 (2005).</li><li>Zanos, S., Richardson, A. G., Shupe, L., Miles, F. P., Fetz, E. E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+Neurochip-2:+an+autonomous+head-fixed+computer+for+recording+and+stimulating+in+freely+behaving+monkeys.">The Neurochip-2: an autonomous head-fixed computer for recording and stimulating in freely behaving monkeys.</a> Neural Systems and Rehabilitation Engineering. IEEE Transactions on. 19 (4), 427-435 (2011).</li><li>Reger, B. D., Fleming, K. M., Sanguineti, V., Alford, S., Mussa-Ivaldi, F. A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Connecting+brains+to+robots:+an+artificial+body+for+studying+the+computational+properties+of+neural+tissues.">Connecting brains to robots: an artificial body for studying the computational properties of neural tissues.</a> Artificial Life. 6 (4), 307-324 (2000).</li><li>Hales, C. M., Roston, J. D., Potter, S. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=How+to+culture,+record+and+stimulate+neuronal+networks+on+micro-electrode+arrays+(MEAs).">How to culture, record and stimulate neuronal networks on micro-electrode arrays (MEAs).</a> Journal of Visualized Experiments. (39), 2056 (2010).</li><li>Gross, G., Rieske, E., Kreutzberg, G., Meyer, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+new+fixed-array+multi-microelectrode+system+designed+for+long-term+monitoring+of+extracellular+single+unit+neuronal+activity+in+vitro.">A new fixed-array multi-microelectrode system designed for long-term monitoring of extracellular single unit neuronal activity in vitro.</a> Neurosci Lett. 6 (2), 101-105 (1977).</li><li>Potter, S. M., DeMarse, T. B. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+new+approach+to+neural+cell+culture+for+long-term+studies.">A new approach to neural cell culture for long-term studies.</a> Journal of Neuroscience Methods. 110 (1), 17-24 (2001).</li><li>le Feber, J., Stegenga, J., Rutten, W. L. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+effect+of+slow+electrical+stimuli+to+achieve+learning+in+cultured+networks+of+rat+cortical+neurons.">The effect of slow electrical stimuli to achieve learning in cultured networks of rat cortical neurons.</a> PLoS One. 5 (1), e8871 (2010).</li><li>Marom, S., Shahaf, G. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Development+learning+and+memory+in+large+random+networks+of+cortical+neurons:+lessons+beyond+anatomy.">Development learning and memory in large random networks of cortical neurons: lessons beyond anatomy.</a> Quarterly reviews of biophysics. 35 (1), 63-87 (2002).</li><li>Shahaf, G., Marom, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Learning+in+networks+of+cortical+neurons.">Learning in networks of cortical neurons.</a> The Journal of Neuroscience. 21 (22), 8782-8788 (2001).</li><li>Pelt, J., Vajda, I., Wolters, P. S., Corner, M. A., Ramakers, G. J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Dynamics+and+plasticity+in+developing+neuronal+networks+in+vitro.">Dynamics and plasticity in developing neuronal networks in vitro.</a> Progress in Brain Research. 147, 171-188 (2005).</li><li>Chiappalone, M., Bove, M., Vato, A., Tedesco, M., Martinoia, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Dissociated+cortical+networks+show+spontaneously+correlated+activity+patterns+during+in+vitro+development.">Dissociated cortical networks show spontaneously correlated activity patterns during in vitro development.</a> Brain Research. 1093 (1), 41-53 (2006).</li><li>Tateno, T., Jimbo, Y., Robinson, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Spatio-temporal+cholinergic+modulation+in+cultured+networks+of+rat+cortical+neurons:+spontaneous+activity.">Spatio-temporal cholinergic modulation in cultured networks of rat cortical neurons: spontaneous activity.</a> Neuroscience. 134 (2), 425-437 (2005).</li><li>Masquelier, T., Deco, G. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Network+bursting+dynamics+in+excitatory+cortical+neuron+cultures+results+from+the+combination+of+different+adaptive+mechanism.">Network bursting dynamics in excitatory cortical neuron cultures results from the combination of different adaptive mechanism.</a> PLoS One. 8 (10), e75824 (2013).</li><li>Wagenaar, D. A., Madhavan, R., Pine, J., Potter, S. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Controlling+bursting+in+cortical+cultures+with+closed-loop+multi-electrode+stimulation.">Controlling bursting in cortical cultures with closed-loop multi-electrode stimulation.</a> The Journal of Neuroscience. 25 (3), 680-688 (2005).</li><li>Martinoia, S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Towards+an+embodied+in+vitro+electrophysiology:+the+NeuroBIT+project.">Towards an embodied in vitro electrophysiology: the NeuroBIT project.</a> Neurocomputing. 58-60, 1065-1072 (2004).</li><li>Mussa-Ivaldi, F. A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=New+perspectives+on+the+dialogue+between+brains+and+machines.">New perspectives on the dialogue between brains and machines.</a> Frontiers in Neuroscience. 4 (1), 44 (2010).</li><li>Warwick, K., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Controlling+a+Mobile+Robot+with+a+Biological+Brain.">Controlling a Mobile Robot with a Biological Brain.</a> Defence Science Journal. 60 (1), (2010).</li><li>Kudoh, S. N., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Vitroid&#8211;the+robot+system+with+an+interface+between+a+living+neuronal+network+and+outer+world.">Vitroid&#8211;the robot system with an interface between a living neuronal network and outer world.</a> International Journal of Mechatronics and Manufacturing Systems. 4 (2), 135-149 (2011).</li><li>Jimbo, Y., Tateno, T., Robinson, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Simultaneous+induction+of+pathway-specific+potentiation+and+depression+in+networks+of+cortical+neurons.">Simultaneous induction of pathway-specific potentiation and depression in networks of cortical neurons.