Miguel Nicolelis：僅通過思考就能實現的人機互動（圖靈訪談）

6月15日~6月18日，圖靈作為技術圖書出版先鋒品牌，受邀參加了第十五屆中國國際軟體和資訊服務交易會（中國軟交會）。本屆中國軟交會以“新IT、新生態、新動能”為主題，匯聚了來自全球的業界精英和代表，全面展示了IT產業的新技術、新思維和新模式，分享了智慧化在醫療、金融、服裝等行業引起的改變，共同探討了企業“智”造之路的務實方案。

作為受邀出版機構，圖靈為參會人員帶來了全新的“智慧閱讀”體驗，用科技和文化元素來打通IT人的圈層文化。“圖靈訪談”作為媒體嘉賓，會議期間對重量級嘉賓米格爾·尼科萊利斯和李傑進行了訪談。

訪談嘉賓：

Miguel Nicolelis（米格爾·尼科萊利斯），美國杜克大學醫學院神經生物學教授、法國科學院院士、巴西科學院院士、暢銷書《腦機穿越》作者、2014年巴西世界盃“機械戰甲”發明者。

他曾被美國科普雜誌《科學美國人》評為全球最具影響力的20位科學家之一，其研究被《麻省理工科技評論》評為10大最具突破性的科技創新之一。

訪談實錄：

決定投身“腦機介面”研究的原因是什麼？

米格爾： 一開始，我只是想研究人腦是怎麼工作的。這是主要的原因。隨著技術的發展，我發現我們能用腦機介面來幫助患有各種疾病的病人。這給了我更大的動力來繼續這方面的研究工作。

您做過很多非常棒的實驗，哪一個最特別？

米格爾： 在外骨骼的幫助下，病人可以再次行走。失去雙腿的病人在世界盃開幕式上可以再次踢足球。（這些對於我來說，都很特別。）我和他們每天都呆在一起，生活了8個月。站起來行走的那一刻對於他們每一個人來說，都是一個巨大的勝利，值得慶祝！

當Juliano在巴西世界盃開幕式上踢開一腳的時候，我就站在他身後的3米遠。我跑向他，一起慶祝。這一刻值得所有人流淚。

“腦機介面”研究的最終目標是什麼？

米格爾： 在醫療方面，目標是恢復病人失去的功能。其他方面的應用有遠端控制機器人，把機器人派到對於人類來說太小、太遠或是太危險的地方。如果能用人類大腦直接控制這些機器人，我們就能獲得更多的靈活性和更好的移動性。

另一件事是，多人通過“大腦網”進行遠端協作。

（圖靈出版的《資訊簡史》《漫談人工智慧》受到“中國軟交會”嘉賓的喜愛）

對下一代計算裝置的展望是怎樣的？對於人類來說，實現大腦控制計算程式的主要挑戰是什麼？

米格爾： 實際上，我認為未來將會看到大腦和計算機之間的連線，從而開發出新的計算裝置。這是一場全面的革命，因為我們的大腦將會跟這些裝置直接相連。我們不再需要打字，只通過思考就能和電腦互動。

最大的挑戰是能夠可靠地提取訊號，並且計算機能夠與大腦互動。這樣我們可以得到一個非常高效的介面用於日常工作，比如處理手機、膝上型電腦和汽車。一旦完善了這個互動，我們將看到腦機介面對整個技術的巨大影響。

我們是否還需要穿戴一些東西來讓大腦連線到數字應用？

米格爾： 是的，我們總是需要一些感測器來捕捉大腦的活動。沒有這些感測器，就無法傳輸大腦的電訊號。感測器要能夠儘可能多地從大腦中獲取資訊，完善與數字平臺互動的能力。對於動物來說，介面會植入大腦。但是對於人類來說，我們會研究一種類似帽子的東西。因為不會植入任何東西，所以對人類來說不會有危險。

這些介面是否有助於治療基因缺陷引起的疾病？

米格爾： 不，這並不針對基因缺失引起的疾病。你必須意識到世界上有近10億的人患有腦功能紊亂的疾病。如果我們把所有患有大腦功能紊亂的人放在一起，這幾乎就是一箇中國的人口。

