面孔可信任度集群表征的认知特点及神经机制
文献类型：学位论文
| 作者 | 励奇添
|
| 答辩日期 | 2022-06 |
| 文献子类 | 博士 |
| 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 | 陈文锋 |
| 关键词 | 集群表征 面孔集合 面孔可信任度 平行加工机制 奖赏学习 |
| 学位名称 | 理学博士 |
| 学位专业 | 基础心理学 |
| 其他题名 | The cognitive characteristics and neural mechanisms of ensemble representation of facial trustworthiness |
| 中文摘要 | Ensemble representation is a process by which people handle a set of stimuli and extract its gist. Previous researches have explored ensemble representation with facial emotion, identity, attractiveness, and other dimensions. However, face is a very informative stimulus, and a set of faces is able to convey certain social information, such as ensemble trustworthiness. Studies on the trustworthiness judgment with individual faces are ample, but studies on the ensemble representation of facial trustworthiness are still very scarce. In this study, using behavioral experiments, event-related potential technologies, and functional magnetic resonance imaging techniques, we investigated mainly three issues: 1) whether trustworthiness could be process within a set, and what kind of relationship existed between set trustworthiness and member trustworthiness, 2) what were the cognitive characteristics of this process and, 3) what were the neural mechanisms characterized by this process. Study 1 explored whether the trustworthiness of a set of faces could be processed as a whole, and what kind of relationship was between the ensemble trustworthiness of the set and the mean of the members. Experiment 1 found that the scores of the ensemble assessment was significantly higher than the member mean, and there was a positive correlation between the two. The same results were found in Experiment 2. Meanwhile, the D-value between ensemble assessment and member mean differed dependent on set type and participant gender. Study 2 explored the cognitive characteristics of the ensemble representation of trustworthiness. Experiment 3 to Experiment 5 investigated respectively the impact of member variability, simultaneous presentation-sequence presentation conditions, and reward learning tasks on the ensemble representation of trustworthiness. 1) Experiment 3 found a small effect of variation on ensemble representation, and that the difference between ensemble assessment and member mean, and the interaction between set type and participant gender was no longer significant. 3) Experiment 4 introduced the simultaneous presentation-sequence presentation paradigm, and found that the ensemble representation of trustworthiness may be more in line with parallel processing mechanism. 4) Experiment 5 combined ensemble representation task with reward learning task, and found that sets with higher reward probability gained trustworthiness, and previously high-trustworthy sets with lower reward probability lost trustworthiness. Model comparison further provided different initial expected- values based on trustworthiness and different learning rates based on congruency. Study 3 used event-related potential technology and functional magnetic resonance imaging technology to explore event-related potential characteristics and brain activation patterns induced by ensemble trustworthiness. 