</a> Biophysical Journal. 76 (2), 670-678 (1999).</li><li>Chiappalone, M., Massobrio, P., Martinoia, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Network+plasticity+in+cortical+assemblies.">Network plasticity in cortical assemblies.</a> European Journal of Neuroscience. 28 (1), 221-237 (2008).</li><li>Mulas, M., Massobrio, P., Martinoia, S., Chiappalone, M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+simulated+neuro-robotic+environment+for+bi-directional+closed-loop+experiments.">A simulated neuro-robotic environment for bi-directional closed-loop experiments.</a> Paladyn. 1 (3), 179-186 (2010).</li><li>Novellino, A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Connecting+neurons+to+a+mobile+robot:+an+in+vitro+bidirectional+neural+interface.">Connecting neurons to a mobile robot: an in vitro bidirectional neural interface.</a> Computational Intelligence and Neuroscience. 2007 (12725), 1-13 (2007).</li><li>Bologna, L. L., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Investigating+neuronal+activity+by+SPYCODE+multichannel+data+analyzer.">Investigating neuronal activity by SPYCODE multichannel data analyzer.</a> Neural Networks. 23 (6), 685-697 (2010).</li><li>Frega, M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Cortical+cultures+coupled+to+micro-electrode+arrays:+a+novel+approach+to+perform+in+vitro+excitotoxicity+testing.">Cortical cultures coupled to micro-electrode arrays: a novel approach to perform in vitro excitotoxicity testing.</a> Neurotoxicology and Teratology. 34 (1), 116-127 (2012).</li><li>Quian Quiroga, R., Panzeri, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Extracting+information+from+neuronal+populations:+information+theory+and+decoding+approaches.">Extracting information from neuronal populations: information theory and decoding approaches.</a> Nature Reviews Neuroscience. 10 (3), 173-185 (2009).</li><li>Donoho, D. L., Johnstone, J. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Ideal+spatial+adaptation+by+wavelet+shrinkage.">Ideal spatial adaptation by wavelet shrinkage.</a> Biometrika. 81 (3), 425-455 (1994).</li><li>Martinez, J., Pedreira, C., Ison, M. J., Quian Quiroga, R. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Realistic+simulation+of+extracellular+recordings.">Realistic simulation of extracellular recordings.</a> Journal of neuroscience methods. 184 (2), 285-293 (2009).</li><li>Wagenaar, D. A., Pine, J., Potter, S. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Effective+parameters+for+stimulation+of+dissociated+cultures+using+multi-electrode+arrays.">Effective parameters for stimulation of dissociated cultures using multi-electrode arrays.</a> Journal of Neuroscience Methods. 138 (1-2), 27-37 (2004).</li><li>Tessadori, J., Mulas, M., Martinoia, S., Chiappalone, M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=.">.</a> Biomedical Robotics and Biomechatronics (BioRob), 2012 4th IEEE RAS &amp; EMBS International Conference on. , 332-337 (2012).</li><li>Tessadori, J., Bisio, M., Martinoia, S., Chiappalone, M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Modular+neuronal+assemblies+embodied+in+a+closed-loop+environment:+toward+future+integration+of+brains+and+machines.">Modular neuronal assemblies embodied in a closed-loop environment: toward future integration of brains and machines.</a> Frontiers in Neural Circuits. 6, (2012).</li><li>Tessadori, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Neural+Engineering+(NER).">Neural Engineering (NER).</a> , 899-902 (2013).</li><li>Colombi, I., Mahajani, S., Frega, M., Gasparini, L., Chiappalone, M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Effects+of+antiepileptic+drugs+on+hippocampal+neurons+coupled+to+micro-electrode+arrays.">Effects of antiepileptic drugs on hippocampal neurons coupled to micro-electrode arrays.</a> Frontiers in Neuroengineering. 6, 10 (2013).</li><li>Gal, A., Marom, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Entrainment+of+the+intrinsic+dynamics+of+single+isolated+neurons+by+natural-like+input.">Entrainment of the intrinsic dynamics of single isolated neurons by natural-like input.</a> Journal of Neuroscience. 33 (18), 7912-7918 (2013).</li><li>Mainen, Z. F., Sejnowski, T. J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Reliability+of+spike+timing+in+neocortical+neurons.">Reliability of spike timing in neocortical neurons.</a> Science. 268 (5216), 1503-1506 (1995).</li><li>Maccione, A., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+novel+algorithm+for+precise+identification+of+spikes+in+extracellularly+recorded+neuronal+signals.">A novel algorithm for precise identification of spikes in extracellularly recorded neuronal signals.</a> Journal of neuroscience methods. 177 (1), 241-249 (2009).</li><li>DeMarse, T. B., Wagenaar, D. A., Blau, A. W., Potter, S. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+neurally+controlled+animat:+biological+brains+acting+with+simulated+bodies.">The neurally controlled animat: biological brains acting with simulated bodies.</a> Autonomous Robots. 11 (3), 305-310 (1023).</li><li>Bakkum, D. J., Chao, Z. C., Potter, S. M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Spatio-temporal+electrical+stimuli+shape+behavior+of+an+embodied+cortical+network+in+a+goal-directed+learning+task.">Spatio-temporal electrical stimuli shape behavior of an embodied cortical network in a goal-directed learning task.</a> Journal of Neural Engineering. 5 (3), 310-323 (2008).</li><li>Truccolo, W., Hochberg, L. R., Donoghue, J. P. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Collective+dynamics+in+huma+and+monkey+sensorimotor+cortex:+predicting+single+neuron+spikes.">Collective dynamics in huma and monkey sensorimotor cortex: predicting single neuron spikes.</a> Nature Neuroscience. 13 (1), 105-111 (2010).</li></ol>

