The world is full of complex environments in which individuals must plan a series of choices to obtain some desired outcome. In these situations, entire sequences of events, including one’s future decisions, should be considered before taking an action. Backward induction provides a normative strategy for planning, in which one works backward, deterministically, from the end of a scenario. However, this model often fails to account for human behavior. This article proposes an alternative account, decision field theory–planning (DFT-P), in which individuals plan future choices on the fly through repeated forward-looking mental simulations. As they imagine the possible outcomes of their actions, decision makers simulate their future choices moment to moment. A key prediction of DFT-P is that payoff variability produces noisy simulations and reduces sensitivity to value differences. In two experiments, a robust multistage payoff variability effect was found, with preferences becoming weaker as variability increased. A formal comparison showed that DFT-P provided a good account of people’s behavior, while a heuristic model and a flexible version of the backward induction model did not. These results confirm a fundamental prediction of DFT-P, and demonstrate its utility as a tool for understanding how people plan future choices and allocate cognitive resources in multistage decision making.

中文翻译：

---

决策场理论-规划：多阶段决策中动态规划的认知模型。

世界充满了复杂的环境，在这些环境中，个人必须计划一系列选择才能获得某些期望的结果。在这些情况下，在采取行动之前应该考虑整个事件序列，包括一个人未来的决定。向后归纳提供了一种规范的规划策略，其中从场景的末尾开始，确定性地向后工作。然而，这个模型通常无法解释人类行为。本文提出了另一种说法，即决策场理论规划 (DFT-P)，其中个人通过重复的前瞻性心理模拟来即时规划未来的选择。当他们想象自己的行为可能产生的结果时，决策者会时时刻刻模拟他们未来的选择。DFT-P 的一个关键预测是收益可变性会产生嘈杂的模拟并降低对价值差异的敏感性。在两个实验中，发现了强大的多阶段收益可变性效应，随着可变性的增加，偏好变得更弱。正式的比较表明，DFT-P 很好地说明了人们的行为，而启发式模型和后向归纳模型的灵活版本则没有。这些结果证实了 DFT-P 的基本预测，并证明了其作为一种工具的效用，可用于了解人们如何在多阶段决策中规划未来选择和分配认知资源。正式的比较表明，DFT-P 很好地说明了人们的行为，而启发式模型和后向归纳模型的灵活版本则没有。这些结果证实了 DFT-P 的基本预测，并证明了其作为一种工具的效用，可用于了解人们如何在多阶段决策中规划未来选择和分配认知资源。正式的比较表明，DFT-P 很好地说明了人们的行为，而启发式模型和后向归纳模型的灵活版本则没有。这些结果证实了 DFT-P 的基本预测，并证明了其作为一种工具的效用，可用于了解人们如何在多阶段决策中规划未来选择和分配认知资源。