模型判别中的最优实验设计
Optimal experimental design for model discrimination.
作者信息
Myung Jay I, Pitt Mark A
机构信息
Department of Psychology, Ohio State University, Columbus, OH 43210-1351, USA.
出版信息
Psychol Rev. 2009 Jul;116(3):499-518. doi: 10.1037/a0016104.
Abstract
Models of a psychological process can be difficult to discriminate experimentally because it is not easy to determine the values of the critical design variables (e.g., presentation schedule, stimulus structure) that will be most informative in differentiating them. Recent developments in sampling-based search methods in statistics make it possible to determine these values and thereby identify an optimal experimental design. After describing the method, it is demonstrated in 2 content areas in cognitive psychology in which models are highly competitive: retention (i.e., forgetting) and categorization. The optimal design is compared with the quality of designs used in the literature. The findings demonstrate that design optimization has the potential to increase the informativeness of the experimental method.
摘要
心理过程模型可能很难通过实验进行区分，因为要确定那些在区分模型时最具信息量的关键设计变量（例如，呈现时间表、刺激结构）的值并非易事。统计学中基于抽样的搜索方法的最新进展使得确定这些值并进而确定最佳实验设计成为可能。在描述了该方法之后，在认知心理学的两个模型竞争激烈的内容领域进行了演示：记忆保持（即遗忘）和分类。将最佳设计与文献中使用的设计质量进行了比较。研究结果表明，设计优化有可能提高实验方法的信息量。
相似文献
1
Optimal experimental design for model discrimination.模型判别中的最优实验设计
Psychol Rev. 2009 Jul;116(3):499-518. doi: 10.1037/a0016104.
2
Power equivalence in structural equation modelling.结构方程建模中的功率等效性。
Br J Math Stat Psychol. 2010 May;63(Pt 2):257-72. doi: 10.1348/000711009X441021. Epub 2009 Jun 12.
3
Uncovering mental representations with Markov chain Monte Carlo.用马尔可夫链蒙特卡罗方法揭示心理表象。
Cogn Psychol. 2010 Mar;60(2):63-106. doi: 10.1016/j.cogpsych.2009.07.001. Epub 2009 Aug 22.
4
Evolving Improved Sampling Protocols for Dose-Response Modelling Using Genetic Algorithms with a Profile-Likelihood Metric.利用基于似然度度量的遗传算法改进剂量-反应建模的抽样方案。
Bull Math Biol. 2024 May 8;86(6):70. doi: 10.1007/s11538-024-01304-1.
5
Robust designs in longitudinal studies accounting for parameter and model uncertainties - application to count data.纵向研究中考虑参数和模型不确定性的稳健设计——在计数数据中的应用。
J Biopharm Stat. 2020;30(1):31-45. doi: 10.1080/10543406.2019.1607367. Epub 2019 Apr 27.
6
Model discrimination through adaptive experimentation.通过自适应实验进行模型判别。
Psychon Bull Rev. 2011 Feb;18(1):204-10. doi: 10.3758/s13423-010-0030-4.
7
Adaptive stimulus selection for multi-alternative psychometric functions with lapses.具有失误情况的多选项心理测量函数的自适应刺激选择
J Vis. 2018 Nov 1;18(12):4. doi: 10.1167/18.12.4.
8
Thermodynamic integration via differential evolution: A method for estimating marginal likelihoods.通过差分进化进行热力学集成：一种估计边缘似然的方法。
Behav Res Methods. 2019 Apr;51(2):930-947. doi: 10.3758/s13428-018-1172-y.
9
A comparison of computational algorithms for the Bayesian analysis of clinical trials.临床试验贝叶斯分析的计算算法比较。
Clin Trials. 2024 Dec;21(6):689-700. doi: 10.1177/17407745241247334. Epub 2024 May 16.
10
Harnessing the theoretical foundations of the exponential and beta-Poisson dose-response models to quantify parameter uncertainty using Markov Chain Monte Carlo.利用指数和贝塔-泊松剂量反应模型的理论基础，通过马尔可夫链蒙特卡罗方法来量化参数不确定性。
Risk Anal. 2013 Sep;33(9):1677-93. doi: 10.1111/risa.12006. Epub 2013 Jan 11.
引用本文的文献
1
People display consistent recency and primacy effects in behavior and neural activity across perceptual and value-based judgments.在基于感知和价值的判断中，人们在行为和神经活动方面表现出一致的近因效应和首因效应。
Cogn Affect Behav Neurosci. 2025 Mar 26. doi: 10.3758/s13415-025-01285-1.
2
Automating the practice of science: Opportunities, challenges, and implications.科学实践的自动化：机遇、挑战与影响。
Proc Natl Acad Sci U S A. 2025 Feb 4;122(5):e2401238121. doi: 10.1073/pnas.2401238121. Epub 2025 Jan 27.