The authors declare they have no competing financial interests.

在本文中一个神经机器人构造基于神经控制器（ 即 ，神经元从皮质或胚胎大鼠的海马未来的网络），双向连接到一个虚拟机器人装置，提出。机器人，它有传感器和车轮，被迫在一个静态的舞台上与移动障碍，其任务包括避免碰撞。 所描述的过程的第一个和可能最关键的方面是在培养物的制备本身，故障率将趋于即使在最好的技术条件是显著。培养技术的详细描述，然而，在本工作范围之外。作为一般准则，当网络燃烧率达到一个稳定的水平，通常为3周后在体外 ，录音应发生。健康文化的一个粗略指标是自发的电生理活动的若干记录陈荫罴存在ELS（至少20个频道60多个可用）。此类培养物的特征在于神经元间的连接高度。在这种情况下，神经活动通常变得强烈同步，有时，它会显示痫性活动，具有强烈的扣球其次分钟长的静默期37。这两项功能提出一个问题:过度同步将使得不可能区分响应来自不同电极的刺激，而培养物表现出癫痫样活动将要呈现的活动的长的突发的第一个刺激，随后是无声期间响应，而不管任何后续的刺激交付。这两个问题可以通过图案化的培养物35，其中，所述神经元群被分成两个或更多个相互连接的弱子组的就业大力改善。 另一个问题是，神经反应强烈地依赖于传入stimul分布我38,39。在一个封闭的循环试验，在递送刺激是所述传感器的读数，这反过来，是机器人运动的结果，因此，神经反应本身的功能。这意味着，有建立预先什么响应将实际的实验过程中观察到的没有简单的方法。因此，输入输出电极的选择必须依靠逐次逼近。在所描述的协议中，一个两步骤的过程（即，步骤5.5和6.4）被实现来尝试确定的连接图。在第一步骤中，刺激的一个常规程序被输送和响应这种刺激被用于导出一个第一连接地图和一个临时设定的记录信道。这种构造，然后用于执行在步骤6.4中所述的测试驱动器，并选择将实际实验期间监视的记录信道。 在结果部分，具有代表性的导航RESULT，并通过引入学习范式所带来的直接的改进提出。整个协议的描述，其他几个可能的实验变种被提及。例如，在实施HyBrainWare2两大编码系统（线性和随机的），其中开发研究的颞输入变异对神经代码38的确切作用。在直线的情况下，瞬时的刺激率是用户定义的参数和机器人的传感器记录的功能。在随机的情况下，每次实例有一个给定的概率被选择以提供一个刺激。这样概率由HyBrainWare2自动计算，使得预期的刺激速率匹配前一种情况的。的可能性增加的抖动到线性编码提供上述两种情况之间的平滑过渡。以同样的方式，探索不同的参数组合在解码部分可能有助于阐明的PRECI光绽放在体外神经网络本身的作用。机器人的各车轮的速度成比例的事件在相应的输出区域在每次检测到增加的权重参数，而消光系数指示了多少时间，以秒的贡献之一发生失去其值的50％ 。衰减是一个简单的指数。这些变体已经被考虑到HyBrainWare2的当前的设计，但是有更多的研究的可能性是开放如果在软件或实验设置额外的修饰可以被引入。 这里所描述的协议的一个，而显著的限制是开发的定制软件，HyBrainWare2（自由可应要求提供给所有感兴趣的用户）的要求。该软件已被设计为一组特定的型号和制造商的设备（刺激，采集板，MEA放大器）的。虽然它确实有可能去适应它的工作在迪fferent设置，转换需要一定的编程技巧。同样，包括选择只覆盖了一组所有可能通过这样的设置进行调查实验的问题有限。例如，在所提出的架构实现的尖峰检测算法（精确的时间尖峰检测40）在需要改变，如果登记的受试者是从解离的神经网络（ 例如 ，心肌细胞非常不同的几个硬编码参数是完全限定或片）。最后，该学习协议包括短，频率高的交付（2秒的20赫兹的刺激，每个刺激脉冲的相同的那些用于编码感觉信息）的刺激下各障碍物命中。如果机器人击中其右侧的障碍物时，强直性刺激递送至通常为右侧信息码和同样适用于左侧命中的电极。这个协议是硬编码的，不能B用户无需修改退出软件所改变。 而这里提出的设置不是第一个具体化系统用于混合，闭环实验20,23,27,41,42，那些设计在过去都集中在从类似制剂的有限数目由数据的支持单个论文。另一方面，所描述的设置已被用于大量的实验（2012年超过100培养物已被记录）配制剂为不同的模块化和原点，而实验本身寻址不同的问题（ 如强直性刺激的影响，并爆破的相关性，如在目前的成果）。在不久的将来，新实验会议预见到验证既强直和闭环刺激的持续效果及刺激规律性之间的关系，并观察响应。另一个问题要解决是自发的，观察到的活动之间的联系，与加入新的解码算法，考虑到过去的活性和刺激43的历史的想法。

