Hanover Music Lab, Hanover University of Music, Drama and Media, Neues Haus 1, 30175, Hannover, Germany.
Department of Mathematics and Computer Science, University of Cagliari, Cagliari, Italy.
Behav Res Methods. 2024 Jan;56(1):362-378. doi: 10.3758/s13428-022-02048-3. Epub 2023 Jan 17.
HALT (The Headphone and Loudspeaker Test) Part II is a continuation of HALT Part I. The main goals of this study (HALT Part II) were (a) to develop screening tests and strategies to discriminate headphones from loudspeakers, (b) to come up with a methodological approach to combine more than two screening tests, and (c) to estimate data quality and required sample sizes for the application of screening tests. Screening Tests A and B were developed based on psychoacoustic effects. In a first laboratory study (N = 40), the two tests were evaluated with four different playback devices (circumaural and intra-aural headphones; external and laptop loudspeakers). In a final step, the two screening tests A and B and a previously established test C were validated in an Internet-based study (N = 211). Test B showed the best single-test performance (sensitivity = 80.0%, specificity = 83.2%, AUC = .844). Following an epidemiological approach, the headphone prevalence (17.67%) was determined to calculate positive and negative predictive values. For a user-oriented, parameter-based selection of suitable screening tests and the simple application of screening strategies, an online tool was programmed. HALT Part II is assumed to be a reliable procedure for planning and executing screenings to detect headphone and loudspeaker playback. Our methodological approach can be used as a generic technique for optimizing the application of any screening tests in psychological research. HALT Part I and II complement each other to form a comprehensive overall concept to control for playback conditions in Internet experiments.
HALT（耳机和扬声器测试）第二部分是 HALT 第一部分的延续。本研究（HALT 第二部分）的主要目标是：(a) 开发用于区分耳机和扬声器的筛选测试和策略；(b) 提出一种将超过两种筛选测试结合起来的方法学方法；(c) 估计筛选测试应用的数据质量和所需样本量。测试 A 和测试 B 是基于心理声学效应开发的。在第一项实验室研究（N = 40）中，使用四种不同的播放设备（全耳式和入耳式耳机；外部和笔记本电脑扬声器）评估了这两个测试。在最后一步，在基于互联网的研究（N = 211）中验证了这两个筛选测试 A 和 B 以及之前建立的测试 C。测试 B 显示出最佳的单测试性能（敏感性= 80.0%，特异性= 83.2%，AUC =.844）。按照流行病学方法，确定耳机的流行率（17.67%）以计算阳性和阴性预测值。为了用户导向、基于参数的筛选测试选择和筛选策略的简单应用，编写了一个在线工具。HALT 第二部分被认为是规划和执行筛查以检测耳机和扬声器播放的可靠程序。我们的方法学方法可用于优化任何心理研究中筛选测试的应用的通用技术。HALT 第一部分和第二部分相互补充，形成了一个全面的总体概念，以控制互联网实验中的播放条件。
