A wide range of empirical factors influences the perception of angles. In this scene, the subtenses of the four angular objects are identical, each measuring approximately 90° on the printed page.

In viewing this stimulus, people invariably perceive the square on the brighter surround (left) to be darker than the square on the darker surround (right).

As can be seen by masking out the context, the gray diamonds on the upper and lower front face of the cubes are identical, but look very different in the unmasked scene. When the information in a scene indicates that, on statistical grounds, two targets are differently reflective surfaces under different illuminants, they look differently bright (or light).

When diverse statistical information about the sources of a visual stimulus is made mutually consistent, two equiluminant surfaces can be made to appear remarkably different. In this case, the empirical information indicates a high probability that the two surfaces are differently reflective materials under different levels of illumination, making the same patch in one part of the scene look very dark gray, and in another part of the scene very light gray. (Image by Beau Lotto - lottolab.org)

Just as achromatic patches can be made to look differently bright by empirical information, color patches can be dramatically affected by empirical cues about the amount of illumination they are likely to be under. In this example, the effect is primarily on the color brightness of the relevant tiles rather than on sensations of hue or saturation as such.

In viewing this scene, people invariably perceive the surface of the top block to be darker than the bottom block. Click on the play button to see that these territories are in fact identical. Covering the center section, including the shaded gradients, shows that the source of the difference is this component of the scene.

It has been known for more than 150 years that spectrally identical patches can look differently colored when placed in spectrally different surrounds. The two central targets here are identical, as can be seen by masking out the surround.

Statistical information about probable sources strongly affects color, as well as lightness and/or brightness. Here, two spectrally identical patches (indicated with a dot) appear gray-green when viewed in isolation, but look reddish and bluish, respectively, when viewed in the context of empirical information that makes different surfaces under different chromatic illuminants a highly likely source of the stimulus.

In this demonstration, physically identical patches that appear neutral gray when viewed in isolation can be made to look either yellow or blue, depending on the context in which they appear. Although quite unbelievable at first glance, the reality of the demonstration can be confirmed by masking out the context.

In this remarkable example, differently colored patches can be made to look more or less the same color (red in this case) by empirical information, which changes the probable spectral sources of the differently colored tiles. These demonstrations show that color contrast and constancy are both manifestations of the same empirically determined visual strategy for seeing color.

Neural Activity and the Growth of the Brain
Purves, D
Cambridge University Press, 1994

This book is now out of print, but the short lecture series may still be of interest. Brain growth is considered at a macroscopic level by examining brain maps and their molecular substructure, and at a cellular level by investigating the neuronal interactions that influence the formation and maintenance of these structures. The ways that experience influences the maturation of the brain at both macroscopic and microscopic levels are described, and some of the conventional wisdom about these issues re-examined. Anyone interested in how the brain stores information may find the lectures of interest.

Body and Brain: A trophic theory of neural connections
Purves, D
Harvard University Press, 1988

This book has recently gone out of print. It may nonetheless be of interest to many readers because of its broad, historical coverage of the relationship between the growth of the body and the complementary growth and organization of the brain. A particular focus is the development of neurons and their synaptic connections, and the mediation of these interactions by trophic agents. The link between somatic targets and their innervation is considered using simple systems such as the neuromuscular junction and the innervation of autonomic ganglion cells as examples of processes that are presumably characteristic of interactions throughout the nervous system.

Table of Contents
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
Glossary
Bibliography
Index