The Substrate That Dreams Are Made On: An Evaluation of Current Neurobiological Theories of Dreaming

Abstract

Theories regarding ambiguous consciousness states, such as dreaming, often attract questions regarding the scientific status of the experiments on which they are based. Rarely, however is the scientific status of the theory itself scrutinized. There are basic principles of theory construction that can provide a framework for evaluating current neurobiological theories of dreaming. This chapter places particular emphasis on the activation-synthesis (AS) and activation, input and modulation (AIM) models, developed by Hobson and colleagues over the past three decades (Hobson and McCarley, Am J Pschiatry 134:1335–1348, 1977; Hobson et al., Behav Brain Sci 23:793–842, 2000). This theory set was chosen as it can be considered one of the most widely cited and publicized neurobiological theories of dreaming today. Our aim in this chapter is not to criticize this work specifically, but to draw attention to the problems of theory development presently inherent in all dream research. The nature of the assumptions which underlie dream theories, the logic of argument, and the validity of methodologies used in collecting the empirical evidence, are scrutinized according to principles of theory construction and validity. We argue that modern theories of dreaming, whilst evolving ad hoc modifications in the face of new and sometimes anomalous evidence, are essentially unfalsifiable, and by definition do not qualify as scientific theories. However, as methodologies and technologies improve, particularly in the areas of sleep stage recording and neuroimaging, a new paradigm for the neurobiology of dreaming may emerge.