1) Experiment 6 found difference of P1 amplitudes between heterogeneous sets and homogenous sets, and difference of N170 amplitudes between the left hemisphere and the right hemisphere. Meanwhile, the EPN amplitude induced by female medium trustworthiness sets was smaller than high trustworthiness sets, while the LPP amplitude induced was greater than high trustworthiness sets. 2) Experiment 7 found that ensemble trustworthiness processing employed the right angular gyrus, the left temporal middle gyrus and the right frontal middle/superior gyrus. The results from Study 1 showed that facial trustworthiness could be represented as a whole, but there was a difference between the ensemble estimation and the member mean. Moreover, the size and the direction of the difference would change, dependent on set type and participant gender. The results from Study 2 showed that member variation only had a small effect upon the ensemble representation of trustworthiness, and the process was more in line with the parallel mechanism. Also, ensemble assessment was flexible, and could influence reward learning tasks. The results from Study 3 explored further about the procedure and brain activation of trustworthiness ensemble representation. This study systematically explored the cognitive characteristics of the ensemble representation of facial trustworthiness, and laid a foundation for the study of ensemble representations of further social information. Also, this study preliminarily explored the neural mechanism of the ensemble representation of facial trustworthiness, and provides experimental evidence for its event-related potential characteristics and relevant brain regions. |
| 英文摘要 | 集群表征是人们对刺激集合进行加工并提取其整体特征的过程。前人研究已经发现，人们能够对面孔情绪、身份、吸引力等维度进行集群表征。但是，面孔是一种信息量非常丰富的刺激，面孔集合能够传递许多的社会信息，例如一群人整体上是否可信，即面孔集合的平均可信任度。单个面孔的可信任度研究比较充分，但是对面孔可信任度的集群表征研究还非常少。本研究通过行为实验、事件相关电位技术、功能磁共振成像技术，对可信任度的集群表征进行了探索，主要探讨三个问题：1）面孔可信任度能否被集群表征，集合平均可信任度与成员可信任度均值之间是什么关系，2）可信任度的集群表征有什么样的认知特点，成员平均值、变异性、单位时间信息加工量以及社会互动会对其产生何种影响，3）可信任度集群表征的神经机制是什么样的，单面孔加工和集群表征诱发的事件相关电位有何差异，其脑激活模式又有何不同。 研究一探讨了面孔的可信任度能否被集群表征，集合的平均可信任度与成员可信任度均值之间是什么样的关系。实验一发现，集群评价的分数显著地高于成员均值，且两者存在正相关。实验二也发现了类似结果。另一方面，集群评价与成员均值的偏差会因为集合种类和被试性别的变化而发生变化。 研究二在研究一的基础上探索了面孔可信任度集群表征的认知特点，实验三到实验五分别探讨了成员变异性、同时呈现——序列呈现条件，以及奖赏学习对可信任度集群表征的影响。1）实验三发现成员变异性对可信任度集群表征的影响较小，同时集群评价与成员均值的差异，集合种类和被试性别的交互作用不再显著。2）实验四引入了同时呈现——序列呈现范式，发现可信任度的集群表征可能更符合平行加工机制。3）实验五将集群表征任务与奖赏学习任务相结合，发现奖赏学习任务中奖励概率较高的面孔集合，其可信任度无论是集群评价还是成员均值都在前后测间发生了上升。同时，在奖赏学习任务中奖励概率较低的高可信任度集合，其可信任度的集群评价在前后测间发生了下降，而成员均值则没有显著变化。进一步的模型拟合还发现了基于面孔集合可信任度的初始期望差异和基于一致性的学习率差异。 研究三采用事件相关电位技术和功能磁共振成像技术，探索面孔可信任度集群表征的事件相关电位特征和脑激活模式。1）实验六发现异质集合和同质集合诱发的 P1 波幅存在差异，而整体上 N170 的波幅在左半球小于右半球。同时，女性中可信任度集合诱发的 EPN（Early Posterior Negativity）波幅小于高可信任度集合，诱发的 LPP（Late Positive Potential）波幅大于高可信任度集合。2）实验七发现面孔集合加工可能会激活右侧角回、左侧颞中回和右侧额中回/额上回区域，而单面孔加工相比可能会激活壳核。 研究一的结果表明，面孔可信任度能够被集群表征，但是集群评价与成员均值存在差异，整体上出现了高估的现象。另外这种差异的大小和方向会因为集合种类和被试性别的变化而改变。研究二的结果表明，成员变异性对可信任度集群表征的影响较小，同时可信任度的集群表征可能更符合平行加工机制，且具有一定的可塑性，能够对奖赏学习产生影响。研究三进一步探索了可信任度集群表征的时间进程和脑激活模式。 本研究系统探讨了面孔可信任度集群表征的认知特点，为面孔集合社会信息的集群表征研究打下了基础。同时本研究初步探索了可信任度集群表征的神经机制，对其相关电位特征和特异性脑区提供初步的实验证据。 |
| 语种 | 中文 |
| 源URL | [http://ir.psych.ac.cn/handle/311026/43155]  |
| 专题 | 心理研究所_认知与发展心理学研究室 |
| 推荐引用方式 GB/T 7714 | 励奇添. 面孔可信任度集群表征的认知特点及神经机制[D]. 中国科学院心理研究所. 中国科学院大学. 2022. |
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。