Transferring the principles of the evolutionary learning theory to the world of feelings means, first, that feelings-i.e., appraisals-are considered nothing else than “emergent properties”  that the evolutionary selection acts upon.

Johnston considers feelings to be “proximate values” whose task has become to steer the “two-loop” (phylogenetic and ontogenetic) learning model by selecting the best-fitting hypothesis from several hypotheses generated by the sensory input. From the establishment of lasting behavior patterns up to more volatile memory storage, feelings decide what is kept and what is let go according to the reproduction advantage. The tuning spectrum of hedonic versus unhedonic does it. Johnston exemplifies this with many amusing observations and, in the end, with an artificial intelligence creature, a dog-like sniffer whose sensory and feeling systems decide whether it can sniff successfully to get hold of its prey or runs into starvation.

One most disenchanting for human narcissism but most plausible example of Darwinian learning theory applied to emotions is Johnston’s work on beauty and erotic partner selection, for which he won a scientific award. To prove that our feelings for beauty and sex appeal do not follow arbitrary individual patterns but the stern rules of the “net reproduction gain,” Johnston invited the world to choose the most attractive female and male faces of those he posted on the World Wide Web. Statistical evaluation demonstrated that from full lips (due to little fat cushions after estrogen rise in puberty) to the hip-waist ratio for the pelvis most likely to deliver without complications, every sign of sexual attraction is in fact related to the “net reproduction gain,” i.e., the likelihood to win the most fertile and successfully delivering woman. The attractiveness of male faces seemed to rest mainly with signs of health and physical strength. Furthermore, according to the selection experiment, our feelings are tuned to the age of highest fertility. Again, these cues cause exaggerated feelings, in the sense of a “discriminant amplifier,” to waste not even a minimal reproduction advantage.

The same learning principles underlie memory. Johnston claims that memory stores only experience related to events that matter for biological survival. Memories work by indexing episodes according to their hedonic tone-the basis for man’s reasoning on future events. The ability of “mind reading” is based on the comparison of currently perceived and formerly experienced social situations and their past and presumptive future social sequelae. It is obvious that this ability has an enormous impact on the competence to adjust to and influence other individuals’ feelings and hence to improve the reproduction rate.

Social feelings have played an increasing role in evolution for man’s reproduction advantage. Experimental games like “the prisoner’s dilemma” suggest that modesty and cooperation may do better than directly acting on the individual’s instant advantage. Johnston describes a set of feelings particularly adjusted to this somewhat paradoxical condition: an elaborate emotional monitoring system for the mutuality of affection, protection, and concern. If the reciprocity is broken by recklessness, a “pop out feeling” indicates the loss of the “investment”-hopefully not too late. In addition to anger, “pop out emotions” are fear and love-Johnston comments at length on their serendipity and their potential reproduction advantages. Emotional framing for decisions by a hedonic tone is also seen in the context of gene survival. The group size, for example, plays a crucial role in the amount of reciprocal interactions of social altruism offered by the group members to each other. Johnston quotes studies that relate the maximum group size allowing for altruism to the size of hunting bands in prehistoric times-the root of an adjustment that may no longer fit, like many others: the “naturalist’s fallacy.” The nepotistic, oligarchic regimes in some countries may resemble these mechanisms, however.

Johnston’s book is both amusing and plausible, but its basic theories are not novel. For the clinician, the harvest of the great factual knowledge in this book is not as rich as it is for the theoretician and biologist. The mechanisms of learning, the language of emotions in primary relationships, and the consequences of its failure or distortion (2) do not play major roles in the book, although it starts with a clinical case vignette of schizophrenia. The neurophysiology of brain systems that generate and process emotions is touched upon only very briefly. There is a vast body of research that strives to disentangle the universally stable basic patterns of emotions from their individually changeable parts (3–5). The microsocial and individual developments of distorted patterns of emotional language dealt with in trauma research could also build a bridge to clinical aspects, which obviously was not an intention of the author.

What lasts as a stimulating and original message of the book for the clinician is the generalization of principles of the Darwinian transgenerational “learning theory” concerning DNA under the selection process to other biological systems in general. Although DNA does not “learn” but just happens to change under the pressure of selection, the task of balancing the preservation of functions and their modification is the same in truly learning biological systems. Johnston’s application of the “Russian doll” learning processes to neuropsychology seems promising.

The strength of this book, namely its rigorous Darwinism and the relentless uncovering of what is seen as the ultimate purpose of feeling, nevertheless causes disillusionment and even disenchantment. Historical comparisons come to mind.

After the overstretched adoration of rationality in the late enlightenment toward the end of the eighteenth century, when reason was put on the altar, the Sturm and Drang period swept away logic and returned to the realm of feelings as a colorful but less calculable world in its own right. Johnston has not indulged in those worlds which are best represented by art and religion. Certainly he would find “net-reproductive gains” in societies with superior ethical and religious systems; and fine arts may be considered tentative appraisals that at some time will lead to “emergent properties,” but a little bit of illusion may also be necessary to survive.

References
1. Groß  M:  Hitzeschockprotein hilft der Evolution auf die Sprünge: Spektrum der Wissenschaft. Scientific American German Edition 1999, pp 12–16
2. Harris PL:  The child’s understanding of emotion:  developmental change and the family environment. J Child Psychol Psychiatry 1994; 35:3–28
3. Russell JA: Is there universal recognition of emotion from facial expression? a review of the cross-cultural studies. Psychol Bull 1994; 115:102–141
4. Ekman P: Strong evidence for universals in facial expressions: a reply to Russell’s mistaken critique. Psychol Bull 1994; 115: 268–287
5. Izard CE:  Innate and universal facial expressions:  evidence from developmental and cross-cultural research. Psychol Bull 1994; 115:288–391

PROF. DR. CHISTOPH MUNDT
Heidelberg, Germany
Why We Feel: The Science of Human Emotions, by Victor S. Johnston. Cambridge, Mass., Perseus Books, 1999, 224 pp., $26.00.