Whether its Chinese social media like Sina Weibo， or Western media like Instagram and Facebook， videos can go viral in mere hours. The wide range of viral videos suggests that popular concepts are largely random.

According to scientists from Stanford University， US， the virality of a video can be predicted by looking at how certain areas of a persons brain are activated within the first few seconds of a video. This method has been called neuroforecasting.

The team made the finding by recruiting 36 volunteers to watch a range of videos while being scanned with an fMRI—a machine that can detect the changes in blood oxygen and flow.

Participants were shown 32 different videos and were monitored according to their brain responses in four different areas. The results showed that specific brain activity during the first four seconds of a video could effectively predict a persons thoughts on a video and whether they would keep watching.

Using the fMRI results， the Stanford team consistently saw increased activity in the nucleus accumbens and decreased activity in the anterior insula parts of the brain while the participants were watching the most popular of the 32-video selection.

According to the study， these two brain regions are related to the feeling of anticipation we get when were not certain of somethings outcome.

Stanford neuroscientist， and the studys author， Brian Knutson， said， “If we examine our subjects choices to watch the video or even their reported responses to the videos， they dont tell us about the general response online… Only brain activity seems to forecast a videos popularity on the Internet.”

He added， “Future research might also systematically deconstruct video content so content creators can use these findings to make their videos more popular.”

不管是在新浪微博等中國社交媒體，還是在照片墻、臉書等西方社交媒體上，視頻都能在短短幾小時內(nèi)迅速火爆。包羅萬象的爆款視頻表明，流行話題大多數(shù)都是隨機出現(xiàn)的。

美國斯坦福大學(xué)的科學(xué)家們認(rèn)為，一段視頻的火爆程度能夠通過觀察這段視頻開頭幾秒如何激活人們大腦特定區(qū)域來做出預(yù)測。這一方法被稱為神經(jīng)預(yù)測。

科研團隊招募了36名志愿者，讓他們一邊看一系列視頻，一邊接受功能性磁共振成像監(jiān)測——該設(shè)備能夠發(fā)現(xiàn)血氧和血流的變化，從而得出結(jié)論。

實驗讓參與者們觀看了32段不同的視頻，并對大腦四個不同區(qū)域的反應(yīng)進行了監(jiān)測。結(jié)果顯示，視頻開頭前四秒引起的特定大腦活動能夠有效預(yù)測人們對視頻的看法，以及他們是否會繼續(xù)看下去。

當(dāng)參與者們觀看32段視頻中最受歡迎的視頻時，斯坦福團隊根據(jù)功能性磁共振成像的結(jié)果，一致發(fā)現(xiàn)大腦伏核部分的活動增加，前腦島部分的活動減少。

研究表明，大腦這兩塊區(qū)域與我們面對未知結(jié)果的預(yù)期感覺有關(guān)。

斯坦福大學(xué)神經(jīng)學(xué)家、該研究的負責(zé)人布萊恩·克努森表示：“如果我們觀察參與者們選擇觀看的視頻主題，或者甚至是實驗報告中對于視頻的反應(yīng)，都無法體現(xiàn)網(wǎng)上的大致反響……只有大腦活動似乎能夠預(yù)測視頻在網(wǎng)上的流行程度?！?/p>

他還補充道：“未來的研究或許還會系統(tǒng)地拆析視頻內(nèi)容，這樣內(nèi)容創(chuàng)作者便能利用這些發(fā)現(xiàn)，來制作更受歡迎的視頻?！?/p>

Word Study

go viral 像病毒一樣傳播開來

In the new world of social media， board actions instantly go viral.

activate /'?kt?ve?t/ v. 使活動;激活

The burglar alarm is activated by movement.

recruit /r?'kru?t/ v. 吸收;征募

They recruited several new members to the club.

detect /d?'tekt/ v. 發(fā)現(xiàn);查明

The tests are designed to detect the disease early.

selection /s?'lek?n/ n. 被挑選的人（或物）;挑選

A selection of readers comments are published below.