译自《经济学人》2018.1.4文章 Reading the brain from the outside--Can brain activity be deciphered withoutopening up the skull?
PATRICK KAIFOSH’S left hand lies flat on thetable in front of him. Occasionally his fingers twitch or his palm rises upslightly from the surface. There is nothing obvious to connect these movementswith what is happening on the tablet in front of him, where a game of asteroidsis being played. Yet he is controlling the spaceship on the screen as it spins,thrusts and fires.
Patrick Kaifosh的左手平放在他身前的桌上，他的手指偶尔抽动下，手掌偶尔轻轻从桌面抬起。你很难把这些动作和他面前显示屏上的行星游戏联系起来，然而实际上他正通过这些动作控制着屏幕上飞船的旋转、推进和开火。
What enables him to do so is a sweatbandstudded with small gold bars that sits halfway up his left forearm. Each barcontains a handful of electrodes designed to pick up the signals of motor units(the combination of a motor neuron, a cell that projects from the spinal cord,and the muscle fibres it controls). These data are processed bymachine-learning algorithms and translated into the actions in the game. DrKaifosh, a co-founder of CTRL-Labs, the startup behind the device, has learnedto exercise impressive control over these signals with hardly any obviousmovement.
之所以能这样做是因为在他左前臂中部有一块特殊的防汗带，上面镶有很小的黄金条，每个黄金条上都有一些用来捕捉人体运动单元信号的电子元件（运动单元是运动神经的组合：一个从脊髓伸出的神经细胞和由它控制的肌纤维），这些数据再通过机器学习算法处理转换成游戏里面的动作。CTRL-Labs的创始人之一Kaifosh博士，同样也是这个设备的创始者，已经学会用难以察觉的细微动作来熟练控制这些信号了。
Some say that the claims of Dr Kaifosh andThomas Reardon, his co-founder, that CTRL-Labs has created a brain-machineinterface are nonsense. The sweatband is nowhere near the brain, and thesignals it is picking up are generated not just by the firing of a motor neuronbut by the electrical activity of muscles. “If this is a BCI, then the movementof my fingers when I type on a keyboard is also a brain output,” sniffs oneresearcher.Krishna Shenoy, who directs the neural prostheticssystems lab at Stanford University and acts as an adviser to the firm, thinksit is on the right side of the divide. “Measuring the movement of the hand ismotion capture. They are picking up neural activity amplified by the muscles.”
一些人认为Kaifosh博士和他的合伙人Thomas Reardon宣称他们的CTRL实验室已经研发出脑机接口技术（BCI）的言论是荒谬的。他们的绑带并没有靠近大脑，收集的信号不光是运动神经的活动，还有肌肉的电反应。“如果这是脑机接口技术，那我敲击键盘时手指的活动也可以认为是脑活动。”一位研究人员嘲弄着说。斯坦福大学神经修复系统实验室的主导人同时也是CTRL-LAB的咨询师Krishna Shenoy则认为这个设备是个分水岭，“衡量手的运动是一种动作捕捉技术，它收集那些被肌肉放大的神经活动。”
It is not reasonableto expect consumers to undergo brain surgery
Whatever the semantics, it isinstructive to hear the logic behind the firm’s decision to record the activityof the peripheral nervous system, rather than looking directly inside the head.The startup wants to create a consumer product (its potential uses includebeing an interface for interactions in virtual reality and augmented reality).It is not reasonable to expect consumers to undergo brain surgery, say thefounders, and current non-invasive options for reading the brain provide noisy,hard-to-read signals. “For machine-learning folk, there is no question whichdata set—cortical neurons or motor neurons—you would prefer,” says Dr Reardon.
“期待消费者忍受脑外科手术太不合理。”
无论这句话意义如何，这个公司所做决定背后的逻辑都具有指导意义：记录外围神经系统的活动，而非直接进入大脑内部。发起者希望创造一个消费者直接使用的产品（它的潜在应用包括用于虚拟现实和增强现实的交互界面）。让消费者忍受大脑手术太不合理，创办人说，而现有非侵入式解读大脑方式提供的信号又太嘈杂、不易解读。“对于熟悉机器学习的人来说，更倾向于使用哪种数据集显而易见——皮层神经元数据或运动神经元数据。”Reardon博士说。
This trade-offbetween the degree of invasiveness and the fidelity of brain signals is a bigproblem in the search for improved BCIs. But plenty of people are trying tofind a better way to read neural code from outside the skull.
