夜航船

Hindsight Bias 后见之明

结果一旦揭晓,人就容易重新解释当时的线索,高估自己原本的预测能力,觉得我早就知道会这样。这是后见之明。

The Social Animal 里 Aronson 的原文:

it's also likely that you are acting on the hindsight bias, which refers to our tendency to overestimate our powers of prediction once we know the outcome of a given event. (Location 254) Numerous studies have demonstrated this bias, from predicting the outcomes of elections ("I always knew he'd win, even if the polls didn't say so") to predicting how a given social psychology experiment will turn out ("Oh, please, it was obvious the participants would obey those cruel instructions"). (Location 256)

人倾向于把零散线索组织成故事,这也可以联系到 故事的力量。它和 火鸡问题 有一个对照:火鸡对未来过度自信,后见之明让人高估自己对过去的理解。两者都会让世界显得更可预测,也可能让下一次冒险显得更安全。

卡尼曼写决策者受到事后审判时说:

When the outcomes are bad, the clients often blame their agents for not seeing the handwriting on the wall—forgetting that it was written in invisible ink that became legible only afterward.

——Daniel Kahneman,Thinking, Fast and Slow

结果不好,不表示当时的决定一定愚蠢;结果很好,也不表示过程合理。小数定律让小样本显得比实际可靠极端表现之后,常常只是回归均值 还会继续制造材料,让事后的故事看起来比数据本身整齐。

在结果揭晓之前写下判断和理由,事后才有记录可供核对。这也是写作的一种用处。

真正拿它来检查时,要在“我早就知道”出现后去找旧记录。找不到,就别把结果揭晓后的熟悉感当成当时的预测。

——收入 思维模型 · 心理学

看它的邻域 →