Sampling Effects through Self-Truncated Information Search

本延续提案的目的是为德国联邦研究基金支持的一个研究项目再获得一年的延长资金。虽然用这笔赠款进行的研究(致力于“通过自我截断信息搜索的抽样效应”)不仅在经验结果和出版物产出方面取得了成功，而且在理论进展、会议活动和与其他研究小组富有成效的合作方面也取得了成功，但所获得的结果提出了新的挑战性问题，需要继续进行研究。特别是，到目前为止，我们在自我截断信息搜索的影响以及我们所称的布伦斯威克抽样和瑟斯顿抽样的相互作用方面获得的经验结果证明要强得多，并导致了比我们最初预期更精细的相互作用。因此，不追求这些有希望的新想法将是一种遗憾。具体地说，我们打算将拟议的持续资金用于四个方面的研究：(1)我们打算进行一系列新的计算机模拟，以了解通过自我截断搜索经常带来的信息量与决策质量之间的负相关。(2)将专门设计一个实验，以测试不同的方法来操纵瑟斯顿抽样对印象形成的影响。(3)另一个实验的目的是加深我们对组间判断中停止规则的理解，解释不同截断策略导致的组外同质性和组内异质性。此外，(4)我们计划再进行一次实验，将自我截断效应的概念应用到有意义的建议采纳的应用语境中。

The purpose of the present continuation proposal is to obtain extended funding for one additional year for a research project supported by the Deutsche Forschungsgemeinschaft. Although the research conducted with this grant (devoted to “Sampling effects through self-truncated information search”) was successful not just in terms of empirical results and publication output, but also in terms of theoretical progress, conference activities, and fruitful collaboration with other research groups, the obtained findings have raised new challenging questions that call for continued research. In particular, the empirical results we have obtained so far on the impact of self-truncated information search and on the interplay of what we call Brunswikian and Thurstonian sampling turned out to be much stronger, and led to more refined interactions than we had expected originally. It would thus be a pity not to pursue these promising new ideas. Specifically, we intend to use the proposed continued funding for four lines of research: (1) We intend to conduct a new series of computer simulations to understand the negative correlation between amount of information and decision quality that is regularly brought about through self-truncated search. (2) One experiment will be specifically designed to test distinct ways of manipulating Thurstonian sampling effects on impression formation. (3) The aim of another experiment is to deepen our understanding of stopping rules in the context of intergroup judgment, explaining outgroup homogeneity and ingroup heterogeneity as a result of different truncation strategies. Furthermore, (4) we plan to conduct one more experiment in which we apply the notion of self-truncation effects in the meaningful applied context of advice taking.