Commentary on “Integrating the Psychoanalytic and Neurobiological Views of Panic Disorder”

Alexander, Feigelson, & Gorman embark on an intellectually precarious journey in attempting to unite vastly dissimilar realms of inquiry that are both temporally and conceptually distant. The paper describing this effort demonstrates an impressive breadth of knowledge and an underlying generosity of spirit; however, the enterprise is too ambitious to succeed. The authors are trying to link nineteenth-century Freudian ideas to developing twenty-first-century neurobiological models of panic, using results of twentieth-century fear-conditioning experiments as a bridge. Neither psychoanalysis nor neuroscience is sufficiently well specified for such an enterprise. Conditioning experiments have provided interesting data, but a great deal is still unknown about fear conditioning, especially in humans. To provide just a few examples, there is some evidence that conditioned learning in infants follows different principles from those in adults. When and how do such differences occur across the lifespan? How does fear learning interact with specific contextual variables, such as the state of the attachment system? How does the attachment system work? There are many such unresolved questions, so it is premature to create an integrated model using fear conditioning as its bedrock.

While psychoanalysts theorize about normal psychological functioning, they do not engage in intensive observation of normal behavior. Instead, this group concerns itself with a theory of the mind developed from observations in the consulting-room. In the case of this paper, the observations of Freud himself play a central role, mostly unmodified by late twentieth-century work. Neuroscientists work in a very different way. Investigation pertains to studying properties of specific brain functions. Research is carried out primarily by nonclinical scientists, and investigators use accumulating answers to specific questions in order to develop provisional mo