在侵入度和信号保真度之间做权衡是BCI（脑机接口技术）研发中的主要问题，但是更多人尝试寻找从头骨外部解读大脑信号的更好方法。
The simplest wayto read electrical activity from outside is to conduct an electroencephalogram(EEG). And it is not all that simple. Conventionally, it has involved wearing acap containing lots of electrodes that are pressed against the surface of the scalp.To improve the signal quality, a conductive gel is often applied. That requiresa hairwash afterwards. Sometimes the skin of the scalp is roughened up to get abetter connection. As a consumer experience it beats going to the dentist, butnot by much.
从外部读取脑电活动最简单的方法就是进行脑电波扫描（EEG）。然而这并不简单，依照惯例，需要带上包含很多电极的帽子，这些电极压在头皮表面。为了增强信号质量，还需要涂抹导电膏，因此用后还需洗发。有时还要把头皮弄得更粗糙些以便更好连接。这种消费体验比看牙医好不到哪去。
Once on, eachelectrode picks up currents generated by the firing of thousands of neurons,but only in the area covered by that electrode. Neurons that fire deep in thebrain are not detected either. The signal is distorted by the layers of skin,bone and membrane that separate the brain from the electrode. And muscleactivity (of the sort that CTRL-Labs looks for) from eye and neck movements orclenched jaws can overwhelm the neural data.
一旦戴上这种帽子，每个电极将收集上千个神经元释放出的电流，但只限于被电极覆盖的区域。且大脑深层神经元释放的信号不会被探测到，信号被大脑和电极间层层的皮肤、骨头、筋膜干扰扭曲。眼部颈部的肌肉活动（CTRL实验室寻找的那种）或者口部的咬合也会淹没神经数据。
Even so, someEEG signals are strong enough to be picked up pretty reliably. An“event-related potential”, for example, is an electrical signal that the brainreliably gives off in response to an external stimulus of some sort. One such,called an error-related potential (Errp), occurs when a user spots a mistake. Researchersat MIT have connected a human observer wearing an EEG cap to an industrialrobot called Baxter as it carried out a sorting task. If Baxter made a mistake,an Errp signal in the observer’s brain alerted the robot to its error;helpfully, if Baxter still did not react, the human brain generated an evenstronger Errp signal.
尽管如此，一些信号还是强到可以被可靠地收集。例如“事件相关电信号”，它是大脑对特定外部刺激发出的一种可靠电信号。其中一种叫“错误相关电信号”（ERRP），当用户犯错时会产生。麻省理工学院的研究员将一台名为Baxter的正在做分类工作的工业机器人连接到一名戴着脑部扫描头盔（EEG）的人类观察者身上。Baxter一旦犯错，观察者大脑发出的“错误相关电信号（ERRP）”将警告机器人犯错；有帮助的是，如果Baxter没有任何反应，人类大脑将发出更强的错误相关电信号（ERRP）。
If the cap fits
Neurable, a consumer startup, hasdeveloped an EEG headset with just seven dry electrodes which uses a signalcalled the P300 to enable users to play a virtual-reality (VR) escape game.This signal is a marker of surprise or recognition. Think of the word “brain” andthen watch a series of letters flash up randomly on a screen; when the letter“b” comes up, you will almost certainly be giving off a P300 signal. InNeurable’s game, all you have to do is concentrate on an object (a ball, say)for it to come towards you or be hurled at an object.Ramses Alcaide,Neurable’s boss, sees the potential for entertainment companies like Disney(owner of the Star Wars and Marvel franchises) to license the software in themeparks and arcade games.
如果帽子合适
Neurable是一个由用户发起的公司，他们开发了一种头戴式脑部扫描设备（EEG），它包含7个使用P300信号的电极，依靠它用户可以玩逃亡虚拟现实游戏。这种信号可以反映出大脑的惊奇反应或识别反应。想着一个词“brain”，然后看到一串字母在屏幕上随机显现，当字母‘b’出现时，你的大脑就会发出一个P300信号。在Neurable的游戏中，你需要做的就是集中注意力于某件物体上（比如一个球），然后它就会到自动你手中或是被掷向其他物体。Neurable的老板Ramses Alcaide认为一些娱乐公司，比如迪士尼，可能在它们的主题公园或游乐场应用该软件。
Thorsten Zander of theTechnische Universität in Berlin thinks that “passive” EEG signals (those thatare not evoked by an external stimulus) can be put to good use too. Researchhas shown that brainwave activity changes depending on how alert, drowsy or focuseda person is. If an EEG can reliably pick this up, perhaps surgeons, pilots ortruck drivers who are becoming dangerously tired can be identified. Studieshave shown strong correlations between people’s mental states as shown by anEEG and their ability to spot weapons in X-rays of luggage.
