Abstract
Objects in our environment are subject to manifold transformations, from simple rigid changes like rotation or scaling, to complex processes like twisting, bending or biological growth. Although many transformations have drastic effects on an object’s physical appearance, we are often able to identify stable objects across such changes, and in many cases have strong subjective impressions of the type of transformation (e.g., in case of a crumpled can). This suggests the brain is equipped with sophisticated mechanisms for inferring both invariant object representations across transformations, and objects’ causal history. Despite this, relatively little is known about how shape, spatial relationships and transformations themselves are inferred when objects undergo change. We used a simple, intuitive task to measure participants’ shape representations across spatial transformation. We presented an untransformed 2D shape (‘base shape’) on the left side of the screen and its transformed counterpart on the right. On each trial, a dot was superimposed at a given location on the base shape and the participant’s task was to place a dot at the corresponding location on the right side of the screen. Across trials many locations were probed. The resulting ‘perceptual correspondence maps’ allowed us to measure how well participants perceive shape across transformations; the extent to which they could infer the type of transformation; as well as its effects on the representation of space. We find that shape representations are remarkably robust against spatial transformations, especially rigid transformations. Performance is affected by the type and magnitude of transformation, as well as contour saliency. We also find that the representation of space within and around a shape is transformed in line with the shape transformation, as if shape features establish local coordinate frames. Together, these findings suggest sophisticated mechanisms for the inference of shape, space and correspondence across transformations.
Meeting abstract presented at VSS 2015