脑功能的许多功能目前无法复制的人工系统。大脑的快速处理复杂的感官信息，并生成，在反应能力，精确的电机命令是​​本身就已经超越了目前的先进设备，最先进的。但它从过去的经验中学习，以适应不同条件的能力使得它如此大大优于人类开发的控制系统。到目前为止，试图复制或利用这种可塑性遇到收效甚微，与大脑的内部运作的理解已经躲避科学家的把握。其中一个主要的问题，同时调查大脑和行为之间的关系是不能正确地访问系统中所有的变量:理想地，最佳的实验装置将允许同时记录和刺激到大量的神经元的，长期稳定的，监测突触位置和重量，可控双向的directio与环境的相互作用纳尔。在同时跟踪所有这些变量的困难导致的脑行为的关系的研究中，在两个完全不同的尺度:无论是与行为的动物，没有很好地控制的实验条件下1-7或具有小的，孤立的部分，如部分神经元组织，与系统8的无整体图。在后一种情况下，虽然没有设计出的实验装置允许完全监测所有涉及即使是简单的神经网络的运作参数的，良好的折衷是由生长在微电极阵列（MEAs）的9离解的神经元提供的。这些设备，出生于70年代10的端，具有比传统的电生理学技术几个优点:第一，记录的可能性，并刺激一次（通常为60电极）一神经网络中的许多不同位置。此外，的MEA与细胞的结合几乎是无创，允许观察在同一网络长时间的时间，最多的数月11。电刺激对解离的培养物的生理作用已被广泛研究由于这些设备，揭示了许多性能在较高尺度观察（如，例如，塑性和简单的记忆处理12-14）是保守的，尽管结构的损失。在培养中生长，这些网络开始显示在体外 （DIV）7天左右15,16自发活动。网络活动往往具有进一步增长从根本上改变;首先作为单一尖峰聚集成脉冲串（朝第二周结束）17，后来由于它改变成具有高度复杂的同步，非周期的网络模式脉冲串18，这代表一个网络的成熟状态。有人建议19，该同步行为，有点类似，在体内观察到的记录上睡觉动物英格斯，由缺乏感觉输入造成的。 一种不同的方法试图更好地理解信息编码已采取通过进行闭环实验，其中不同类型的信号被用来控制所述神经网络本身11,20-23的刺激。在这些实验中，能够与环境的相互作用的外部代理已被用于产生供给至神经网络，其中，反过来，产生电机的命令的执行机构的感觉信息。这允许如何神经系统的动态和自适应特性响应于感应环境的变化演变意见。 一个设置执行"体现神经生理学"实验，开发，其中一个轮子的传感器平台（物理机器人或它的虚拟模型）在舞台上移动约而其速度分布是由神经元的活性确定系统（ 即鼠神经元的人口超过一MEA培养）。机器人的特征在于，它的两个独立控制的车轮的速度分布和通过的距离传感器的当前读数。的距离传感器的确切性质是不相关的;它们可以是有源或无源光学传感器或超声波传感器。显然，这一问题并不适用于虚拟机器人，其中传感器可以被设计为具有任何所需的功能的情况下。 在本文描述的实验中，所使用的机械手总是虚拟实现，用6距离传感器指向30°，从机器人朝着两个方向60°和90°。三个左和右传感器的活性的平均值和生物培养物的活性是通过这样的"超级传感器"（这将只是被称为"左"，并在剩下的"正确"的传感器所收集的信息驱动这项工作）。在原描述栏实际上可以应用于具有相当小的调整的物理机械手。由机器人（物理或虚拟）收集的信息被编码在一系列被用来操纵所述生物神经网络，这在物理上由机器人分开的活性的刺激。刺激本身都是相同的，因此不进行编码的任何信息。什么是相关的是其频率:刺激速率增加时，机器人接近的障碍物时，与不同的递送​​位点的编码，从机器人的左和右的"眼睛"的感觉信息。神经网络将呈现不同的响应于输入刺激的列车:解码算法的任务是向所得的网络活动转化为用于控制机器人的轮子的命令。给定一个"完美"的网络行为（ 例如 ，具有可靠，完全分开的反应从不同的电极刺激），这将resul吨机器人驾驶的舞台上没有击中任何障碍。大多数网络目前行为的理想有很大不同，因此一个简单的学习方案介绍:在激活时，强直刺激（高频刺激短暂时间，20赫兹刺激2秒，设计灵感来自于24,25描述协议）继与障碍物碰撞被传递。如果强直性刺激导致局部加强的网络连通性，这将导致在所述机器人的导航功能的逐步增加。 HyBrainWare2，发表于26的定制软件的改进版本，是开发来处理系统（刺激器，数据采集，处理和可视化，机器人通信或模拟）的不同设备的控制的核心架构。该软件已经开发我们的实验室，并免费索取。该软件提供与接口数据采集​​板:一旦用户开始数据采集从GUI，该软件控制所述采集板以开始采样和数据从记录电极来的A / D转换。该数据然后可被记录，显示到屏幕或实时分析，以检测峰值，根据由用户设置的选项（详见程序部分）。此外，软件内，编码（翻译的感觉信息转换成电刺激）和解码（译记录活动的进电机的命令对机器人）的定义的算法必须被指定。特别是，我们的设置是比较方便相比设计在过去的27中，由于几乎所有的变量都可以由用户访问的类似系统在开始实际实验之前右，而所有的记录的信息被自动保存在与兼容的格式神经数据分析工具箱28。 以下步骤部分描述了一个学习实验上分离的大鼠海马培养:所有的培养和实验参数提供了用于这种特定制剂，并可能需要进行修改，如果需要不同的生物基片是要使用的。类似地，所描述的实验中利用闭环架构的调查强直性刺激的学习效果，但结构本身是足够灵活的解离的神经网络的不同特性的研究中使用。所提出的实验主要变种在讨论部分的进一步解释。

<table><tbody><tr><td>Stimulus Generator 4002</td><td>Multi Channel Systems</td><td></td><td></td></tr><tr><td>MEA1060-Inv-BC</td><td>Multi Channel Systems</td><td></td><td></td></tr><tr><td>TC02</td><td>Multi Channel Systems</td><td></td><td></td></tr><tr><td>NI 6255 Acquisition Card</td><td>National Instruments</td><td></td><td></td></tr><tr><td>Microsoft Visual Studio 2008</td><td>Microsoft</td><td></td><td></td></tr><tr><td>2078P Multichannel System-National Instruments adapter board</td><td>Developed at University of Genova (Italy)</td><td></td><td></td></tr><tr><td>Matlab 2010</td><td>Mathworks</td><td></td><td></td></tr><tr><td>HyBrainWare2</td><td></td><td></td><td>&lt;a href=&quot;https://docs.google.com/forms/d/1ffdr_a5OhzuuAduiPvoLJIvR4x_fDfsKvuM6RcRPLhU/viewform&quot;&gt;HyBrainWare2: Contact Information&lt;/a&gt;</td></tr></tbody></table>

closed-loop neuro-robotic experiments to test computational properties of neuronal networks

Information coding in the Central Nervous System (CNS) remains unexplored. There is mounting evidence that, even at a very low level, the representation of a given stimulus might be dependent on context and history. If this is actually the case, bi-directional interactions between the brain (or if need be a reduced model of it) and sensory-motor system can shed a light on how encoding and decoding of information is performed. Here an experimental system is introduced and described in which the activity of a neuronal element (i.e., a network of neurons extracted from embryonic mammalian hippocampi) is given context and used to control the movement of an artificial agent, while environmental information is fed back to the culture as a sequence of electrical stimuli. This architecture allows a quick selection of diverse encoding, decoding, and learning algorithms to test different hypotheses on the computational properties of neuronal networks.

