CAREER: Neural Mechanisms of Stereotypic Vision

通过文化学习，人们对不同类型的人和社会类别（如种族、性别、年龄）形成简化的联想（刻板印象）。我们可能在意识层面上不同意这些刻板印象，但它们也储存在我们的意识之外。一旦被激活，这些无意识的刻板印象会导致我们在评估他人和与他人互动时产生无意识的偏见。过去的研究表明，无意识的刻板印象只有在大脑完成对一个人的视觉处理后才会被激活。与此相反，目前的项目测试了刻板印象联想是否会影响大脑最初的视觉处理，从而在如何感知人脸或物体时产生系统的扭曲。这些扭曲使面孔或物体在视觉上看起来比现实中更符合刻板印象。这种刻板的视觉现象与日常的信念形成了鲜明的对比，即视觉感知是对现实的准确反映。通过扭曲与刻板印象联想一致的视觉感知，这种影响可能加深和维持储存的刻板印象联想。通过更好地理解形成刻板印象的神经机制及其改变能力，该项目可以为减少意外偏见的新技术铺平道路。这样做，该项目可以解决在刻板印象联想中创造长期变化的困难。这个职业项目建立在乔纳森·弗里曼博士尖端的脑成像和行为方法的基础上，以测试他的刻板印象视觉模型。通过他的专业软件，一个电脑鼠标跟踪技术将被用来测量手的运动轨迹，当他们看到人脸或物体时，他们会在屏幕上做出反应。尽管有意识的反应，手的轨迹同时吸引到一个未选择的替代提供了一个敏感的衡量刻板印象对感知过程的影响与毫秒精度。这些行为影响将与功能性磁共振成像（fMRI）模式分析相关联，从而评估感知偏差与大脑中面部或物体视觉表征的偏差之间的关系。七项研究将测试：(a)刻板印象是否会对面部或物体的视觉表征产生偏见，这是由涉及面部处理和视觉预测的大脑区域的相互作用驱动的；(b)这种刻板印象的影响是自动的，而不是有意识和故意的；(c)这种影响是灵活的，不局限于已有的社会类别（如种族、性别、年龄），也会出现在新获得的类别中，重要的是，可以通过反刻板印象干预来减少。该项目将推进与功能磁共振成像模式分析相结合的前沿行为方法。它们有可能在心理学、神经科学、认知科学和相关领域产生变革，为揭示“隐藏”认知状态的神经表征提供一种范式。通过这些研究获得的知识将用于为公众开发新的鼠标跟踪平台，以提高对微妙形式的偏见的认识，为未来的研究生成数据，并为改进技术以减少现实世界的偏见铺平道路。

Through cultural learning, people develop simplified associations (stereotypes) about different types of people and social categories (e.g., race, gender, age). We may not agree with these stereotypes at a conscious level, but they are also stored outside our awareness. Once activated, these unconscious stereotypes can lead to involuntary biases in how we evaluate and interact with others. Past research suggests that unconscious stereotypes are activated only after the brain is finished with visually processing a person. Contrary to this, the present project tests whether stereotypic associations influence the brain's initial visual processing, in turn creating systematic distortions in how a face or object is perceived. These distortions make faces or objects appear more visually consistent with stereotypes than they are in reality. Such a phenomenon of stereotypic vision contrasts with everyday beliefs that visual perception is an accurate reflection of reality. By distorting visual perception in line with stereotypic associations, such effects may deepen and maintain stored stereotypic associations. By providing a better understanding of the neural mechanisms that underlie stereotypic vision and their capacity for change, this project could pave the way for new techniques to reduce unintended biases. In doing so, this project could address the difficulty in creating long-term change in stereotypic associations. This CAREER project builds on Dr. Jonathan Freeman's cutting-edge brain-imaging and behavioral methodology to test his model of stereotypic vision. Through his specialized software, a computer mouse-tracking technique will be used to measure the trajectory of hand movements en route to response alternatives on the screen while they are presented with faces or objects. Despite conscious responses, hand trajectories' simultaneous attraction to an unselected alternative provides a sensitive measure of the effect of stereotypes on the process of perception with millisecond precision. These behavioral effects will be linked with functional magnetic resonance imaging (fMRI) pattern analyses, permitting an assessment of how biases in perception relate to biases in a face or object's visual representation in the brain. Seven studies will test whether (a) stereotypes bias the visual representation of a face or object, driven by the interplay of brain regions involved in face processing and visual predictions; (b) this effect of stereotypic vision is automatic, rather than conscious and deliberate; (c) this effect is flexible, such that it is not limited to pre-existing social categories (e.g., race, gender, age) but arises with newly acquired categories, and importantly, can be reduced by counter-stereotype interventions. This project will advance a cutting-edge behavioral methodology in tandem with fMRI pattern-analyses. They have the potential to be transformative across psychology, neuroscience, cognitive science, and related fields, providing a paradigm to reveal the neural representation of 'hidden' cognitive states. Knowledge gained through these studies will be leveraged to develop new mouse-tracking platforms for the general public to raise awareness of subtle forms of bias, generate data for future research, and pave the way for improved techniques to reduce real-world biases.