Matching multicomponent objects from different viewpoints: mental rotation as normalization?

Novel multicomponent objects were created, and 3 distractors were created for each object by changing the relations between the parts of the object. In a set of 5 experiments, target objects were presented as a motion sequence of multiple views or as a single view. Participants were asked to determine whether an image of an object, viewed from another viewpoint, was the same as the target object. The axis of rotation was aligned with one of the environmental axes or with the main axis of the object. The effects of viewpoint on performance imply that the matching of objects is viewpoint dependent and requires a process of normalization to undo the differences between the perceptual description and the stored object descriptions. The lack of a systematic effect of the axis of rotation, however, suggests that this normalization is best understood as not involving a 3-D transformation of stored 3-D object models.