标题含有负面词汇会让人们更可能浏览以提升点击量

较难

议论文

热点话题

一项新的研究发现，标题中含有负面词汇会让人们更可能浏览以提升点击量，而在头条新闻中的正面词汇会降低人们点击它们的可能性。

Millions of people now get their news online. But with so much online content to consume and so little time to consume it, editors at news outlets know that writing “catchy” headlines is more important than ever. So, what makes a headline catchy? A new study published in Nature Human Behaviour found that negative emotion words in headlines made it more likely that consumers would click through and read the story. In contrast, positive emotion words decreased click rates.

现在有数百万人通过互联网获取新闻。但是由于要消耗大量时间来阅读海量的网络内容，新闻媒体编辑知道撰写“引人注目”的标题比以往任何时候都更加重要。那么，什么样的标题才算是引人注目的呢？《Nature Human Behaviour》杂志发表的一项新研究发现，标题中使用的负面情感词汇可以提高消费者的点击率和阅读量。相反，积极情感词汇则会降低点击率。

The idea that people are more attentive to negative things is not a new one. Journalists have always had this intuition (captured in phrases like “if it bleeds, it leads”), and psychology studies have backed it up. People dislike losses more than they like gains, and they overweigh negative information in many contexts. Even infants pay more attention to negative stimuli. This negativity bias makes some sense from an evolutionary perspective: If you miss out on an opportunity to forage for some delicious berries, you will be sad…but if you accidentally eat a poison mushroom, you will be dead.

人们更关注负面事物并不是什么新鲜事。作为记者，他们一直有这种直觉（体现在“只要能见红，就能上头条”这样的短语中），心理学研究也支持了这一点。人们对遭受损失的厌恶程超过了对获得受益的喜爱，而且在许多情境下会过分强调负面信息。即使是婴儿也更注意负面刺激。从进化的角度来看，这种负面倾向是有道理的：如果你错过了采摘美味浆果的机会，你（仅仅）会感到难过，但如果你不小心吃了毒蘑菇，你就会死亡。

Given this negativity bias, do people tend to engage more with negative news content online? Previous research suggests that the answer is yes. For example, people share more content if it makes them angry. But a lot of this research has been done either in a controlled lab setting, or it has been correlational, so we can’t know for sure if negativity actually causes people to consume news more.

鉴于这种负面偏见，人们是否倾向于关注网上更多的负面新闻内容呢？先前的研究表明答案是肯定的。例如，如果人们感到愤怒，他们会分享更多的内容。但许多这项研究要么是在受控的实验室环境中进行的，或者是相关性研究，所以我们不能确定消极情绪是否真的会导致人们更多地消费新闻。

That’s why the dataset analyzed by Robertson and colleagues is so special. First of all, it is data on the number of clicks (“click-through rate” is the technical term) in response to more than 105,000 headlines on the news site Upworthy. Therefore, these are real decisions being made by real people. Moreover, Upworthy often tested out (through random assignment) different headlines for the same exact story, so they essentially ran experiments on their users. Therefore, even when the content of the news story was exactly the same, when editors put negative words like “hate” and “worst” into the headline, more people clicked on it, and when they included positive words like “love,” and “best,” fewer people clicked through.

这就是Robertson及其同事分析的数据集如此特殊的原因。首先，这是新闻网站Upworthy上10.5万多个头条新闻的点击次数（专业术语叫“点击率”）。因此，这些都是真人做出的真实反馈。而且，Upworthy经常（通过随机分配）为同一个确切的故事测试不同的标题，所以他们基本上在用户身上进行了实验。因此，即使新闻报道的内容完全相同，当编辑在标题中加入类似“仇恨”和“最坏”等负面词汇时，点击的人会更多，而当他们加入“爱”和“最好”等正面词汇时，浏览的人会更少。

Not only does this paper tell us more about the factors that influence online news consumption, but it also showcases the nuanced ways in which emotion influences decision-making. The researchers found that words about sadness increased click-through rates, but words about fear actually decreased them (and anger words had no significant effect). Now, we don’t know if people were actually feeling those emotions when they read the headlines, so we will need more studies to explore these effects further. However, given that different discrete emotions might be associated with different appraisals of one’s current situation, it makes sense that not all negative emotions would have the same downstream effects on behavior. For example, previous research has shown that people are more optimistic and risk-seeking when they are angry, but they are more pessimistic and risk-averse（风险厌恶）when they are afraid.

这篇论文不仅告诉了我们更多关于影响在线新闻消费的因素，还展示了情绪影响决策的微妙方式。研究人员发现，和悲伤相关的词会增加点击率，但和恐惧相关的词实际上会降低点击率（而愤怒的词没有显著影响）。现在，我们不知道人们在阅读头条新闻时是否真的感受到了这些情绪，所以我们需要更多的研究来进一步探索这些影响。然而，考虑到不同离散的情绪可能与当前情况的不同评估相关联，有些负面情绪对行为可能产生的影响不会完全相同，这是有道理的。例如，先前的研究表明，人们在愤怒时更乐观和寻求风险，但在害怕时更悲观和避免风险。