Suppr超能文献

Integration of foveal orientation signals: distinct local and long-range spatial domains.

作者信息

Brincat S L, Westheimer G

机构信息

Division of Neurobiology, University of California, Berkeley, California 94720, USA.

出版信息

J Neurophysiol. 2000 Apr;83(4):1900-11. doi: 10.1152/jn.2000.83.4.1900.

Abstract

Human observers can discriminate the orientation of a stimulus configuration composed of a pair of collinear visual patterns much better than that of a single component pattern alone. Previous investigations of this type of orientation signal integration and of other similar visual integrative functions have shown that, for closely spaced elements, there is integration only for stimuli with the same contrast polarity (i.e., both lighter or both darker than the background) but, at greater separations, integration is independent of contrast polarity. Is this effect specific to differences in contrast polarity, which is known to be an important parameter in the organization of the visual system, or might there be a cluster of other stimulus dimensions that show similar effects, indicating a more widespread distinction between the processes limiting integration at local and long-range spatial scales? Here, we report a similar distance dependence for orientation signal integration across stimulus differences in binocular disparity, direction of motion, and direction of figure-ground assignment. We also demonstrate that the selectivity found at short separations cannot be explained only by "end-cuts," the small borders created at the junction of abutting contrasting patterns. These findings imply the existence of two distinct spatial domains for the integration of foveal orientation information: a local zone in which integration is highly selective for a number of salient stimulus parameters and a long-range domain in which integration is relatively unselective and only requires that patterns be roughly collinear.

摘要

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