评论
打印
You might think of your eyes like cameras that just take in light from the world around you
你可能会把眼睛想象成相机，只需要从周围的世界中摄取光线，
to recreate a perfect snapshot in your brain.
就可以在大脑中重现一个完美的快照。
But in reality, your brain does a lot of processing before you even perceive an image.
但实际上，大脑在你感知到图像之前就做了很多处理。
And now, scientists have found a way to hack that process and generate shapes directly on the brain
现在，科学家们已经找到了一种方法来破解这个过程，直接在大脑上生成形状，
so that a person can see them without using their eyes.
这样不用眼睛就能看到它们。
The trick they used to do this reveals how vision is laid out in the brain,
他们所用的技巧揭示了视觉是如何在大脑中分布，
and could even offer a way for people who have lost their sight to process visual signals without their eyes.
甚至可以为失明者提供不用眼睛处理视觉信号的方法。
For those who can see, the process begins when light hits the eyes.
对于那些能看见的人来说，这个过程从光线照射到眼睛时开始。
There, light-sensitive cells send electrical signals down the optic nerve,
在那里，感光细胞通过视神经传递电信号，
where they eventually reach a region in the back of your brain called the visual cortex.
最终到达大脑后部叫做视觉皮层的区域。
At this point, you're still not consciously aware of what you've seen.
在这里，你仍然没有意识到所看到的东西。
Your visual cortex processes the raw visual information it gets from your eyes first, before you become conscious of it.
在你意识到之前，视觉皮层首先处理从眼睛获得的原始视觉信息。
And this region plays a pretty important role in how you see the world.
这个区域对于你如何看到周围的世界扮演着相当重要的角色。
Like, it can produce optical illusions even if your eyes take in signals perfectly.
就像，即使眼睛完美地接收到信号，它也会产生视觉错觉。
But even weirder, if you stimulate the visual cortex, it can make you see things even if your eyes aren't involved at all.
但更奇怪的是，如果刺激视觉皮层，它可以让你看到东西，即使眼睛根本没有参与。
And that's the idea that vision researchers find so intriguing.
这就是视觉研究人员发现的非常有趣的想法。
Scientists began testing this out in the 1950s,
科学家们在20世纪50年代开始对此进行测试，
but one of the most important studies was done by researchers at the University of London in the 1960s.
但其中一项最重要的研究是在20世纪60年代由伦敦大学的研究人员完成的。
They surgically implanted an array of electrodes onto the brains of two participants who had lost their sight.
他们通过外科手术将一组电极植入两名失明参与者的大脑。
The electrodes were specially designed to deliver small electrical currents to different regions of the visual cortex,
这种电极是专门设计用来向视觉皮层的不同区域输送小电流，
activating the brain cells there.
激活那里的脑细胞。
Each time a single electrode was switched on, the participants reported seeing a very small spot of white light.
每次打开一个电极，参与者都会看到一个非常小的白光点。
And when electrodes were activated in different parts of the cortex one at a time,
当大脑皮层不同部位的电极一次一个地被激活时，
that little spot of light showed up in different parts of their field of view.
那一小点光点就会出现在视野的不同部位。
Based on where the subjects saw the spot of light,
根据受试者看到光点的位置，
the scientists were able to connect certain regions of the brain with specific parts of the visual field.
科学家们能够将大脑的某些区域与视野的特定部分连接起来。
In fact, they found that a part of the visual cortex called V
事实上，他们发现视觉皮层的一个叫做V的部分，
is basically an exact projection of your visual field onto the physical structure of your brain.
基本上是视野投射到大脑物理结构上的精确投影。
And, incredibly, their results essentially confirmed maps of human vision that doctors had drawn up
而且，令人难以置信的是，他们的结果基本上证实了
based on soldiers' injuries back in World War I.
医生根据一战中士兵受伤情况绘制的人类视觉图谱。
Because the visual cortex is so neatly arranged,
由于视觉皮层的排列非常整齐，
researchers eventually became interested in seeing how precisely they could place images in someone's visual field just by stimulating the brain.
