摘要：他们称之为“脑腐”。无意义的短视频刺激性刚好足以让你持续观看和滑动，以僵尸般的状态浏览算法推荐的下一个内容；又尚未乏味到让你将可货币化的眼球从屏幕上移开。观众们心情矛盾。这类内容提供了一种放空的方式，同时也提供了一种浪费生命数小时的方式。

The Economist ｜Free exchange

经济学人｜自由交换

Can you make it to the end of this column?

你能坚持读完这篇专栏吗？

Understanding the new economics of attention

理解注意力新经济学

他们称之为“脑腐”。无意义的短视频刺激性刚好足以让你持续观看和滑动，以僵尸般的状态浏览算法推荐的下一个内容；又尚未乏味到让你将可货币化的眼球从屏幕上移开。观众们心情矛盾。这类内容提供了一种放空的方式，同时也提供了一种浪费生命数小时的方式。

经济学家可能很快会将“脑腐”视为一种盗窃手段。该学科日益将注意力建模为一种资源，与土地、劳动力和资本并列。用该领域的行话说， 注意力是稀缺且具有竞争性的 ；花在“脑腐”上的时间无法用于其他事情。专注对大多数工作形式至关重要，它既能辅助生产，也能在休闲中被消耗。要充分获取收益，比如阅读报纸专栏，需要你全神贯注——如果手机在旁边，这很难做到。

将注意力视为稀缺资源，有助于弥合传统“经济人”模型（一个寻求最大化偏好满足的理性优化者）与行为经济学实验中出现的形象（被各种偏见拉离理性思考的人）之间的部分差距。例如，为何消费者始终陷入“左位数偏见”，即他们购买标价2.99美元商品的可能性远高于3.00美元，却无法区分标价2.99美元和2.98美元的商品？

因为注意力是稀缺的，这意味着人们依赖捷径，而2就是比3少。何必费心去看价格的其余部分？2003年，普林斯顿大学的克里斯托弗·西姆斯（后获诺贝尔经济学奖）提出了他所谓的“理性疏忽”模型，其中优化主体一次只能处理这么多信息。他认为，这种模型解释了包括利率和价格在内的各种宏观经济变量对新信息的平滑（而非即时）调整。人们明智地只将有限注意力的一部分用于了解影响市场的新闻。

这种方法有其先驱。赫伯特·西蒙（1975年获图灵奖，1978年获诺贝尔经济学奖）创造了“有限理性”一词。他认为，经济学不应将人建模为全知优化者，而应建模为“满足者”，即在信息环境限制下选择足够好（而非完美）的方案。“信息消耗的东西相当明显：它消耗的是接收者的注意力，”他写道。“ 信息丰富导致注意力贫乏，并需要高效分配这种注意力 。”通过智能手机传播的无尽信息创造了这样一个世界。正如西蒙指出的，无限的欲望和稀缺的资源正是经济问题的定义。

然而，将注意力视为可与劳动力和资本类似分配的资源，也存在一些担忧。 什么算作注意力？ 卡内基梅隆大学的乔治·洛温斯坦和麻省理工学院的扎卡里·沃伊托维奇近期的一篇评论将其定义为： 将一种稀缺、具有竞争性的心智资源选择性分配给信息处理任务。 注意力资源包括工作记忆、选择性专注和视觉焦点。正如洛温斯坦和沃伊托维奇指出的，“注意”既可以表示“看这个”，也可以表示“停止做白日梦”，而遵循任一指令都意味着相似但细微不同的行动。将它们全部归入“注意力”这一单一类别会带来自身问题，正如经济学将各种生产资料归入“资本”这一大类时发现的那样。

此外，尽管注意力可以被引导至不同方向，但它常常是非自愿的。因此， 关于注意力如何分配的经济模型大致分为两类：自上而下和自下而上 。自上而下模型是更典型的经济学范式，其中主体选择稀缺资源的最佳用途。一旦学会开车，驾驶就很枯燥，何不同时打个电话？自下而上模型指出，环境本身会引导人的注意力——例如手机收到新消息时的提示音，或一个孩子跑到路上。洛温斯坦和沃伊托维奇更进一步，提出情绪状态本身可以引导注意力：疼痛难以忽视，无聊令人不快，饥饿让人很难专注于食物以外的任何事情。这种理解可能将经济学家带入新领域。通常他们回避过多关注“经济人”的内心世界。

抵抗“脑腐”很难。西蒙用剪刀的双刃来比喻他的有限理性概念。决策来自个体自身局限与其所处信息环境的结合。问剪刀的哪一片刃完成了裁剪是错误的，因为总是两者共同作用。仅靠意志力不太可能击败那些经过完美调校、通过算法不断调整以最大化用户参与度的干扰机器。环境是充满敌意的。

这应该引起经济学家的关注，因为如果注意力是一种资源，它并非一种受产权保护的资源。巧妙的设计或许能将其从我们这里夺走，并且毫无补偿的希望。智能手机和社交媒体的普及创造了一个极易发生盗窃的世界。有什么能让注意力回归其所有者吗？社会规范或许有帮助：例如，图书馆长期以来坚持保持安静以保护专注力。假以时日，监管或许也能起作用。 但目前，最大的希望是将问题掌握在自己手中，当你想阅读专栏时，请将手机放得远远的。 ■

They callit brainrot. Inane short-form videos just stimulating enough to keep you watching and scrolling, in a zombie-like manner, through whatever the algorithm presents next; not quite dull enough for you to tear your monetisable eyeballs away from the screen. Viewers are ambivalent. Such content offers a way to switch off. It also offers a way to waste hours of your life.

