Colour Perception: Mind and the Physical World

R. Mausfeld & D. Heyer (eds.)

“The colours red, blue and green are real. The colour yellow is a mystical experience shared by everybody.” (Stoppard, 1967).

Colour is a vivid visual experience we (almost) all share. But what exactly is it that we are experiencing when we see red (or yellow)? When two people agree that something looks yellow, how can they know if they are actually sharing an experience or if they have simply learned a common association. For example, we all grow up learning that bananas are yellow, so we know that the colour we experience when looking at a banana must be yellow. But is my experience of yellow the same as yours, and is it meaningful even to ask this question? In some ways our understanding of colour vision has advanced enormously in recent years. Thanks to the advent of single-cell electrophysiology we have learnt a lot about the neural basis of the processing of chromatic information. However, on a philosophical level some fundamental questions are still unresolved. For example, what is the proper metaphysical status of colour? Is it the property of an object or is it a purely subjective experience?

Colour Perception: Mind and the Physical World is an eclectic and rather esoteric collection of chapters resulting from a year spent by its authors at the Zentrum für interdisziplinäre Forschung of the University of Bielefeld. The format of the book is of sixteen chapters each followed by a commentary or commentaries, usually from authors of other chapters. The subject matter ranges from computational accounts of issues such as colour constancy to philosophical discussions of the realism, or otherwise, of colour.

The essential problem of colour constancy, the principal focus of Chapters 7-11, is how our visual systems factor the wavelength composition of light incident on our retinas into a multiplicative combination of the spectrum of the illumination and the surface reflectance properties of objects. The solution of this inverse problem would allow the observer to discount the effects of illumination and thus recover veridically the surface reflectances (colours) of objects, making colour a valuable aid in object recognition. According to this view, colour is a property of objects and our perception of colour depends upon neural mechanisms dedicated to solution of this problem. Thus, implicit in computational accounts of colour constancy is a particular perspective on the ontological status of colour.

Of the chapters in this volume, one stands out from the rest as being a result of genuine interdisciplinary collaboration resulting from dialogues in Bielefeld. Don MacLeod is visual psychophysicist from UC San Diego while Tassilo von der Twer is a mathematician from Wuppertal. Together they determine the optimal form of the neuronal response function for coding a distribution of chromatic inputs under a range of different assumptions about the statistics of neuronal noise. They show that when a single stimulus dimension can be represented by more than one neuron then an opponent split-range code enhances coding efficiency. For example, the existence of separate ‘on’ and ‘off’ channels in the coding of luminance allows each neuron to devote its full dynamic range to coding only one half of the input range. These results have implications far beyond the perception of colour. As the commentary from Michael Webster puts it, “the power of the approaches they develop are hard to over emphasize”.

Webster himself has contributed a very strong chapter on colour perception and adaptation, though much of the content is based on an earlier review paper (Webster, 1996). In it he looks mechanistic and functional basis of light and contrast adaptation, adaptation of post-receptoral chromatic channels, and adaptation and the statistics of natural images. The chapter by Michael D’Zmura also provides a concise review of recent work on the processing of chromatic information in the human visual system, in particular his own elegant noise-masking work measuring the spectral properties of chromatic detection mechanisms (D’Zmura & Knoblauch, 1998). These are chapters I certainly envisage myself referring to in the future.

Parts of this book will be of interest to researchers from a range of disciplines who are interested in colour vision, but there is little to attract a wider non-specialist audience. Comparisons with Color vision: from genes to perception (Gegenfurtner & Sharpe, 2001) are inevitable. The latter is strong on the neurobiology of vision, while the present volume is pitched at more theoretical and philosophical level. I must admit that my attitude to much of the philosophical discussion in this book is summed up in the title of Paul Whittle’s commentary on Gary Hatfield’s chapter. Hatfield writes on “Objectivity and subjectivity revisited: colour as a psychobiological property”. Whittle entitles his commentary “Why is this game still being played?”. One thing is for sure: despite a long history of study from a range of perspectives and notwithstanding significant theoretical and neurobiological advances, the experience of the colour yellow has lost none of its mystery.

References
D’Zmura M, Knoblauch K, 1998, “Spectral bandwidths for the detection of color” Vision Research 38 3117-3128.
Gegenfurtner K R, Sharpe L T (eds.), 2001 Color vision: from genes to perception (Cambridge University Press, Cambridge).
Stoppard T, 1967 Rosencrantz and Guildenstern are dead (Faber, London).
Webster M A, 1996, “Human color perception and its adaptation” Network: Computation in Neural Systems 7 587-634