研究人员最终对仅通过刺激大脑就能将图像精确投射到某人的视野中产生了兴趣。
Scientists thought that if they could stimulate V to make those spots of light, now known as phosphenes, show up wherever they wanted,
科学家们认为，如果他们能刺激V部位，使这些称为压眼闪光的光点出现在他们想要的任何地方，
maybe they could combine them to make shapes, like letters of the alphabet.
也许他们可以将它们结合起来，形成各种形状，比如字母表中的字母。
Scientists hypothesized that it would work just like how individual pixels on a computer monitor come together to display text.
科学家们假设，其工作原理就像电脑显示器上的单个像素组合在一起显示文本一样。
Unfortunately, over the next few decades, no one found much support for that idea.
不幸的是，在接下来的几十年里，没人找到太多支持这种想法的证据。
Then, in a study published in May of 2020 in the journal Cell, researchers finally had a breakthrough.
随后，2020年5月发表在《细胞》杂志上的一项研究中，研究人员终于有了突破。
But at first, they ran into problems, too.
但一开始，他们也遇到了问题。
They found that stimulating multiple regions of V at the same time with an array of electrodes didn't produce an image of multiple phosphenes.
他们发现，用电极阵列同时刺激V部位的多个区域不会产生多个压眼闪光的图像。
Instead, participants typically reported seeing one bigger blob of light.
取而代之的是，参与者通常会看到更大的光点。
And that gave them a clue.
这给他们提供了线索。
They speculated that the currents being applied to each region weren't staying isolated,
他们推测，施加在每个区域的电流并不孤立，
and they were combining to activate more regions of V than they were meant to.
它们结合在一起激活了比预期更多的V部位。
So the team decided to try something different.
所以，研究团队决定尝试一些不同的东西。
Instead of trying to create a whole shape at once, they would use electrodes to trace the outline of shapes on the brain.
他们不会试图一次创造出一个完整的形状，而是用电极追踪大脑中形状的轮廓。
This would avoid the problem of having multiple electrodes active at once and prevent their signals from blending together.
这将避免同时激活多个电极的问题，并防止它们的信号混合在一起。
It took a little creativity, the researchers had to find a way to trace a continuous path on the brain,
这需要一点创造力，研究人员必须找到在大脑上追踪连续路径的方法，
and they only had 24 electrodes, so they couldn't connect the dots very smoothly.
而且他们只有24个电极，所以他们不能很顺利地连接这些点。
But by manipulating electric currents, they were able to guide the phosphene down a smooth path and trace out letters on the brain.
但是通过控制电流，他们能够引导压眼闪光沿着一条平滑的路径前进，并在大脑中找到字母。
And amazingly, it worked!
令人惊讶的是，这样做成功了！
Participants who had lost their sight not only recognized these letters,
失明的参与者不仅能认出这些字母，
they could even trace out similar versions of them with their fingers.
甚至能用手指找出它们的相似版本。
One participant was even able to recognize a sequence of letters at a rate of about 1 every 2 seconds with 92% accuracy.
一名参与者甚至能以每2秒1个字母的速度识别字母序列，准确率为92%。
This was incredibly exciting for the researchers, because ever since the 1960s, when scientists first started studying phosphenes,
研究人员对此感到非常兴奋，因为自20世纪60年代科学家们第一次开始研究压眼闪光以来，
their primary motivation was to create a kind of prosthetic for people who had lost function in their eyes.
他们的主要动机就是为眼睛功能丧失的人制造假体。
And having a way of tracing recognizable shapes directly onto the visual cortex opens up a lot of possibilities.
拥有一种直接在视觉皮层上追踪可识别形状的方法，就有了很多可能性。
For example, by having cameras study the environment, modern computer vision algorithms could draw cues on the visual cortex
例如，通过让摄像机研究环境，现代计算机视觉算法可以在视觉皮层上提取线索，
that people who have lost their sight could potentially use to navigate the world.
失明者可以利用这些线索来游览世界。
Even though that's a long way off, figuring out how to write on the brain is a huge first step.
尽管这还有很长的路要走，但弄清楚如何在大脑上写字是巨大的第一步。
