Lecture Notes in Computer Science, 2000, Volume 1811/2000, 395-404, DOI: 10.1007/3-540-45482-9_64

Temporal Structure in the Input to Vision Can Promote Spatial Grouping

Randolph Blake and Sang-Hun Lee

View Related Documents

Abstract

This familiar “Mother Goose” nursery rhyme captures the essence of what has become a central problem in the field of visual science: how can an aggregate of individuals working together reassemble a complex object that has been broken into countless parts? In the case of vision, the “horses and men” comprise the many millions of brain neurons devoted to the analysis of visual information, neurons distributed over multiple areas of the brain (Van Essen et al. 1992; Logothetis, 1998). And, in the case of vision, the source to be reassembled corresponds to the panorama of objects and events we experience upon looking around our visual world. In contemporary parlance, this “reassembly” process has been dubbed the binding problem, and it has become a major focus of interest in neuroscience (Gray, 1999) and cognitive science (Treisman 1999).

Fulltext Preview

Image of the first page of the fulltext document