整合编码对联结再认的影响及其认知神经机制

PSYCH OpenIR > 认知与发展心理学研究室

	整合编码对联结再认的影响及其认知神经机制
其他题名	The Effect of Unitization on Associative Recognition and its Neural Mechanisms
	赵敏芳
	2015-11
摘要	双加工理论认为情景记忆由熟悉性与回想两种不同的认知加工过程所支持。一般认为，仅回想能够支持联结记忆。然而，近年有研究发现当刺激材料能够被编码为一个整合表征时，熟悉性也能够支持联结记忆，被称之为“整合假说”。为了系统验证整合假说，本研究从日常记忆角度出发，使用事件相关电位技术，以面孔和动作短语为刺激材料，通过操纵面孔特征间或动作和物体之间的整合程度来考察整合对联结再认提取过程的影响及其认知神经机制。研究一利用面孔正立或倒立呈现以操纵特征整合程度，以考察整合编码对特征间联结再认提取过程的影响。使用联结再认任务范式，实验1，学习阶段，让被试识记由内部特征以及外部特征配对组成的正立面孔和倒立面孔；测试阶段，要求被试区分学习过的“相同”、“重组合”以及“新”面孔。结果表明正立面孔的联结再认成绩显著高于倒立面孔的成绩。“相同vs.新”和“相同vs.重组合”的早期额区新旧效应均出现在正立面孔条件而非倒立面孔条件。“相同vs.新”的晚期顶区新旧效应未出现在正立面孔条件下的联结提取过程。实验2，仅考察正立面孔条件，并通过加强学习以提升被试的整体记忆水平。结果重复了早期额区新旧效应，并观察到与回想相关的晚期顶区新旧效应。研究一证实当面孔特征可以被整合为一个融合表征时(正立面孔)，熟悉性能够支持特征间的联结再认。研究二利用SPT 和VT 编码以操纵动作与物体的整合程度来考察整合编码对动作与物体的联结再认的影响。其次，再进一步考察动作与物体的实验前关联是否影响联结再认。学习阶段，被试使用SPT 与VT 编码方式识记日常动作短语(动作-物体对，如，“切西瓜”)(实验3)和怪异动作短语(如，“种锤子”)(实验4)；测试阶段，要求被试区分“相同”、“重组合”以及“新”短语。ERP 结果发现： 1)对于日常短语：SPT(而非VT)编码后的联结再认观察到“相同vs.新”和“相同vs.重组合”的早期额区新旧效应。两个编码条件，均发现“相同vs.新”的晚期顶区新旧效应；仅SPT 编码条件，发现“相同vs.重组合”的晚期新旧效应。 2)对于怪异短语：在SPT 和VT 编码条件，均存在“相同vs.新”的早期额区新旧效应及晚期顶区新旧效应。仅VT 编码条件，发现“相同vs.重组合”的晚期新旧效应。研究二结果表明对于日常动作短语，SPT 编码可以促进熟悉性与回想在联结再认的作用，从而提升联结记忆水平。仅在VT 编码下，怪异性可以促进熟悉性与回想在联结再认的作用，两者继而提高联结记忆成绩。总而言之，本论文的研究从实现整合的两大途径——利用固有属性和编码策略考察了整合假说，以上结果强化并支持当刺激材料可以被整合为一个融合的表征时，熟悉性能够支持联结再认的提取。
其他摘要	Dual process models of episodic recognition propose that two different cognitive processes–familiarity and recollection–can contribute to memory performances. Associative recognition is traditionally believed to be solely supported by ecollection. However, recent studies have challenged this view by demonstrating that familiarity can also contribute to associative recognition when to-be-remembered stimuli are perceived as a unitized representation. This is usually referred to as the “unitization hypothesis”. Our primary goal is to address this important issue in four event-related potential (ERP) studies in order to further understand the underlying mechanisms of associative recognition that closely related to everyday memory for actions and faces. Experiment 1 and Experiment 2 examined whether both familiarity and recollection could contribute to associative recognition of face features when faces are presented upright. In Experiment 1, participants studied upright and inverted faces consisting of internal features (specific mouth, nose, and eyes) paired with external features (specific hair, chin, and ears). During testing, participants were required to discriminate between intact, recombined and new faces. Recombined faces consisted of internal and external features taken from two different studied faces. Results revealed that early frontal old-new effects from comparisons of both “intact vs. new” and “intact vs. recombined” were observed in the upright but not inverted faces. In the inverted faces condition, we observed the late parietal old-new effect from comparison of “intact vs. new” (associated with recollection), but this effect was not observed in the upright faces condition. In Experiment 2, we presented only upright faces and improved memory via a more extensive learning phase. We replicated the early old-new effects for intact faces with respect to both new and recombined faces. With higher memory performance, the late parietal old-new effects were also observed. We take these results as further evidence for the contribution of familiarity to associative recognition if facial features are encoded as a coherent whole, as in upright faces. In Experiment 3 and Experiment 4, we investigated whether familiarity could contribute to associative retrieval of action and object, when adopting SPT (self-performed task) during encoding. Additionally, we are also interested in whether the SPT effect and memory processes during associative recognition were influenced by the strength of the pre-existing associations between action and object. During study phase, participants studied ordinary action-object pairs (e.g., “cut the watermelon”) (Experiment 3) and bizarre phrases (e.g., “plant the hammer”) (Experiment 4) in the SPT or VT condition. At test, they discriminated between intact, recombined, or new pairs. ERP results revealed that: 1) For ordinary phrases, early frontal old-new effects from comparisons of “intact vs. new” and “intact vs. recombined” were observed after SPT, but not after VT. In both encoding conditions, late parietal old-new effects from comparisons of “intact vs. new” were observed; additionally, following SPT, late ERPs to intact phrases were more positive than recombined phrases. 2) For bizarre phrases, early frontal old-new effect from comparison of “intact vs. new” was observed after both SPT and VT encoding. We observed late parietal old-new effect from comparison of “intact vs. new” after both SPT and VT encoding; additionally, only after VT, late ERPs to intact phrases were more positive than recombined phrases for bizarre phrases. These results demonstrate that enactment in SPT encourages both familiarity and recollection during associative recognition of action and object for ordinary phrases, which further enhances recognition performance. Bizarreness encourages both familiarity and recollection only after VT encoding, which led to bizarreness effect in the VT condition. Taken together, these findings reinforce the assumption that familiarity supports associative recognition during retrieval if a unitized representation was established during encoding.
学科领域	基础心理学
关键词	联结记忆整合熟悉性回想 Erp
学位类型	博士
语种	中文
学位专业	心理学
学位授予单位	中国科学院研究生院
学位授予地点	北京
文献类型	学位论文
条目标识符	http://ir.psych.ac.cn/handle/311026/19802
专题	认知与发展心理学研究室
作者单位	中国科学院心理研究所
推荐引用方式 GB/T 7714	赵敏芳. 整合编码对联结再认的影响及其认知神经机制[D]. 北京. 中国科学院研究生院,2015.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
赵敏芳-博士学位论文.pdf（2570KB）	学位论文		限制开放	CC BY-NC-SA	请求全文