的开发实验框架允许的信息的神经元的培养和虚拟实现的物理机器人之间的交换是否是可能的测试34。 图1示出了在20分钟的实验在不同的条件下由虚拟机器人行进获得路径的几个样品:从左至右适当闭环试验，一个"空的MEA&#39;机器人实验（无细胞铺在MEA此控制实验）和一个开环机器人实验（刺激速率是代替编码传感器信息常数）表示。代表轨迹确认神经元和人造元件之间的双向相互作用是必要的，以便获得所述机器人的良好的导航性能。然而，机器人经历对障碍物几命中。 在图2中 ，机器人的导航性能，电子xpressed作为后续的命中之间行进的像素中，示出在不同的条件。前两列显示在上述（&#39;空MEA&#39;和开环配置）中提到的对照实验行驶距离的分布，而第三和第四列显示性能不具有和具有分别强直性刺激的以下各输送撞到障碍物。的强直性刺激（参见步骤，点7.2）的引入显著提高两个连续命中之间的行驶距离，因此提高了机械手35的导航性能。 在图3中 ，用不同的解码条件的机器人的导航性能呈现。为此，不同的舞台结构已经通过。这有助于量化竞技场36内的机器人的成功导航:如在过程的步骤4.1所描述的，机器人提出了一系列的短轨道。成功率是简单地通过给出的轨道数成功越过的轨道数的比率。特别是，在实验过程中，进行突发分离和尖峰的一个实时识别。解码范式从一个不同，因为突发和孤立的尖峰的相对权重的另一个（参见程序，指向3.5-3.6和讨论）。 <img alt="图1" src="/files/ftp_upload/52341/52341fig1highres.jpg" width="700" /> 图1:一个神经网络和机器人具有的信息的双向交换该图显示由机器人在20分钟的实验行进3代表路径。特别是浅绿色的区域都是免费的机器人移动，而深绿色像素代表了不可逾越的障碍，该机器人可以通过距离传感器感知。在每次试验中，机器人开始我n个主场的左上部分，并前往其最终位置，描绘成一个大的粉红色圆点。较小的黑点代表碰上障碍物，而彩色编码路径提供机器人运动的时间演变的指示。右边的颜色条表示经过的时间，从该实验在几分钟的开头。（A）的路径，随后闭环实验期间。（B）的路径，随后一个"空"的MEA实验期间（无细胞是在MEA中培养表面）。（C）的路径，随后开环试验过程中（刺激率在整个实验常数）。 <a href="/files/ftp_upload/52341/52341fig1highres.jpg" target="_blank">请点击此处查看图的放大版本。</a> <img alt="图2" src="/files/ftp_upload/52341/52341fig2.jpg" /> 图2:性能结果受封闭-loop和强直性刺激。该图报告涵盖在不同的条件下随后的命中之间的机器人的距离的分布。特别是，在第一两个分布代表对照实验（&#39;空&#39;为录音没有神经元镀在MEA中，"开环"（OL）为实验用&#39;盲&#39;机器人）。最后两列所代表的&#39;闭环&#39;条件无（CL）和带（CL + TS）通过强直性刺激（TS）的实施的学习协议，被输送到下面的针对障碍物的命中的培养。在每个框中，中央水平段代表该分布中，空的平方的平均值，该中心杆延伸到第一和第三个四分位数和晶须延伸到第 5和第 95百分位数的中值。离群表示为钻石。采用方差的Kruskall-Wallis单因素分析统计已经完成秩:学生 - 纽曼两比较两两比较发现，所有的中位值与P &lt;0.05显著不同。 <img alt="图3" src="/files/ftp_upload/52341/52341fig3.jpg" /> 图3:解码影响机器人性能上面的图表示的概率，对于给定的解码算法中，机器人成功通过短轨道在一个有限的时间导航。实验本身期间实时地执行脉冲串和孤立的尖峰的鉴定。在第一种情况下（&#39;尖峰&#39;）的所有检测到的尖峰呈现相同的相对权重，对于第二和第三分布为0的重量为，分别设置的，孤立的尖峰和尖峰属于一个突发。最后两列表示在所有的尖峰会占的情况下获得的结果，但使用不同的相对权重，根据它们的位置。特别是在第四列中分离尖峰被给予比突发事件更高的相对权重，而加权倒相对于第五列数据的解码。在每个曲线图中，中央线表示的分布，空平方的平均值的中值，而中央杆延伸到第一和第三个四分位数和晶须延伸到第 5和第 95百分位数。利用方差的Kruskall-Wallis单因素分析上行列统计已经完成。的成对比较（学生 - 纽曼-Keuls法，）显示，显著差异（p &lt;0.05），可以将列1和4,2和4,2和5,2和3之间进行观察。

Watch this Scientific Journal Video about Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks at JoVE.com

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

In this paper, an experimental framework to perform closed-loop experiments is presented, in which information processing (i.e., coding and decoding) and learning of neuronal assemblies are studied during the continuous interaction with a robotic body.

