Suppr超能文献

视觉层级对全局和局部特征的时间分离

Temporal dissociation of global and local features by hierarchy of vision.

作者信息

Ishizu Tomohiro, Ayabe Tomoaki, Kojima Shozo

机构信息

Department of Psychology, Graduate School of Human Relations, Keio University, Mita, Minato-ku, Tokyo, Japan.

出版信息

Int J Neurosci. 2009;119(3):373-83. doi: 10.1080/00207450802540524.

Abstract

Most objects in our environment are organized hierarchically with a global whole embedding its local parts, but the way we recognize these features remains unclear. Using a visual masking paradigm, we examined the temporal dissociation between global and local feature as proposed in Reverse Hierarchy Theory, RHT (Ahissar & Hochstein, 2000), where global and local information are assumed to be processed, respectively, by feed-forward and feedback systems. We found that in a long Stimulus Onset Asynchrony (SOA) condition, both global and local information were recognized adequately. However, in a short SOA condition, global information was recognized correctly while local recognition was critically disrupted. Consistent with RHT, it is suggested that local information is processed in a feedback system; this processing is then interrupted by the mask stimulus presented later at the primary visual area. Global information, by contrast, is transferred via a feed-forward system, and is not disrupted by the mask.

摘要

我们环境中的大多数物体都是层次化组织的,整体包含局部,但我们识别这些特征的方式仍不清楚。我们采用视觉掩蔽范式,检验了反向层次理论(RHT,Ahissar & Hochstein,2000)中提出的全局和局部特征之间的时间分离,该理论假定全局和局部信息分别由前馈和反馈系统处理。我们发现,在长刺激起始异步(SOA)条件下,全局和局部信息都能得到充分识别。然而,在短SOA条件下,全局信息能被正确识别,而局部识别则受到严重干扰。与RHT一致,这表明局部信息是在反馈系统中处理的;随后,该处理会被初级视觉区域稍后呈现的掩蔽刺激打断。相比之下,全局信息通过前馈系统传递,不会被掩蔽所干扰。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验