Npgrj_NN_1519 1..3

While it is often assumed that objects can be recognized irrespective of where they fall on the retina, little is known about the mechanisms underlying this ability. By exposing human subjects to an altered world where some objects systematically changed identity during the transient blindness that accompanies eye movements, we induced predictable object confusions across retinal positions, effectively ‘breaking’ position invariance. Thus, position invariance is not a rigid property of vision but is constantly adapting to the statistics of the environment. Any given object can cast an essentially infinite number of different images on the retina, owing to variations in position, scale, view, lighting and a host of other factors. Nonetheless, humans effortlessly recognize familiar objects in a manner that is largely invariant to these transformations. The ability to identify objects in spite of these transforms is central to human visual object recognition, yet the neural mechanisms that achieve this feat are poorly understood, and transform-tolerant recognition remains a major stumbling block in the development of artificial vision systems. Even for variations in the position of an image on the retina, arguably the simplest transform that the visual system must discount, little is known about how invariance is achieved. Several authors have proposed that one solution to the invariance problem is to learn representations through experience with the spatiotemporal statistics of the natural visual world1–4. Visual features that covary across short time intervals are, on average, more likely to