Skip to main content


A pair of neuroscientists finds that investigating emotions is easier done than said

The Neuroscience of Emotion: A New Synthesis

Ralph Adolphs and David J. Anderson
Princeton University Press
372 pp.
Purchase this item now

Ask a roomful of neuroscientists to define the term “emotion” and you will trigger a lively discussion. Some will argue that emotions involve conscious experiences that can be studied only in humans. Others might counter that insects and other invertebrates exhibit some of the emotion building blocks seen in mammals. Some will contend that different emotions correspond to anatomically distinct areas of the brain, whereas others argue that emotions are produced in a highly distributed manner. Still others will bring up the 19th-century psychologist William James’s argument that emotions are a consequence, not a cause, of behavior.

In The Neuroscience of Emotion, Ralph Adolphs and David J. Anderson argue that before we can study it, we must first define what we mean by “emotion.” Only then, they maintain, can we form appropriate and testable hypotheses.

Colleagues at Caltech, the authors bring different experimental backgrounds to the topic of emotion. Adolphs studies the neural basis of human social behavior. Anderson uses rodents and fruitflies as model organisms to investigate how internal states elicit emotional behaviors. Their book is less a catalog of recent neuroscientific discoveries and more a conceptual framework for investigating emotional behaviors both in humans and in other animals.

Adolphs and Anderson begin by contending that emotions are biological phenomena that cause behavioral and physiological changes in the brain and body and—in some species—subjective feelings. If emotions are a class of internal brain states expressed in measurable ways, they argue, we can study the neurobiological implementation of these states separately from subjective conscious feelings, meaning both humans and other animals are potential subjects. They go on to define, in detail, the basic properties of an emotion, including valence, scalability, persistence, automaticity, and generalization.

There is a tight logic running throughout The Neuroscience of Emotion that integrates theories of emotions, recent studies, and commonsense analogies. The authors confront, for example, the popular but erroneous view that fear is in the amygdala. This belief arose because activity in the amygdala can be correlated with feelings of fear in humans. But just as monitoring the speedometer of a car enables us to predict how fast it is moving without telling us anything about how the car works, increased activity in a particular brain region does not necessarily tell us anything about the causal mechanism of emotion.

Cutting-edge methods (such as optogenetics and pharmacogenetics) allow for the functional dissection of the neural circuits that cause emotional behavior in animals. Using these techniques, we have learned, for example, that the central nucleus of the amygdala contains as many as seven different neuronal subtypes that participate in phenomena including fear, learning, and appetitive behaviors. Conventional drugs and electrical stimulation tend to act indiscriminately on all cells, meaning it’s possible that such interventions might cancel out any potential effects a treatment might have had by stimulating or inhibiting conflicting cell types simultaneously.

Chapters on emotions in rodents highlight aspects of the emotional experience that we currently do not understand and provide specific suggestions for future research. How are some of the basic properties of emotions, such as persistence and generalizability, encoded, for example? And how do cortical projections from subcortical areas allow for interactions between emotions and cognitive processes?

Verbal reports by human subjects can conflate actual emotional states and the conscious experience of emotions. To combat this, the authors suggest designing experiments to emphasize one component or the other. Subjects could be presented with a spider, for example, to incite fear, or—if the subjects’ perceptions of the experience are desired—they could be asked to recollect emotional experiences from their lives.

The book finishes with a critique of current theories of emotion and suggestions for future research. The authors ask, for example, if there might be emotions that are unique to a certain species—and even if there might be emotions unique to individuals within a species. And will we one day understand emotions well enough to be able to build robots that have an internal emotional life?

Adolphs and Anderson openly acknowledge that they do not provide a comprehensive theory of emotion. Indeed, despite the existence of substantial research on the subject, we are left with the impression that we actually know remarkably little about emotions at present. But their enthusiasm for the topic is genuine and makes The Neuroscience of Emotion compelling and engaging.

About the author

The reviewer is at the Department of Biology, Barnard College, New York, NY 10027, USA.