基于眼动的内隐社会认知测量研究
文献类型：学位论文
| 作者 | 程曦 |
| 答辩日期 | 2024-06 |
| 文献子类 | 硕士 |
| 授予单位 | 中国科学院大学 |
| 授予地点 | 中国科学院心理研究所 |
| 其他责任者 | 蔡华俭 ; 杨紫嫣 |
| 关键词 | 内隐刻板印象 内隐态度 眼动 内隐联系测验 |
| 学位名称 | 理学硕士 |
| 学位专业 | 基础心理学 |
| 其他题名 | Exploring Implicit Social Cognitive Measurement through Eye Tracking |
| 中文摘要 | The effective measurement of implicit social cognition has been a central concern since the advent of this research field. Researchers have developed various measurement tasks, including priming paradigms, classification paradigms, and paper-and-pencil tests, but they face issues such as low predictive and content validity and the inability to assess implicit social cognition in early childhood and infancy. This paper proposes a novel approach by integrating the measurement methods based on priming and classification paradigms with eye-tracking technology to assess individual implicit social cognition through eye movement patterns. The aim is to design an innovative and widely applicable measurement method, the "Eye Movement Implicit Test" (EMIT). This project consists of two studies to measure two common facets of implicit social cognition - implicit stereotypes and implicit attitudes - to validate the effectiveness of this method. Each study encompasses two sub-studies, one grounded in the categorical paradigm and the other in the priming paradigm. In these studies, participants view images that are either congruent or incongruent with implicit stereotypes, attitudes, and self-esteem under two experimental paradigms while their eye movements (gaze, saccades, and pupil responses) are recorded. Differences in eye movement patterns between congruent and incongruent conditions are employed to construct implicit measurement indicators. Implicit measurement results from the corresponding Implicit Association Test and explicit tests serve as criteria for evaluating eye movement-based implicit measurement. The results of Study 1 indicated that in the category paradigm experiment, the stereotype condition significantly positively predicted gaze time, number of gazes, number of regressions, while significantly negatively predicting first arrival time and initial scanning amplitude. The measurement index "E score," based on the number of regressions, demonstrated good reliability and validity, serving as a measurement index for individual differences in implicit stereotypes; the smaller the value, the higher the individual's implicit stereotypes. In the eye-tracking experiment based on the priming paradigm, the stereotype condition significantly negatively predicted first fixation time, initial scanning amplitude, and initial fixation ratio. The "E score" of first fixation time and initial fixation ratio demonstrated good reliability and validity, serving as measurement indices for individual differences in implicit stereotypes; higher values indicate higher levels of individual implicit stereotypes. The results of Study 2 revealed that in the category paradigm experiment, the racial attitude condition significantly positively predicted participants' gaze time, first fixation time, first saccadic angle, average pupil size, and maximum pupil size. The "E score" of first saccadic angle, average pupil size, and maximum pupil size demonstrated good reliability, measuring individual explicit racial attitudes; higher values indicate lower levels of individual explicit racial bias. In the priming paradigm experiment, the racial attitude condition significantly negatively predicted participants' gaze time, number of gazes, first fixation time, first saccadic angle, and final fixation ratio. The "E score" of first fixation time demonstrated good reliability, serving as a measurement index for individual differences in implicit racial attitudes; lower values indicate higher levels of individual implicit racial bias. The "E score" of gaze time also demonstrated good reliability, measuring individual explicit racial attitudes; lower values indicate higher levels of individual explicit racial bias. This study aims to provide an initial confirmation of the reliability and validity of implicit measurement indices based on eye movement technology. It offers a novel reference point for indirectly assessing implicit social cognition, significantly contributing to the theoretical development, measurement, and research on the development of implicit social cognition. |
| 英文摘要 | 自内隐社会认知兴起以来，如何对其进行有效的测量一直是研究者们关注的 核心问题之一。迄今，研究者们已经基于启动范式、分类范式、纸笔测验等开发 了很多测量任务。这些任务虽被广泛应用，但仍存在许多问题，比如：预测效度 和内容效度较低、使用存在一定的认知能力要求、难以测量婴幼儿时期的内隐社 会认知等。本文将传统内隐测量中的启动范式和分类范式与眼动技术相结合，期 望设计一种新的、有效的、应用范围广泛的内隐社会认知测量方法——眼动内隐 测验。具体来说，本文由两项研究组成，每项研究又包含两项子研究，分别考察 被试在分类范式和启动范式下观察内隐刻板印象、内隐态度的相容或不相容图片 时的注视、眼跳、瞳孔变化三类眼动的情况，并采用相容和不相容两种条件下的 眼动差异构建个体差异测量的指标，最后以对应的内隐联系测验和外显测验的结 果作为效标，检验内隐眼动测验的信效度。 研究一的结果表明：基于分类范式的实验中，刻板印象条件能显著正向预测 注视时间、注视次数、回视次数，能显著负向预测首次到达时间、首次扫视幅度， 其中基于回视次数构建的测量指标“E 分数”具有良好的信效度，可以作为内隐 刻板印象个体差异的测量指标，其值越小表明个体的内隐刻板印象越高；基于启 动范式的眼动实验中，刻板印象条件能显著地负向预测首次凝视时间、首次扫视 幅度和首次注视比例，其中首次凝视时间的“E 分数”和首次注视比例的“E 分 数”具有良好的信效度，可以作为内隐刻板印象个体差异的测量指标，其值越高 说明个体的内隐刻板印象越高。研究二的结果表明：基于分类范式的实验中，态 度条件能显著正向预测被试的注视时间、首次凝视时间、首次眼跳角度、平均瞳 孔尺寸和最大瞳孔尺寸，其中首次眼跳角度、平均瞳孔尺寸和最大瞳孔尺寸的“E 分数”具有良好的信度，且能测量个体的外显种族态度，其值越低说明个体的外 显种族偏见越高；基于启动范式的实验中，态度条件能显著负向预测被试的注视 时间、注视次数、首次凝视时间、首次眼跳角度和末次注视比例，其中首次凝视 时间的“E 分数”具有良好的信效度，可以作为内隐种族态度个体差异的测量指 标，其值越低说明个体的内隐种族偏见越高，注视时间的“E 分数”也具有良好 的信度且能测量个体的外显种族态度，其值越低说明个体的外显种族偏见越高。 本研究初步证实了以眼动为基础构建的内隐测量指标有良好的信效度，为内 隐社会认知的间接测量提供一种新的参考指标，对内隐社会认知的理论发展、内 隐社会认知的测量以及婴幼儿的内隐社会认知发展研究具有重要的意义。 |
| 语种 | 中文 |
| 源URL | [http://ir.psych.ac.cn/handle/311026/47947]  |
| 专题 | 心理研究所_社会与工程心理学研究室 |
| 推荐引用方式 GB/T 7714 | 程曦. 基于眼动的内隐社会认知测量研究[D]. 中国科学院心理研究所. 中国科学院大学. 2024. |
入库方式： OAI收割
来源：心理研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。