大多數患有腦功能障礙的病人與基因方面並沒有關係。它們大部分來自脊髓損傷、帕金森和阿爾茨海默病症。我們需要對這將近10億的病人進行治療，我認為這項技術會有所幫助。

您對“不可能”的理解是什麼？

米格爾： 當我還是個孩子的時候，祖母告訴我，“不可能的事就是還沒有人花費足夠的時間去做的可能的事”。我十分相信這一點。我的祖母是一個很有智慧的女人。我認為成為一名科學家與她對我的影響有著很大的關係。她教會了我如何思考大事情。

隨著研究發展，你的新發現是什麼？

米格爾： 事實上，我正在寫一本新書，因為有很多新的想法出現。我在書中提到了一點，而且我認為現在很有必要講的一點是，有越來越多的證據表明我們的大腦遠比計算機優越。人工智慧不會征服人類的大腦。這就是我決定寫一本新書的原因。

---

What reasons are there making you dedicate yourself to brain-machine interfaces?

First, I want to investigate how the brain works. That's the main reason. As technology develops, I realize that we could use that to treat patients suffering from various diseases. It gives extra motivation to continue this approach.

You've done lots of fantastic experiments. Which one do you think is most special?

The patient with exoskeleton could walk again and the patient who has lost legs could kick again on the opening ceremony of world cup. I lived with these patients every day for 8 months. The moment they were up and walk is a big victory and celebration for each one of them. That's the most profound thing I have done, I think.

When Juliano kicked the ball on the ceremony, I was three meters behind him. After that kicking, I run to him for celebration. At that time, I almost cried because it's like a miracle.

What's the ultimate goal for brain-machine interfaces?

On the medical front, the goal is to restore functions to patients who have lost them. There will be other applications in controlling robots in places where it's too small, too far or too dangerous. If we can control robots directly with our brain, we could gain more flexibility and mobility.

The other thing is that multiple people could collaborate at distance through brain net.

What's your outlook for next generation computational devices? What's the main challenge for humans to realize brain control of computational Apps?

I actually think that the future will see the links between brains and computers to develop new computational devices. It's a total revolution in the way we talk about computers, because our brain is going to be directly linked to these devices. We don't need to type anymore, if we can basically interact with computers just by thinking.

The biggest challenge is to be able to extract signals reliably and that computers are able to interact with the brain. So we get a very efficient interface that can be used for routine tasks, like dealing with laptops, cars and cellphones. Once we perfect this link, we'll see tremendous impacts on technology.

Do we still need some kind of things to wear so as to connect brains to digital Apps?

Yes, we always need some sensors to capture brain activities. Without that, we couldn't transmit the electrical signals from our brains. We really need to develop the sensors to the point that we could get as much as possible from the brain to interact with digital platforms.

For animals, we implant. For human beings, what we're developing is something like a cap. You don't need to implant anything. That's the reason why it's not dangerous.

Would the interfaces be helpful for diseases resulted from genetic defects?

No, that's not for genetic diseases. You have to realize that there're almost one billion people in the world with brain disorder. If we put all the brain-disordered people together, it's almost one China of patients.

Most of the patients with brain disorder have nothing to do with genetic reasons. They come from lesions of spinal cord, Parkinson and Alzheimer's disease. So we need treatments for these almost one billion people. I think the technology will help.

What's your understanding of "impossibility"?

My grandmother told me when I was a kid that “the impossible is just the possible that nobody put time enough to make”. I believe in that. She was a very intellectual woman. I think to become a scientist has strong relationships with her influences on me. She taught me how to think big things.

As your research moves on, what's your new finding?

In fact, I'm finishing a new book, because there are so many new ideas coming out. One thing I've talked a little bit in the book and I think it's very important to say now is that there is more and more evident showing our brains are far superior than computers. AI will not overcome the human brain. That's the reason why I decide to write a new book.

---

更多精彩，加入圖靈訪談微信！