3
Designing optimal behavioral experiments using machine learning.使用机器学习设计最优行为实验。
Elife. 2024 Jan 23;13:e86224. doi: 10.7554/eLife.86224.
4
Visuospatial information foraging describes search behavior in learning latent environmental features.视空间信息搜索描述了在学习潜在环境特征中的搜索行为。
Sci Rep. 2023 Jan 20;13(1):1126. doi: 10.1038/s41598-023-27662-9.
5
A Bayesian perspective on severity: risky predictions and specific hypotheses.贝叶斯视角下的严重程度：风险预测和具体假设。
Psychon Bull Rev. 2023 Apr;30(2):516-533. doi: 10.3758/s13423-022-02069-1. Epub 2022 Aug 15.
6
Empirical underidentification in estimating random utility models: The role of choice sets and standardizations.估计随机效用模型中的经验不足识别：选择集和标准化的作用。
Br J Math Stat Psychol. 2022 May;75(2):252-292. doi: 10.1111/bmsp.12256. Epub 2021 Nov 8.
7
SweetPea: A standard language for factorial experimental design.甜豌豆：析因实验设计的标准语言。
Behav Res Methods. 2022 Apr;54(2):805-829. doi: 10.3758/s13428-021-01598-2. Epub 2021 Aug 6.
8
Digital Learning Games for Mathematics and Computer Science Education: The Need for Preregistered RCTs, Standardized Methodology, and Advanced Technology.用于数学和计算机科学教育的数字学习游戏：对预先注册的随机对照试验、标准化方法和先进技术的需求。
Front Psychol. 2020 Oct 15;11:2127. doi: 10.3389/fpsyg.2020.02127. eCollection 2020.
9
Hybrid algorithms for generating optimal designs for discriminating multiple nonlinear models under various error distributional assumptions.用于在各种误差分布假设下生成用于区分多个非线性模型的最优设计的混合算法。
PLoS One. 2020 Oct 5;15(10):e0239864. doi: 10.1371/journal.pone.0239864. eCollection 2020.
10
Rapid, precise, and reliable measurement of delay discounting using a Bayesian learning algorithm.使用贝叶斯学习算法快速、准确、可靠地测量延迟折扣。
Sci Rep. 2020 Jul 21;10(1):12091. doi: 10.1038/s41598-020-68587-x.
本文引用的文献
1
A model for recognition memory: REM-retrieving effectively from memory.一种再认记忆模型：REM-有效地从记忆中提取。
Psychon Bull Rev. 1997 Jun;4(2):145-66. doi: 10.3758/BF03209391.
2
Sequential optimal design of neurophysiology experiments.神经生理学实验的序贯优化设计
Neural Comput. 2009 Mar;21(3):619-87. doi: 10.1162/neco.2008.08-07-594.
3
In search of abstraction: the varying abstraction model of categorization.寻找抽象：分类的不同抽象模型。
Psychon Bull Rev. 2008 Aug;15(4):732-49. doi: 10.3758/pbr.15.4.732.
4
Does response scaling cause the generalized context model to mimic a prototype model?反应缩放是否会导致广义语境模型模仿原型模型？
Psychon Bull Rev. 2007 Dec;14(6):1043-50. doi: 10.3758/bf03193089.
5
Optimization by simulated annealing.模拟退火优化。
Science. 1983 May 13;220(4598):671-80. doi: 10.1126/science.220.4598.671.
6
Bayesian adaptive estimation of threshold versus contrast external noise functions: the quick TvC method.阈值与对比度外部噪声函数的贝叶斯自适应估计：快速TvC方法
Vision Res. 2006 Oct;46(19):3160-76. doi: 10.1016/j.visres.2006.04.022. Epub 2006 Jun 19.
7
Common and distinctive features in stimulus similarity: a modified version of the contrast model.刺激相似性中的共同特征与独特特征：对比模型的修正版
Psychon Bull Rev. 2004 Dec;11(6):961-74. doi: 10.3758/bf03196728.
8
A theory of interactive parallel processing: new capacity measures and predictions for a response time inequality series.一种交互式并行处理理论：响应时间不等式系列的新容量度量与预测
Psychol Rev. 2004 Oct;111(4):1003-35. doi: 10.1037/0033-295X.111.4.1003.
9
Assessing the distinguishability of models and the informativeness of data.评估模型的可区分性和数据的信息量。
Cogn Psychol. 2004 Aug;49(1):47-84. doi: 10.1016/j.cogpsych.2003.11.001.
10
PRIMARY MEMORY.主存储器
Psychol Rev. 1965 Mar;72:89-104. doi: 10.1037/h0021797.
Zotero 插件