多视点学习条件下空间表征的朝向特异性

让被试在没有外界参照线索的条件下，从三个观察视点学习物体场景，首次以局部场景再认范式探讨了空间记忆中场景表征的朝向特异性。结果表明：（1）多视点学习条件下，场景空间表征依然是依赖于特定朝向的，不支持空间表征独立于朝向和空间表征依赖于多个观察朝向的观点。（2）场景空间表征的朝向依赖性不仅表现于方位判断任务，也同样表现于场景再认任务。（3）场景本身的内在结构对空间表征中参照系的选取有重要影响。

ORIENTATION SPECIFICITY OF SPATIAL REPRESENTATION IN MULTI-VIEW LEARNING

The experiment was designed to test the orientation specificity of the spatial representation under multi-view learning condition without any cues of external reference frame. Participants learned the layout in the cylinder room from three viewpoints (0, 90 and 225, respectively), and then were tested in another room using the partial scene recognition tasks. The results showed that all headings aligned with the symmetric axis of the layout were represented, supporting the orientation-specific theory, against the orientation-free spatial representation or multiple view-dependent visual memory representation. The intrinsic property of the layout (e.g. symmetry) might play an important role in selecting a spatial reference frame to represent the location of objects in the layout.