柏林科技大学的Thorsten Zander认为一些被动脑电波信号（那些不是被外界刺激引发的信号）也能被很好地利用。研究显示脑波活动的变化取决于人大脑的警觉度、昏沉度或注意力集中程度。如果脑电波扫描技术（EEG）能可靠地收集这些信号，那些处于危险疲惫状态的外科医生、飞行员、卡车司机就能被识别出来了。研究表明，EEG技术反映出的人脑状态跟人识别安检机中武器的能力有很强的相关性。
Yet the uses ofEEGs remain limited. In a real-world environment like a cockpit, a car or anairport, muscle activity and ambient electricity are likely to confound anyneural signals. As for Neurable’s game, it relies not solely on brain activitybut also deploys eye-tracking technology to see where a player is looking. DrAlcaide says the system can work with brain signals alone, but it is hard for auser to disentangle the two.
然而EEG技术的应用还是很有限的，在类似驾驶舱、汽车或机场的真实环境中，肌肉活动和外界的电流很容易和神经信号相混淆。至于Neurable的游戏，它不单单依赖大脑活动，还需应用人眼追踪技术追踪玩家视线。Alcaide博士说系统可以只需要大脑信号，但对玩家来说要把视觉信号和脑信号相分离很难。
Othernon-invasive options also have flaws. Magnetoencephalography measures magneticfields generated by electrical activity in the brain, but it requires a specialroom to shield the machinery from Earth’s magnetic field. Functional magneticresonance imaging (fMRI) can spot changes in blood oxygenation, a proxy forneural activity, and can zero in on a small area of the brain. But it involvesa large, expensive machine, and there is a lag between neural activity andblood flow.
其他非侵入式方法也有瑕疵。
脑磁图描技术可以测量脑电活动产生的磁场，但它需要一个特殊空间来隔离地球磁场。功能性磁场共振成像技术可以探测到血氧变化，它可作为神经活动的替代指征，也能精确测量大脑中的一小块区域。但这需要大型而昂贵的机器，且血氧变化相对神经活动有滞后性。
If any area islikely to yield a big breakthrough in non-invasive recording of the brain, itis a variation on fNIRS, the infrared technique used in the experiment to allowlocked-in patients to communicate. In essence, light sent through the skull iseither absorbed or reflected back to detectors, providing a picture of what isgoing on in the brain. This technique does not require bulky equipment, andunlike EEG it does not measure electrical activity, so it is not confused bymuscle activity. Both Facebook and Openwater are focusing their efforts on thisarea.
在非侵入式记录大脑活动的方法中可能有突破的是改进的FNIRS技术，实验中用这种红外技术可以让锁住的病人正常交流。大体原理是，透过头骨的光线被吸收或是被反射回探测器形成大脑活动图片。这项技术不需要庞大的机器，它不像EEG技术测量大脑电活动，所以也不会受肌肉活动影响。Facebook和Openwater公司都着重在该领域进行研究。
The obstacles toa breakthrough are formidable, however. Current infrared techniques measure anepiphenomenon,blood oxygenation(the degree ofwhich affects the absorption of light), rather than the actual firing ofneurons. The light usually penetrates only a few millimetres into the cortex.And because light scatters in tissue (think of how your whole fingertip glowsred when you press a pen-torch against it), the precise source of reflectedsignals is hard to identify.
然而距离突破性的进展还有很多的困难，目前的红外技术可以测量附带现象和血氧含量（反映了红外光的吸收量），而不是真正的神经发射。红外光只能穿透头皮几毫米，由于光在组织中会发生散射（想想当你按压红外笔时整个指尖都红了），反射信号也很难被精确识别。
Facebook is notsaying much about what it is doing. Its efforts are being led by MarkChevillet, who joined the social-media giant’s Building 8 consumer-hardwareteam fromJohns HopkinsUniversity. To cope with the problem oflight scattering as it passes through the brain, the team hopes to be able topick up on bothballistic photons, which passthrough tissue in a straight line, and what it terms “quasi-ballistic photons”,which deviate slightly but can still be traced to a specific source. The clockis ticking. Dr Chevillet has about a year of a two-year programme left todemonstrate that the firm’s goal of brain-controlled typing at 100 words aminute is achievable using current invasive cell-recording techniques, and toproduce a road map for replicating that level of performance non-invasively.
