John Nash said of his delusions “Its kind of like a dream. In a dream it’s typical not to be rational”. His words express a common idea that dreams and delusions are both states characterised by absent or compromised “reality testing” (Bentall, 2004). This intuition is now supported by a research program which compares neural activity, cognitive processing and phenomenology in dreaming and waking (A. Hobson, 1999 and J. Hobson, 1999).
This program depends on the fact that the basic cognitive architecture of the mind does not change during dreams. What changes in dreams are the patterns of neural activation, driven by brainstem neuroregulation, which drive processing in this cognitive architecture. Some cognitive processes, such as late states of perceptual processing, remain relatively intact in dreams, producing the characteristic stream of imagery. Others, such as logical argumentation, volitional control and planning are absent or reduced. Thus in REM dreams we experience cognitive fragments such as images and sensations juxtaposed incongruously in vignettes and scenarios rather than coherently organised in narratives or explanations ( Braun et al., 1997, 1998; Desseilles et al., 2011; Domhoff, 2003; A. Hobson, 1999, J. Hobson, 1999, Manni, 2005, Roehrenbach and Landis, 1995, Schwartz and Maquet, 2002 and Solms, 2007).
This phenomenon reflects an important cognitive distinction between feature binding (the construction of a integrated perceptual or quasi-perceptual representation, Ashby et al., 1996) and contextual binding (the ability to organise such representations into a coherent metacognitive structure such as a narrative or theoretical explanation. See Revonsuo and Tarkko (2002)). Feature binding is a relatively modular process implemented in localised neural circuits whose processing properties are fairly rigid (Ashby et al., 1996; Coltheart, 1999 and David, 1994). Context binding is a more flexible cognitive process dependent on the maintenance of coordinated activity in widely distributed circuitry. Consequently context binding is more vulnerable to disintegration when resources required to stabilise and synchronise the necessary distributed activation patterns are withdrawn (Cleeremans, 2003 and Maia and Cleeremans, 2005).
In this paper I apply some of these ideas to help explain the phenomenon of delusional misidentification. In these delusions people report seeing “doubles”, imposters, people in disguise, people changing appearance and identity (Breen et al., 2000, Ellis, 1991, Ellis, Luautè, et al., 1994, Ellis, Whitley, et al., 1994, Luaute and Bidault, 1994 and Tzavaras et al., 1986). The phenomenology of these delusions can be explained in terms of abnormalities of feature binding, combined with abnormalities in context binding. The abnormality of feature binding produces a representation in which elements normally bound together such as face, name, autonomic response to a familiar person, and identity may dissociate (Noreika et al., 2010; Kahn and Hobson, 2003, Kahn and Hobson, 2005 and Kahn et al., 2002). For example if a representation of a feature such as the face of a spouse is not bound to information which drives autonomic response to a familiar person the result will be an incongruity: the patient sees a person who appears familiar but the patient does not have a characteristic autonomic reaction to that person. This incongruity, produced by the failure to bind the familiar face to a signal initiating autonomic response, then leads to delusional context binding: “my wife has been replaced by an imposter”. The delusion binds the incongruous representation into a larger explanatory context which accounts for the fact that perception of the familiar person does not produce the “familiar” autonomic response ( Ellis and de Pauw, 1994, Ellis and Lewis, 2001, Spier, 1992 and Young et al., 1993).
