Seeing Spatial Form

  • Michael R. M. Jenkin &
  • Laurence R. Harris
Oxford University Press: 2005. 449 pp. $38.99 | ISBN: 0-195-17288-4

Seeing Spatial Form covers an apparently eclectic mix of topics in visual perception, from 'pictorial relief' and 'vision in flying' to 'neuronal damage' and 'dopamine receptors'. In the introduction, the editors, Michael Jenkin and Laurence Harris, relate these and other seemingly disparate themes to the general topic of 'spatial form' — the detection of (often hidden) shapes in the visual world. In my view, detecting objects' shapes in the enormous complexity of retinal signals is one of the most elaborate feats achieved by the human brain. One excellent example of this feat is the detection of camouflaged items, which is proving a tough problem for artificial intelligence.

The book does not provide a systematic treatment of the perception of spatial form in vision, so how did the editors decide what to include? They were guided by three principles: the choice of first-rate contributors; the adoption of a systems-analytical and model-driven approach; and the desire to select topics close to David Martin Regan's heart. This is because the book is a Festschrift in honour of Regan's 68th birthday. So, instead of an integrated, textbook-like coverage of the perception of spatial form, the book provides rather specialized, mostly very well written, chapters that are more or less closely related to the topic, by some of the most prominent people in the field.

The individual chapters provide excellent reviews on state-of-the-art research in a wide range of topics, all related to Regan's interests. And wide they are, at a time when many researchers devote their life to a single topic, or even to a single technique. Regan is more of a Renaissance man, not only being interested in and using psychophysical methods, both in normal observers and patients suffering from different deficits of visual perception, but also sum-potential recordings, 'active' vision with eye movements, functional magnetic resonance imaging, neuropharmacology and developmental methods. Accordingly, the book covers a wide range of topics, and a wide range of methods for studying them.

It is not possible to mention all these topics in a short review. Suffice it to say that Regan's friends and colleagues produced an excellent Festschrift, with well structured, well formulated and well argued synopses on many facets of visual perception, not just spatial form. Even a reader seasoned in vision research will find not just new pieces of knowledge, but plenty of new insights — this, in my opinion, makes it an ideal book to choose for those times when you're stranded on a desert island. It will tell you about all those topics in vision research that you always wanted to discover but did not have the time to read about.

Let me give you four examples. You probably know, or would have guessed, that attention plays an important role in programming saccades, or rapid eye movements. But were you aware that planning and executing them 'costs' cognitive resources, and that a batsman cannot follow an approaching ball with his eyes but needs to estimate its time of arrival with an accuracy better than one-hundredth of a second? Second, it turns out that the processing of motion-defined form may be defective in patients even though simple motion detection is intact. Third, although retinal images are flat and the detection of surfaces represents an early stage of visual computation, perception is always inherently three-dimensional, even in the pictorial space of photographs. Finally, great progress has been made in understanding the neural basis of 'form vision'. For example, three-dimensional shape can be perceived from as few as two frames of a motion sequence.

If, once you have read the book, you want to know even more, no problem: there are selected references after each chapter, and a CD-ROM with additional material to keep you busy for quite a while.