Yiyang Chen, Heather R Daly, Mark A Pitt, Trisha Van Zandt
{"title":"Assessing the distortions introduced when calculating d': A simulation approach.","authors":"Yiyang Chen, Heather R Daly, Mark A Pitt, Trisha Van Zandt","doi":"10.3758/s13428-024-02447-8","DOIUrl":null,"url":null,"abstract":"<p><p>The discriminability measure <math><msup><mi>d</mi> <mo>'</mo></msup> </math> is widely used in psychology to estimate sensitivity independently of response bias. The conventional approach to estimate <math><msup><mi>d</mi> <mo>'</mo></msup> </math> involves a transformation from the hit rate and the false-alarm rate. When performance is perfect, correction methods must be applied to calculate <math><msup><mi>d</mi> <mo>'</mo></msup> </math> , but these corrections distort the estimate. In three simulation studies, we show that distortion in <math><msup><mi>d</mi> <mo>'</mo></msup> </math> estimation can arise from other properties of the experimental design (number of trials, sample size, sample variance, task difficulty) that, when combined with application of the correction method, make <math><msup><mi>d</mi> <mo>'</mo></msup> </math> distortion in any specific experiment design complex and can mislead statistical inference in the worst cases (Type I and Type II errors). To address this problem, we propose that researchers simulate <math><msup><mi>d</mi> <mo>'</mo></msup> </math> estimation to explore the impact of design choices, given anticipated or observed data. An R Shiny application is introduced that estimates <math><msup><mi>d</mi> <mo>'</mo></msup> </math> distortion, providing researchers the means to identify distortion and take steps to minimize its impact.</p>","PeriodicalId":8717,"journal":{"name":"Behavior Research Methods","volume":" ","pages":"7728-7747"},"PeriodicalIF":4.6000,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Behavior Research Methods","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.3758/s13428-024-02447-8","RegionNum":2,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/7/3 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"PSYCHOLOGY, EXPERIMENTAL","Score":null,"Total":0}
引用次数: 0
Abstract
The discriminability measure is widely used in psychology to estimate sensitivity independently of response bias. The conventional approach to estimate involves a transformation from the hit rate and the false-alarm rate. When performance is perfect, correction methods must be applied to calculate , but these corrections distort the estimate. In three simulation studies, we show that distortion in estimation can arise from other properties of the experimental design (number of trials, sample size, sample variance, task difficulty) that, when combined with application of the correction method, make distortion in any specific experiment design complex and can mislead statistical inference in the worst cases (Type I and Type II errors). To address this problem, we propose that researchers simulate estimation to explore the impact of design choices, given anticipated or observed data. An R Shiny application is introduced that estimates distortion, providing researchers the means to identify distortion and take steps to minimize its impact.
心理学中广泛使用可辨别度量 d ' 来估算灵敏度,而不考虑反应偏差。估算 d ' 的传统方法包括对命中率和误报率进行转换。当表现完美时,必须使用校正方法来计算 d ' ,但这些校正会扭曲估计值。在三项模拟研究中,我们发现实验设计的其他属性(试验次数、样本大小、样本方差、任务难度)也会导致 d ' 估计值失真,这些属性与校正方法的应用相结合,会使任何特定实验设计中的 d ' 失真变得复杂,并在最坏的情况下误导统计推断(第一类和第二类错误)。为了解决这个问题,我们建议研究人员模拟 d' 估计,以探索设计选择对预期或观察数据的影响。我们介绍了一个 R Shiny 应用程序,它可以估算 d ' 失真,为研究人员提供识别失真并采取措施尽量减少其影响的方法。
期刊介绍:
Behavior Research Methods publishes articles concerned with the methods, techniques, and instrumentation of research in experimental psychology. The journal focuses particularly on the use of computer technology in psychological research. An annual special issue is devoted to this field.