Facebook没有透漏太多进展。他们的研究由加入了约翰霍普金斯大学社会传媒巨头 “8幢用户-硬件团队”的Mark Chevillet领导。为了处理光透过大脑后的散射问题，该团队希望能同时收集在组织中沿直线传播的弹道光子，该光子称为“准弹道光子”，虽然有略微偏差但仍可被准确追溯至源头。时间很紧迫，Chevillet博士还剩1年时间去实现公司一个2年项目的目标，该目标是利用现有非侵入式细胞记录技术达到大脑控制打字每分钟100字，并绘制出那个水平的脑图谱。
Openwater ismuch less tight-lipped. Ms Jepsen says that her San Francisco-based startupuses holography to reconstruct how light scatters in the body, so it canneutralise this effect. Openwater, she suggests, has already created technologythat has a billion times the resolution of an fMRI machine, can penetrate thecortex to a depth of 10cm, and can sample data in milliseconds.
Openwater公司则没有那么守口如瓶了，Jepsen女士说他们在弗朗西斯科的项目使用全息摄影术修正光线在体内的散射，这可以中和散射影响。她也透露Openwater公司已开发出效果比FMRI好十亿倍的技术，可以渗透到皮层下方深达10厘米，也可以以毫秒的速度取样。
Openwater hasyet to demonstrate its technology, so these claims are impossible to verify.Most BCI experts are sceptical. But Ms Jepsen has an impressive background inconsumer electronics and display technologies, and breakthroughs by theirnature upend conventional wisdom. Developer kits are due out in 2018.
Openwater公司还需进一步阐明他们的技术，所以这些言论还无法证实，大多数脑机接口技术专家对此都持怀疑态度。但Jepsen女士在消费电子显示技术领域有很很强的背景，而突破性进展经常会颠覆人们的传统观念。开发工具将于2018年待售。
In the meantime,other efforts to decipher the language of the brain are under way. Some involveheading downstream intothe peripheral nervous system.One example of that approach is CTRL-Labs; another is provided by Qi Wang, atColumbia University, who researches the role of thelocus coeruleus,a nucleus deep in the brain stem that plays a role in modulating anxiety andstress. Dr Wang is looking at ways of stimulating the vagus nerve, which runsfrom the brain into the abdomen, through the skin to see if he can affectthelocus coeruleus.
同时，其他破译大脑语言的努力也在持续进行中，其中一些最终跟外围神经系统技术合流。其中一个例子是CTRL实验室，另外一个来自哥伦比亚大学的王奇，他研究了位于脑干深处的原子核“蓝斑核”调节焦虑和压力的作用。王先生致力于透过皮肤刺激从脑部通向腹部的迷走神经，观察这些刺激能否作用于蓝斑核。
Others arelooking at invasive approaches that do not involve drilling through the skull.One idea, from a firm called SmartStent, using technology partly developed withthe University of Melbourne, is to use a stent-like device called a “stentrode”that is studded with electrodes. It is inserted via a small incision in theneck and then guided up through blood vessels to overlie the brain. Once thedevice is in the right location, it expands from about the size of a matchstickto the size of the vessel and tissuegrows into its scaffolding,keeping it in place. Human trials of the stentrode are due to start next year.
另些人尝试侵入式的方法，但不用钻通颅骨。其中一个来自于SmartStent公司的想法，部分利用墨尔本大学研发的技术，使用一个镀有电极的支架状设备stentrode，通过颈部小切口植入体内，再通过血管流向大脑，一旦到达准确位置，它就会从火柴杆大小展开到血管大小，周围组织会把它包裹固定住。Stentrode技术的人体试验将在明年进行。
Another approachis to put electrodes under the scalp but not under the skull. Maxime Baud, aneurologist attached to the Wyss Centre, wants to do just that in order tomonitor the long-term seizure patterns of epileptics. He hopes that once thesepatterns are revealed, they can be used to provide accurate forecasts of when aseizure is likely to occur.
另一种方法是把电极放在头皮下面而不是头骨下面。隶属于怀斯中心的神经学家Maxime Baud想通过这种方法跟踪捕获癫痫病的长期规律，他希望发现这些规律，并用来精确预测癫痫病的发病时间。
Yet others thinkthey need to go directly to the source of action potentials. And that meansheading inside the brain itself.
然而一些人认为他们应该直接进入到产生行动电势的源头，那意味着要进入大脑内部。