Kelareva Elena, Mewing James, Turpin Andrew, Wirth Anthony
University of Melbourne, Melbourne, Victoria, Australia.
Atten Percept Psychophys. 2010 Oct;72(7):2003-12. doi: 10.3758/APP.72.7.2003.
In this article, we present a new strategy for locating a point on a psychometric function (threshold determination) for yes-no procedures, called EntFirst. Our results show that it performs better than many existing strategies for estimating the 50% threshold that are commonly used in nonlaboratory settings. We also provide a review of existing algorithms for finding thresholds, with an emphasis on identifying the types of problems for which each algorithm is useful. Finally, we address a number of issues that are not adequately covered in the literature, including choosing an appropriate loss function to evaluate the performance of an algorithm for a given problem.
在本文中,我们提出了一种新策略,用于在是-否程序的心理测量函数上定位一个点(阈值确定),称为EntFirst。我们的结果表明,对于估计非实验室环境中常用的50%阈值,它比许多现有策略表现更好。我们还对现有的用于寻找阈值的算法进行了综述,重点是确定每种算法适用的问题类型。最后,我们讨论了文献中未充分涉及的一些问题,包括为给定问题选择合适的损失函数来评估算法的性能。