Why do some (and only some) observations prompt people to ask "why?" We propose a functional approach to "Explanation-Seeking Curiosity" (ESC): the state that motivates people to seek an explanation. If ESC tends to prompt explanation search when doing so is likely to be beneficial, we can use prior work on the functional consequences of explanation search to derive "forward-looking" candidate triggers of ESC-those that concern expectations about the downstream consequences of pursuing explanation search. Across three studies (N = 867), we test hypotheses derived from this functional approach. In Studies 1-3, we find that ESC is most strongly predicted by expectations about future learning and future utility. We also find that judgments of novelty, surprise, and information gap predict ESC, consistent with prior work on curiosity; however, the role for forward-looking considerations is not reducible to these factors. In Studies 2-3, we find that predictors of ESC form three clusters, expectations about learning (about the target of explanation), expectations about export (to other cases and future contexts), and backward-looking considerations (having to do with the relationship between the target of explanation and prior knowledge). Additionally, these clusters are consistent across stimulus sets that probe ESC, but not fact-seeking curiosity. These findings suggest that explanation-seeking curiosity is aroused in a systematic way, and that people are not only sensitive to the match or mismatch between a given stimulus and their current or former beliefs, but to how they expect an explanation for that stimulus to improve their epistemic state.

中文翻译：

---

寻求解释的好奇心的功能方法

为什么有些（而且只有一些）观察会促使人们问“为什么？” 我们提出了一种“寻求解释的好奇心”（ESC）的功能方法：激励人们寻求解释的状态。如果 ESC 在这样做可能有益时倾向于提示解释搜索，我们可以使用先前关于解释搜索的功能后果的工作来推导出 ESC 的“前瞻性”候选触发器——那些关注对追求的下游后果的期望解释搜索。在三项研究 (N = 867) 中，我们测试了源自这种功能方法的假设。在研究 1-3 中，我们发现 ESC 最受对未来学习和未来效用的期望预测。我们还发现对新颖性、惊喜和信息差距的判断可以预测 ESC，与之前关于好奇心的工作一致；然而，前瞻性考虑的作用不能归结为这些因素。在研究 2-3 中，我们发现 ESC 的预测因子形成三个集群，对学习的期望（关于解释的目标）、对输出的期望（对其他案例和未来背景）和向后考虑（与解释目标与先验知识之间的关系）。此外，这些集群在探测 ESC 的刺激组中是一致的，但不是寻求事实的好奇心。这些发现表明，寻求解释的好奇心是以一种系统的方式引起的，人们不仅对给定刺激与他们当前或以前的信念之间的匹配或不匹配敏感，