闭环神经机器人实验来检验神经网络的计算性能

Information coding in the Central Nervous System (CNS) remains unexplored. There is mounting evidence that, even at a very low level, the representation ...

closed-loop-neuro-robotic-experiments-to-test-computational

Universidad San Sebastian

Research

JoVE Journal

Neuroscience

10.2K 次观看.  Istituto Italiano di Tecnologia. In this paper, an experimental framework to perform closed-loop experiments is presented, in which information processing (i.e., coding and decoding) and learning of neuronal assemblies are studied during the continuous interaction with a robotic body.

视频: Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

An Experimental Platform to Study the Closed-loop Performance of Brain-machine Interfaces

The non-stationary nature and variability of neuronal signals is a fundamental problem in brain-machine interfacing. We developed a brain-machine interface to assess the robustness of different control-laws applied to a closed-loop image stabilization task. Taking advantage of the well-characterized fly visuomotor pathway we record the electrical activity from an identified, motion-sensitive neuron, H1, to control the yaw rotation of a two-wheeled robot. The robot is equipped with 2 high-speed video cameras providing visual motion input to a fly placed in front of 2 CRT computer monitors. The activity of the H1 neuron indicates the direction and relative speed of the robot's rotation. The neural activity is filtered and fed back into the steering system of the robot by means of proportional and proportional/adaptive control. Our goal is to test and optimize the performance of various control laws under closed-loop conditions for a broader application also in other brain machine interfaces.

We use a closed-loop fly-machine interface to investigate general principles in neuronal control.

研究脑机接口的闭环性能的实验平台

The non-stationary nature and variability of neuronal signals is a fundamental problem in brain-machine interfacing. We developed a brain-machine ...

科研

神经科学

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Brain-Robot Interaction (BRI), which provides an innovative communication pathway between human and a robotic device via brain signals, is prospective in helping the disabled in their daily lives. The overall goal of our method is to establish an SSVEP-based experimental procedure by integrating multiple software programs, such as OpenViBE, Choregraph, and Central software as well as user developed programs written in C++ and MATLAB, to enable the study of brain-robot interaction with humanoid robots.
This is achieved by first placing EEG electrodes on a human subject to measure the brain responses through an EEG data acquisition system. A user interface is used to elicit SSVEP responses and to display video feedback in the closed-loop control experiments. The second step is to record the EEG signals of first-time subjects, to analyze their SSVEP features offline, and to train the classifier for each subject. Next, the Online Signal Processor and the Robot Controller are configured for the online control of a humanoid robot. As the final step, the subject completes three specific closed-loop control experiments within different environments to evaluate the brain-robot interaction performance.
The advantage of this approach is its reliability and flexibility because it is developed by integrating multiple software programs. The results show that using this approach, the subject is capable of interacting with the humanoid robot via brain signals. This allows the mind-controlled humanoid robot to perform typical tasks that are popular in robotic research and are helpful in assisting the disabled.

The overall goal of this method is to establish an SSVEP-based experimental procedure by integrating multiple software programs to enable the study of brain-robot interaction with humanoid robots, which is prospective in assisting the sick and elderly as well as performing unsanitary or dangerous jobs.

SSVEP基础实验步骤与人形机器人脑机器人互动

Brain-Robot Interaction (BRI), which provides an innovative communication pathway between human and a robotic device via brain signals, is prospective in ...

工程学

A Fully Automated and Highly Versatile System for Testing Multi-cognitive Functions and Recording Neuronal Activities in Rodents

We have developed a fully automated system for operant behavior testing and neuronal activity recording by which multiple cognitive brain functions can be investigated in a single task sequence. The unique feature of this system is a custom-made, acoustically transparent chamber that eliminates many of the issues associated with auditory cue control in most commercially available chambers. The ease with which operant devices can be added or replaced makes this system quite versatile, allowing for the implementation of a variety of auditory, visual, and olfactory behavioral tasks. Automation of the system allows fine temporal (10 ms) control and precise time-stamping of each event in a predesigned behavioral sequence. When combined with a multi-channel electrophysiology recording system, multiple cognitive brain functions, such as motivation, attention, decision-making, patience, and rewards, can be examined sequentially or independently.

In this report, we present a fully automated and highly versatile system capable of simultaneously testing multi-cognitive behaviors and recording neuronal activities for rodents.

一个多认知功能测试，并记录在啮齿动物的神经元活动的完全自动化和高度灵活的系统

We have developed a fully automated system for operant behavior testing and neuronal activity recording by which multiple cognitive brain functions can be ...

Designing and Implementing Nervous System Simulations on LEGO Robots

We present a method to use the commercially available LEGO Mindstorms NXT robotics platform to test systems level neuroscience hypotheses. The first step of the method is to develop a nervous system simulation of specific reflexive behaviors of an appropriate model organism; here we use the American Lobster. Exteroceptive reflexes mediated by decussating (crossing) neural connections can explain an animal's taxis towards or away from a stimulus as described by Braitenberg and are particularly well suited for investigation using the NXT platform.1 The nervous system simulation is programmed using LabVIEW software on the LEGO Mindstorms platform. Once the nervous system is tuned properly, behavioral experiments are run on the robot and on the animal under identical environmental conditions. By controlling the sensory milieu experienced by the specimens, differences in behavioral outputs can be observed. These differences may point to specific deficiencies in the nervous system model and serve to inform the iteration of the model for the particular behavior under study. This method allows for the experimental manipulation of electronic nervous systems and serves as a way to explore neuroscience hypotheses specifically regarding the neurophysiological basis of simple innate reflexive behaviors. The LEGO Mindstorms NXT kit provides an affordable and efficient platform on which to test preliminary biomimetic robot control schemes. The approach is also well suited for the high school classroom to serve as the foundation for a hands-on inquiry-based biorobotics curriculum.