Economists may soon start to think of brainrot as a means of theft. Increasingly, the discipline is modelling attention as a resource, alongside land, labour and capital. Attention is scarce and rivalrous in the field’s jargon; time spent on brainrot cannot be spent on something else. Focus, being vital to most forms of work, aids production and can be consumed in leisure. Getting the most out of, say, reading a newspaper column requires your full attention, which can be hard to provide if your phone is nearby.

Treating attention as a scarce resource helps bridge some of the gap between traditional models of Homo economicus—a rational optimiser seeking to maximise satisfaction of their preferences—and the figure who emerges from the experiments of behavioural economics, who is pulled away from rational thinking by any number of biases. Why do consumers consistently fall for “left-digit bias”, for instance, meaning they are far more likely to buy something for $2.99 than $3.00 while not distinguishing between items that cost $2.99 and $2.98?

Well, attention is scarce, which means that people rely on shortcuts and $2 is less than $3. Why bother to read the rest of the price? In 2003 Christopher Sims of Princeton University, later awarded a Nobel prize in economics, developed what he called a “rational inattention” model, in which optimising agents can process only so much information at a time. Such a model explains the smooth (rather than instant) adjustment of various macroeconomic variables, including interest rates and prices, to new information, he suggested. People sensibly devote just a portion of their limited attention to learning about market-moving news.

There are antecedents to this approach. Herbert Simon, who won both the Turing award for artificial intelligence in 1975 and the Nobel prize in economics in 1978, coined the term “bounded rationality”. Instead of an all-knowing optimiser, economics should model people as “satisficing”, he argued, choosing options that are simply good enough, rather than perfect, given the limits of their informational environment. “What information consumes is rather obvious: it consumes the attention of its recipients,” he wrote. “A wealth of information creates a poverty of attention and the need to allocate that attention efficiently.” Endless information distributed through smartphones creates such a world. And as Simon pointed out, unlimited wants and scarce resources are the definition of an economic problem.

Yet there are some concerns about treating attention as a resource that can be allocated along the lines of labour and capital. What counts as attention? A recent review by George Loewenstein of Carnegie Mellon University and Zachary Wojtowicz of the Massachusetts Institute of Technology defines it as the selective allocation of a scarce, rivalrous mental resource to an information-processing task. Attentional resources include working memory, selective concentration and visual focus. As Messrs Loewenstein and Wojtowicz point out, “pay attention” could either mean “look at this” or “stop daydreaming”, and following either command would mean similar, but subtly different, actions. Aggregating all of them into a single category, “attention”, comes with its own problems, as economics has discovered when putting diverse means of production into the single bucket of “capital”.

Moreover, although attention can be steered in one direction or another, it is often involuntary. Economic models of how it is allocated therefore divide roughly into two categories: top-down and bottom-up. The top-down model is a more typical economic paradigm, in which the agent chooses the best use of a scarce resource. Once you have learned how to do it, driving is dull, so why not make a phone call at the same time? The bottom-up model points out that the environment itself directs a person’s attention—the ping from the phone as a new message arrives, for instance, or a child running out into the road. Messrs Loewenstein and Wojtowicz take this a little further, suggesting that emotional states can themselves direct attention: pain is hard to ignore, boredom is unpleasant and hunger makes it difficult to focus on anything but food. Such an understanding could draw economists into new territory. Normally they shy away from caring too much about the inner life of Homo economicus.

Information warfare

Resisting brainrot is hard. Simon used the two blades of a pair of scissors to illustrate his concept of bounded rationality. Decisions come from a combination of the individual’s own limitations and the informational environment in which they operate. Asking which blade of the scissors did the cutting is a mistake, since it is always both. Willpower alone is unlikely to defeat perfectly tuned distraction machines with algorithms that constantly adjust to maximise user engagement. The environment is hostile.

That should concern economists, for if attention is a resource, it is not one covered by property rights. Cunning design might be able to take it from us, and do so with no hope of restitution. The spread of smartphones and social media has created a world in which it is very easy to steal. Can anything return attention to its owners? Social norms might help: libraries, for instance, have long insisted on silence in order to preserve focus. In time, regulation might, too. For the moment, however, the best hope is to take matters into your own hands and, when you are trying to read a column, put your phone far, far away. ■

来源：左右图史