SBP: Disentangling Implicit and Explicit Bias from Responses to Exemplars and Categories

在招聘和录取决定中有据可查的偏见往往会阻止合格的个人参加劳动力和教育。目前许多解决这个问题的尝试都集中在减少“内隐”或“无意识”偏见上。与此同时，科学家们越来越多地辩论和批评内隐偏见作为一个概念。虽然对隐性偏见的认识已经变得司空见惯，但相关的科学还没有提供一个坚实的基础来建立干预措施。该项目的中心目标是通过识别人们对群体和群体标签的反应与特定个人和个人目标的差异，加强对内隐偏见的科学理解。它将揭示希望解决偏见的研究人员和组织是否应该关注“无意识”偏见，或者是否可以通过关注群体与个人的偏见来设计更有效的干预措施。该项目将有助于解决我们对偏见如何运作的理论理解中对隐性偏见的普遍误解。此外，它将产生解决在现实世界中的偏见有价值的见解。这个项目涉及12个实验，将解决一个关键的混淆，阻碍了能力得出强有力的结论有关的研究隐性偏见：混淆之间的测量类型（即，显式对隐式）和刺激类型（即，类别与范例）。传统上，研究外显偏见的研究通常要求参与者对抽象的社会类别做出反应（例如，黑人和白色人的类别），而隐式测量通常涉及这些类别的特定范例（例如，黑人和白色人的脸）。目前的项目使用基于计算机的反应时任务和行为的措施，探讨核心主题的内隐-外显关系。例如，常见的内隐和外显偏差测量之间的低相关性反映了（a）加工意识的差异，还是这些低相关性源于（B）刺激类型的差异？此外，以类别为重点的评价和以范例为重点的评价是否有不同的后果？通过区分外显和内隐测量，以及区分类别和范例，我们可以对偏见的原因和后果有新的认识，这对理论，实践，该奖项反映了NSF的法定使命，并通过使用基金会的智力价值进行评估，被认为值得支持和更广泛的影响审查标准。

Well-documented biases in hiring and admissions decisions often prevent qualified individuals from participating in the workforce and in education. Many current attempts to address this problem focus on reducing “implicit” or “unconscious” bias. At the same time, scientists are increasingly debating and critiquing implicit bias as a concept. While awareness of implicit biases has become commonplace, the associated science does not yet provide a solid foundation on which to build intervention efforts. The central goal of this project is to strengthen the scientific understanding of implicit bias by identifying differences in how people respond to groups and group labels, versus specific individuals and individual targets. It will reveal whether researchers and organizations wishing to address bias should be focusing on “unconscious” bias, or whether more effective interventions could be designed by focusing on bias toward groups versus individuals. The project will help address popular misconceptions of implicit bias from our theoretical understanding of how bias operates. In addition, it will generate valuable insights for addressing bias in real-world settings.This project involves 12 experiments that will tackle a key confound that obstructs the ability to draw strong conclusions about the research on implicit bias: the confound between type of measure (i.e., explicit vs. implicit) and stimuli type (i.e., categories vs. exemplars). Traditionally, research examining explicit bias typically requires participants to respond to abstract social categories (e.g., the categories Black people and White people), whereas implicit measures typically involve specific exemplars of those categories (e.g., faces of Black and White individuals). The current project uses computer-based reaction time tasks and behavioral measures to probe core themes regarding implicit-explicit relations. For example, do low correlations between common implicit and explicit measures of bias reflect (a) differences in processing awareness or do those low correlations derive from (b) the differences in stimuli type? Moreover, do category-focused evaluations and exemplar-focused evaluations have distinct consequences? By identifying the distinction between explicit and implicit measures on the one hand, and the distinction between categories and exemplars on the other, we can gain new insights into the causes and consequences of bias, with important implications for both theory, practice, and discrimination intervention programs.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.