Karlsson Niklas, Juliusson Asgeir, Grankvist Gunne, Gärling Tommy
Department of Psychology, Göteborg University, Sweden.
Acta Psychol (Amst). 2002 Nov;111(3):309-22. doi: 10.1016/s0001-6918(02)00056-2.
This research investigates the sunk-cost effect or escalation defined as the irrational tendency to choose to continue to invest money, time, or effort following unsuccessful investments. Building on previous research demonstrating a loss-sensitivity principle in sequential decision making, the hypothesis was proposed that a loss-minimization goal would lead to stronger effects of sunk outcomes (prior gains and losses) than would a gain-maximizing goal. The hypothesis was investigated in three experiments with undergraduates responding to investment decision scenarios. Although the tendency to continue investments was always larger for gain-maximizing than for loss-minimizing goal instructions, the sunk-outcome effect was stronger in the former case. However, when the decisions were personal and concerned lower stakes rather than business investments involving large amounts of money, the expected stronger effect of sunk outcomes was found for loss-minimizing goal instructions. Another finding was that the expected value was never ignored, thus suggesting that future research should focus on the joint effects of the expected value and sunk outcomes.
本研究探讨了沉没成本效应或升级，其定义为在投资未成功后选择继续投入资金、时间或精力的非理性倾向。基于先前在序贯决策中证明损失敏感性原则的研究，提出了一个假设，即损失最小化目标比收益最大化目标会导致沉没结果（先前的收益和损失）产生更强的影响。该假设在三项实验中进行了研究，实验对象为对投资决策情景做出反应的本科生。尽管收益最大化目标指令下继续投资的倾向总是大于损失最小化目标指令下的倾向，但沉没结果效应在前一种情况下更强。然而，当决策是个人性质的且涉及较低风险，而非涉及大量资金的商业投资时，在损失最小化目标指令下发现了沉没结果预期更强的效应。另一个发现是预期价值从未被忽视，因此表明未来的研究应关注预期价值和沉没结果的联合效应。