An approach to neural network modeling on the LEGO Mindstorms robotics platform is presented. The method provides a simulation tool for invertebrate neuroscience research in both the research lab and the classroom. This technique enables the investigation of biomimetic robot control principles.

LEGO机器人神经系统仿真设计和实现

We present a method to use the commercially available LEGO Mindstorms NXT robotics platform to test systems level neuroscience hypotheses. The first step ...

Real-time Electrophysiology: Using Closed-loop Protocols to Probe Neuronal Dynamics and Beyond

Experimental neuroscience is witnessing an increased interest in the development and application of novel and often complex, closed-loop protocols, where the stimulus applied depends in real-time on the response of the system. Recent applications range from the implementation of virtual reality systems for studying motor responses both in mice1 and in zebrafish2, to control of seizures following cortical stroke using optogenetics3. A key advantage of closed-loop techniques resides in the capability of probing higher dimensional properties that are not directly accessible or that depend on multiple variables, such as neuronal excitability4 and reliability, while at the same time maximizing the experimental throughput. In this contribution and in the context of cellular electrophysiology, we describe how to apply a variety of closed-loop protocols to the study of the response properties of pyramidal cortical neurons, recorded intracellularly with the patch clamp technique in acute brain slices from the somatosensory cortex of juvenile rats. As no commercially available or open source software provides all the features required for efficiently performing the experiments described here, a new software toolbox called LCG5 was developed, whose modular structure maximizes reuse of computer code and facilitates the implementation of novel experimental paradigms. Stimulation waveforms are specified using a compact meta-description and full experimental protocols are described in text-based configuration files. Additionally, LCG has a command-line interface that is suited for repetition of trials and automation of experimental protocols.

Closed-loop protocols are becoming increasingly widespread in modern day electrophysiology. We present a simple, versatile and inexpensive way to perform complex electrophysiological protocols in cortical pyramidal neurons in vitro, using a desktop computer and a digital acquisition board.

实时电:采用闭环协议来探测神经动力学和超越

Experimental neuroscience is witnessing an increased interest in the development and application of novel and often complex, closed-loop protocols, where ...

Design, Surface Treatment, Cellular Plating, and Culturing of Modular Neuronal Networks Composed of Functionally Inter-connected Circuits

The brain operates through the coordinated activation and the dynamic communication of neuronal assemblies. A major open question is how a vast repertoire of dynamical motifs, which underlie most diverse brain functions, can emerge out of a fixed topological and modular organization of brain circuits. Compared to in vivo studies of neuronal circuits which present intrinsic experimental difficulties, in vitro preparations offer a much larger possibility to manipulate and probe the structural, dynamical and chemical properties of experimental neuronal systems. This work describes an in vitro experimental methodology which allows growing of modular networks composed by spatially distinct, functionally interconnected neuronal assemblies. The protocol allows controlling the two-dimensional (2D) architecture of the neuronal network at different levels of topological complexity.
A desired network patterning can be achieved both on regular cover slips and substrate embedded micro electrode arrays. Micromachined structures are embossed on a silicon wafer and used to create biocompatible polymeric stencils, which incorporate the negative features of the desired network architecture. The stencils are placed on the culturing substrates during the surface coating procedure with a molecular layer for promoting cellular adhesion. After removal of the stencils, neurons are plated and they spontaneously redirected to the coated areas. By decreasing the inter-compartment distance, it is possible to obtain either isolated or interconnected neuronal circuits. To promote cell survival, cells are co-cultured with a supporting neuronal network which is located at the periphery of the culture dish. Electrophysiological and optical recordings of the activity of modular networks obtained respectively by using substrate embedded micro electrode arrays and calcium imaging are presented. While each module shows spontaneous global synchronizations, the occurrence of inter-module synchronization is regulated by the density of connection among the circuits.

This manuscript describes a protocol to grow in vitro modular networks consisting of spatially confined, functionally inter-connected neuronal circuits. A polymeric mask is used to pattern a protein layer to promote cellular adhesion over the culturing substrate. Plated neurons grow on coated areas establishing spontaneous connections and exhibiting electrophysiological activity.

设计，表面处理，电镀细胞和神经元的模块化网络功能组成跨接电路的培养

The brain operates through the coordinated activation and the dynamic communication of neuronal assemblies. A major open question is how a vast repertoire ...

Time-dependent Increase in the Network Response to the Stimulation of Neuronal Cell Cultures on Micro-electrode Arrays

Micro-electrode arrays (MEAs) can be used to investigate drug toxicity, design paradigms for next-generation personalized medicine, and study network dynamics in neuronal cultures. In contrast with more traditional methods, such as patch-clamping, which can only record activity from a single cell, MEAs can record simultaneously from multiple sites in a network, without requiring the arduous task of placing each electrode individually. Moreover, numerous control and stimulation configurations can be easily applied within the same experimental setup, allowing for a broad range of dynamics to be explored. One of the key dynamics of interest in these in vitro studies has been the extent to which cultured networks display properties indicative of learning. Mouse neuronal cells cultured on MEAs display an increase in response following training induced by electrical stimulation. This protocol demonstrates how to culture neuronal cells on MEAs; successfully record from over 95% of the plated dishes; establish a protocol to train the networks to respond to patterns of stimulation; and sort, plot, and interpret the results from such experiments. The use of a proprietary system for stimulating and recording neuronal cultures is demonstrated. Software packages are also used to sort neuronal units. A custom-designed graphical user interface is used to visualize post-stimulus time histograms, inter-burst intervals, and burst duration, as well as to compare the cellular response to stimulation before and after a training protocol. Finally, representative results and future directions of this research effort are discussed.

Mouse neuronal cells cultured on multi-electrode arrays display an increase in response following electrical stimulation. This protocol demonstrates how to culture neurons, how to record activity, and how to establish a protocol to train the networks to respond to patterns of stimulation.

