The aims of this book are grand indeed. Rolls addresses such questions as, What are emotions? Why do we have emotions? What are the rules by which emotion operates? What are the brain mechanisms of emotion, and how can disorders of emotion be understood? Why does it feel like something to have an emotion? Why is the brain built to have reward, and punishment, systems? How does the brain produce behavior by using reward, and punishment, mechanisms? And in fact he does succeed remarkably well in presenting a synthesis of experimental data to elucidate many of these questions.

Rolls is an Oxford researcher and a leader in the field of neuropsychological research on emotion and related phenomena, to which he has contributed for the past 30 years. This work serves in part as a review of his remarkable body of work, which ranges from human imaging with emotional stimuli to mapping olfactory and taste cortex in primates, from the neurochemistry of reward to neural network modeling, from emotional changes with frontal lobe damage to the neurophysiology of the visual system and cellular mechanisms of thirst—and much more.

The book opens with definitions of basic emotion-related terms and concepts. It has chapters on the neural mechanisms of hunger and thirst, presents a theory about the nature and purpose of emotion, and reviews the anatomy and neurochemistry of emotion pathways and areas that evoke self-stimulation. It then explores the neural basis of sexual reward (about which more later) and presents a theory of consciousness and its relationship to emotion. A last chapter introduces Rolls' and Treves' view on neural network emotion-related learning, which is detailed in a companion volume (Rolls & Treves, 1998).

This synthesis of emotion research is heavy with Rolls' own work but also integrates findings of many other researchers. Most of the chapters are both detailed and readable, rich in experimental data but using these data to make clear points, so that the reader can easily follow the thread of his arguments and can choose the degree of experimental detail to read. The book works very well as an integrated whole; arguments flow from earlier points and there is continuity in the viewpoint throughout. A particularly appealing feature of the book is that it is generally written at a high enough level of detail and sophistication to satisfy the professional researcher, and yet Rolls assumes little background for his readers and defines even simple terms and concepts as he goes.

Emotion is viewed by Rolls as an adaptive system that helps to map certain classes of stimuli (rewarding and punishing) onto action systems, and to prioritize selections among actions, depending on the magnitude of the available reinforcers and the motivational state of the organism. Rolls details the way in which adaptive (evolutionary) advantage is gained by such phenomena as attaching reinforcing value to a stimulus late in its sensory processing, and outlines such specifics as how food-specific satiety is advantageous and how it is achieved in the brain. I found particularly interesting the extended discussion of the neural representation of primary versus secondary reinforcers, with the latter, surprisingly, proving simpler to specify. Rolls compares his conception of emotion to other conceptions and theories, finding it fundamentally different from the James-Lange conception and that of Damasio (which he regards as neo-James-Langian) but finding his consistent with that of LeDoux. Rolls also presents a novel classification of emotions, based on omission/termination versus presentation of positive versus negative stimuli so that, for example, rage or grief is associated with termination of a positive stimulus, relief with termination of a negative stimulus, and fear with presentation of a negative stimulus.

In a book that attempts to address such a multitude of major questions, it is perhaps inevitable that some areas are less well covered than others. Human individual differences such as personality and psychopathology and complex human emotions such as guilt are not covered in the same detail as other topics and these sections will not be as useful as others.

Rolls also devotes considerable space to discussing the nature of consciousness and the relationship of subjective phenomena (sensory and emotional qualia or feelings) to consciousness. Consciousness is viewed by Rolls as largely dependent on language; in fact, a conscious system is defined as “the ability to have higher-order thoughts” (p. 250); that is, thoughts about thoughts. Furthermore, a “first-order language processor (that might be perfectly competent at language) would not be conscious, in that it could not think about its own or others' thoughts.” This seems to imply that I am conscious when I think about my own thoughts but not when I think about trees, which doesn't seem right, but I leave this to the reader to consider. The ways Rolls deals with qualia does seem to be unconvincing. He does not directly address the question of how subjective phenomena can arise from brain activity, but claims that it is “more parsimonious to hold that we would be conscious of such sensory, emotional, and motivational qualia because they would be being used … in this type of (linguistically based) higher-order thought processing” (p. 251). This seems to suggest that when my cat stretches out on his back to be petted, working his paws and gazing at me through half-closed lids, he is not feeling pleasure, and that when he gets a splinter in his paw, he does not feel pain as I do when I get a splinter in my paw. This contradicts the basic notion of emotion as a kind of motivational processing, or it disconnects emotion (which must exist in cats by Rolls' conception of emotion) from ‘qualia’ in an unnecessary way. But perhaps the subjective experiences of nonlinguistic creatures is fundamentally unknowable.

The speculative chapter on sexual behavior seems out of place in an otherwise scholarly work. Rolls seems to have relinquished his objective, empirical stance in this chapter. It is based almost entirely on a single book by other authors who have a sociobiological perspective on sexual behavior. Among other notions, the views are espoused that there are genes for infidelity which in women are linked to genes for attractiveness (this being defined as being young-looking and thin, thinness being attractive because it is a sign of not being pregnant), and that the purpose of the penile groove and of pelvic thrusting is to vacuum out sperm from other males which may have been deposited in the male's absence. There is no hesitation about generalizing from the sexual behavior of birds or lions to that of humans. No mention is made of cultural influences on sexual behavior or of cultural differences in standards of attractiveness; all is interpreted in simple sociobiological terms. I found this chapter to be uncritical, implausible, and offensive.

Two other features that detracted in a minor way from the enjoyableness of the book were the organization of the chapters (it was not clear why the author separated highly specific chapters on thirst and hunger with chapters on the nature and neural bases of emotion) and the index, which was so skimpy as not to be useful.

Despite the theoretical points with which one could take issue, and the chapter on sexual behavior, which is worth reading only for its shock value, the book as a whole presents a clear and plausible point of view, backed up by a highly specific and comprehensive review of experimental work. It is highly recommended for anyone interested in the neural basis of emotion, and is essential for students of this field.