Introduction

It is an exciting time to be working on the problem of visual object recognition. The fields of visual neuroscience, cognitive science, and computer vision are rapidly converging; they are beginning to speak the same language and define the same set of problems. This chapter is not a scholarly review of any of those fields, and I apologize to the authors of a great deal of work that is relevant to this discussion but is not mentioned here. Instead, my chapter is an attempt to give our perspective on key questions at the intersections of these fields. I use the term “our” because the opinions expressed derive not only from myself but also from the hard work of my students, post-docs, and collaborators. Along the way, I briefly mention some recent research results relevant to this discussion, but mostly refer the reader to the original citations. Although I will not be providing the final answers here, I hope that our perspective among the others in this book will encourage agreement on what those answers might look like and a strategy for finding them.

The authors of the chapters in this book share a strong common interest. Loosely stated, we all want to understand how visual object recognition (categorization and identification) “works.” We have a goal in the mist – there is a mountain out there, and we would like to climb it together.