http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Integration in working memory: Cognitive behavioral and human brain imaging studies
Wheeler, Mary Elizabeth Princeton University 2001 해외박사(DDOD)
소속기관이 구독 중이 아닌 경우 오후 4시부터 익일 오전 9시까지 원문보기가 가능합니다.
The human brain processes information in many parallel streams, but mental life is experienced as a single stream of complex events ranging from episodic memories to object perceptions. Integration or binding involves the reorganization of distributed information to create complex but unified representations. The mechanisms that create and maintain these integrated associations in working memory and attention are essential for conscious thought. Previous work investigated integration in visual working memory (VWM) using simple colored shapes and a change detection paradigm (Wheeler & Treisman, 2001). Results showed that object features from different dimensions could be remembered in parallel. However remembering the binding between two features did not happen automatically and required focused attention. In a proposed model working memory capacity is limited both by the capacity of separate feature stores and by demands on attention networks that integrate distributed information into complex unified thought objects. The current studies further develop this model and investigate the underlying neural basis for integration. Experiments 1–3 show that binding in VWM depends on spatial information, just as does binding in visual perception. Changing the location of objects interferes with memory for the binding but does not interfere with memory for the individual features of those objects, while an irrelevant change of object shape does not interfere. To verify that the paradigm captures memory demands Experiments 4 and 5 vary the length of the memory delay and show that performance declines as delay increases. Memory for binding information declines more quickly with increasing delay and is more vulnerable to interference than feature information but does not show relatively worse perceptual or encoding limits. Experiment 6 uses functional magnetic resonance imaging (fMRI) and shows that brain areas involved in visual attention are also involved in maintaining the binding between features in VWM. Experiment 7 extends beyond vision and uses fMRI to study integration of faces and names. A new area in anterior frontal cortex involved in integration across modalities is characterized. A model of integration in VWM is elaborated and implications of the brain imaging results for the underlying neural mechanisms of the model are discussed.