A computational model of spatial imagery based on object-centered scene representation.
A computational model of spatial imagery based on object-centered scene representation.
复制标题
基于以对象为中心的场景表示的空间图像计算模型。
DOI：
10.1007/978-3-642-10677-4_6
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
N
中科院分区：
文献类型：
--
作者：
Sato;N
海马维持物体-地点联想的记忆，也产生新视角下的场景期待能力。为了实现这种能力，一个物体的距离和方向应该被集成为一个allocentric空间记忆，但它的神经动力学还没有讨论。在本文中，我们提出了一个以对象为中心的场景表示作为海马体中对象位置记忆的一个组成部分。通过使用该表示，可以计算物体在图像视点处的距离和方向，作为以物体为中心的场景和图像场景之间的差异。此外，以物体为中心的场景适用于新位置的物体-位置记忆提取。结果表明，以物体为中心的场景表征在自我中心和他中心的空间表征之间起中介作用，并支持自愿视角下的空间意象。
The hippocampus maintains the memory of object–place associations and also produces the ability of a scene expectation at a novel viewpoint. To implement such capabilities, an objects’ distances and directions should be integrated as an allocentric space memory, while its neural dynamics have not been discussed. In this paper, we propose an object–centered scene representation as a component on object–place memory in the hippocampus. By using the representation, an object’s distance and direction at the imagery viewpoint can be calculated as the difference between object–centered and imagery scenes. Moreover, the object–centered scene is applicable for the object–place memory retrieval at the novel location. It is suggested that the object-centered scene representation mediates between egocentric and allocentric space representation and supports the spatial imagery at the voluntary viewpoint.