时间依赖性增加对微电极阵列刺激神经元细胞培养物的网络响应

Micro-electrode arrays (MEAs) can be used to investigate drug toxicity, design paradigms for next-generation personalized medicine, and study network ...

Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments

Closed-loop neurophysiological systems use patterns of neuronal activity to trigger stimuli, which in turn affect brain activity. Such closed-loop systems are already found in clinical applications, and are important tools for basic brain research. A particularly interesting recent development is the integration of closed-loop approaches with optogenetics, such that specific patterns of neuronal activity can trigger optical stimulation of selected neuronal groups. However, setting up an electrophysiological system for closed-loop experiments can be difficult. Here, a ready-to-apply Matlab code is provided for triggering stimuli based on the activity of single or multiple neurons. This sample code can be easily modified based on individual needs. For instance, it shows how to trigger sound stimuli and how to change it to trigger an external device connected to a PC serial port. The presented protocol is designed to work with a popular neuronal recording system for animal studies (Neuralynx). The implementation of closed-loop stimulation is demonstrated in an awake rat.

This protocol demonstrates how to use an electrophysiological system for closed-loop stimulation triggered by neuronal activity patterns. Sample Matlab code that can be easily modified for different stimulation devices is also provided.

在神经生理学实验中利用神经元刺激活性触发闭环刺激

Closed-loop neurophysiological systems use patterns of neuronal activity to trigger stimuli, which in turn affect brain activity. Such closed-loop systems ...

Automated Multimodal Stimulation and Simultaneous Neuronal Recording from Multiple Small Organisms

Fluorescent genetically encoded calcium indicators have contributed greatly to our understanding of neural dynamics from the level of individual neurons to entire brain circuits. However, neural responses may vary due to prior experience, internal states, or stochastic factors, thus generating the need for methods that can assess neural function across many individuals at once. Whereas most recording techniques examine a single animal at a time, we describe the use of wide-field microscopy to scale up neuronal recordings to dozens of Caenorhabditis elegans or other sub-millimeter-scale organisms at once. Open-source hardware and software allow great flexibility in programming fully automated experiments that control the intensity and timing of various stimulus types, including chemical, optical, mechanical, thermal, and electromagnetic stimuli. In particular, microfluidic flow devices provide precise, repeatable, and quantitative control of chemosensory stimuli with sub-second time resolution. The NeuroTracker semi-automated data analysis pipeline then extracts individual and population-wide neural responses to uncover functional changes in neural excitability and dynamics. This paper presents examples of measuring neuronal adaptation, temporal inhibition, and stimulus crosstalk. These techniques increase the precision and repeatability of stimulation, allow the exploration of population variability, and are generalizable to other dynamic fluorescent signals in small biosystems from cells and organoids to whole organisms and plants.

We present a method for the flexible chemical and multimodal stimulation and recording of simultaneous neural activity from many Caenorhabditis elegans worms. This method uses microfluidics, open-source hardware and software, and supervised automated data analysis to enable the measurement of neuronal phenomena such as adaptation, temporal inhibition, and stimulus crosstalk.

来自多个小生物体的自动多模态刺激和同时神经元记录

Fluorescent genetically encoded calcium indicators have contributed greatly to our understanding of neural dynamics from the level of individual neurons ...

HD-MEA 上的多模态大规模神经记录和分析 

Recording and Analyzing Multimodal Large-Scale Neuronal Ensemble Dynamics on CMOS-Integrated High-Density Microelectrode Array

Large-scale neuronal networks and their complex distributed microcircuits are essential to generate perception, cognition, and behavior that emerge from patterns of spatiotemporal neuronal activity. These dynamic patterns emerging from functional groups of interconnected neuronal ensembles facilitate precise computations for processing and coding multiscale neural information, thereby driving higher brain functions. To probe the computational principles of neural dynamics underlying this complexity and investigate the multiscale impact of biological processes in health and disease, large-scale simultaneous recordings have become instrumental. Here, a high-density microelectrode array (HD-MEA) is employed to study two modalities of neural dynamics - hippocampal and olfactory bulb circuits from ex-vivo mouse brain slices and neuronal networks from in-vitro cell cultures of human induced pluripotent stem cells (iPSCs). The HD-MEA platform, with 4096 microelectrodes, enables non-invasive, multi-site, label-free recordings of extracellular firing patterns from thousands of neuronal ensembles simultaneously at high spatiotemporal resolution. This approach allows the characterization of several electrophysiological network-wide features, including single/-multi-unit spiking activity patterns and local field potential oscillations. To scrutinize these multidimensional neural data, we have developed several computational tools incorporating machine learning algorithms, automatic event detection and classification, graph theory, and other advanced analyses. By supplementing these computational pipelines with this platform, we provide a methodology for studying the large, multiscale, and multimodal dynamics from cell assemblies to networks. This can potentially advance our understanding of complex brain functions and cognitive processes in health and disease. Commitment to open science and insights into large-scale computational neural dynamics could enhance brain-inspired modeling, neuromorphic computing, and neural learning algorithms. Furthermore, understanding the underlying mechanisms of impaired large-scale neural computations and their interconnected microcircuit dynamics could lead to the identification of specific biomarkers, paving the way for more accurate diagnostic tools and targeted therapies for neurological disorders.

作者聚焦：通过 HD-MEA 技术推进大规模神经动力学

Here, we employ HD-MEA to delve into computational dynamics of large-scale neuronal ensembles, particularly in hippocampal, olfactory bulb circuits, and human neuronal networks. Capturing spatiotemporal activity, combined with computational tools, provides insights into neuronal ensemble complexity. The method enhances understanding of brain functions, potentially identifying biomarkers and treatments for neurological disorders.

CMOS集成高密度微电极阵列上多模态大规模神经元集成动力学的记录与分析

Large-scale neuronal networks and their complex distributed microcircuits are essential to generate perception, cognition, and behavior that emerge from ...