In Susan Hurley's shared circuits model (SCM), we see a model that is useful in its intended purposes, though somewhat incomplete both in application and in description, specifically at the neural level. Hurley predicts that disruptions within layer 4 could lead to cognitive deficits, specifically in the capacities for instrumental deliberation and for understanding others' actions. Although her model is successful in predicting some cognitive dysfunction, she does not note relevant problems that can arise when a person fails to adequately differentiate between self and other. Nor does she explore fully the useful role that SCM's layer 2, or, more specifically, the feed-forward model that underlies it, may play both in explicating the self/other delineation and in encouraging the search for the neural correlates involved in the dynamic.

For example, Feinberg and Keenan (Reference Feinberg and Keenan2005) found that the parietal lobe, as well as the right frontal regions, are critical in disorders of patients suffering from a loss of self. In delusional misidentification syndrome (DMS), patients consistently and adamantly misidentify persons, places, objects, or events. In delusional reduplication syndrome (DRS), patients reduplicate or double the misidentified entity. Data from such patients revealed that the greatest number of cases is associated with right frontal and right parietal damage. Disturbances of the self/other relationship are not unique to DMS/DRS, however, but occur in a number of neurological disorders. For example, patients with delusions of control confuse self-produced and externally produced actions and sensations; such delusions of alien control are hallmark symptoms of schizophrenia. Hyperactivity of the parietal cortex and cerebellum occurs in such patients, suggesting that over-activation of a cerebellar-parietal network during self-generated actions is associated with the misattribution of those actions to an alien, external source (Blakemore Reference Blakemore2003). Another instance of deficits in the ability to differentiate self from other that is unexamined yet germane to the target article concerns auditory verbal hallucinations (AVHs) and inner speech. Individuals who experience AVHs report hearing speech in the absence of any external stimulation; that is, they hear in their head a voice or voices other than their own.

Although AVHs are classified as a first-rank symptom of schizophrenia, they may not necessarily signify pathology and may best be understood within the wider context of the development of inner speech (Jones & Fernyhough Reference Jones and Fernyhough2007a; Reference Jones and Fernyhough2007b). In Vygotsky's theory of the social origins of higher mental processes, inner speech represents the end point of a developmental process in which external conversation gradually becomes internalized to form verbal thought (Vygotsky 1934/1987). Like its semi-covert developmental precursor, private speech, inner speech retains the dialogic nature of the external discourse from which it derives. Fernyhough's four-stage model of the development of inner speech as conceptualized by Vygotsky suggests two distinct forms of dialogic inner speech (Fernyhough Reference Fernyhough2004): expanded inner speech, where the give-and-take quality of external conversation permeates the verbal mentation; and a condensed variety of inner speech, where inner speech becomes “thinking in pure meanings” (Vygotsky 1934/1987), having lost most of the acoustics and structure of external dialogue. According to Fernyhough's (2004) theory of AVHs, which draws on Vygotskian ideas about the developmental significance of inner speech, AVHs result from the temporary re-expansion of condensed inner speech, particularly under conditions of stress and cognitive challenge. The acoustic properties of the voices in inner dialogue are thus not attenuated but are experienced fully. The question then is how it is possible that cognition (inner voice) produced by self may be experienced as produced by other.

The cognitive dysfunction that results in the failure to differentiate self from other in inner speech may be explained by a forward model similar to the one underpinning Hurley's layer 2. SCM relies on the forward model of motor control as proposed by Miall (Reference Miall2003) to postulate the subpersonal process that predicts the consequences of motor commands and compares them with the desired state. In her article on delusions of alien control, Blakemore (Reference Blakemore2003) uses this model to explain how an internal predictor uses information about intentions to enable the distinction between self-generated and externally generated sensory events. The forward model is dysfunctional when it cannot accurately predict the sensory consequences of a movement based on the efference copy of the motor command. This results in sensory discrepancy and a failure to cancel the reafference or actual feedback, so that the self-produced movement feels externally caused (Blakemore Reference Blakemore2003; see also Frith et al. Reference Frith, Blakemore and Wolpert2000b). Although developed to explain abnormalities involving overt actions, this forward model has recently been applied to inner speech (Jones & Fernyhough Reference Jones and Fernyhough2007b). Jones and Fernyhough's application proposes a direct causal mechanism leading from a malfunction of the predicted state to the experience of inner speech as being of alien origin. When the brain either produces a degraded predicted state or fails to produce a predicted state at all from the initial inner speech motor command, the consequence is that an emotion of self-authorship is not felt and instead the inner speech is experienced as authored by an other.

For any model of the mind or cognitive functioning to be complete, it must relate to the brain. Thus, we need to understand the neural underpinnings of the predicted-state mechanism proposed by the forward model. This may require investigating networks, such as the interactions between perceptual and motor areas (Jones & Fernyhough Reference Jones and Fernyhough2007b). For example, Leube et al. (Reference Leube, Knoblich, Erb, Grodd, Bartels and Kircher2003) have suggested that neurological activity associated with a deficit in the efference copy mechanism may involve the cortical network that de Vignemont and Fourneret (Reference de Vignemont and Fourneret2004) found implicated in action attribution, including the prefrontal and the parietal cortex, the supplementary motor area, and the cerebellum. In terms of AVHs, Shergill et al. (Reference Shergill, Bullmore, Simmons, Murray and McGuire2000) examined functional magnetic resonance imaging (fMRI) scans of patients with schizophrenia made while the patients were experiencing AVHs. They noted that the pattern of activation observed during AVHs was remarkably similar to that seen when healthy volunteers engaged in auditory verbal imagery (AVI), which is produced when one imagines being spoken to by another person. Specifically, Shergill et al. (Reference Shergill, Bullmore, Simmons, Murray and McGuire2000) observed common activation of bilateral frontal and temporal gyri, along with right-sided precentral and inferior parietal gyri. Increased supplementary motor area activation was associated with healthy participants generating auditory verbal images; however, the supplementary motor area (SMA) was only weakly activated during AVHs. Other studies have suggested a role for the right anterior cingulate gyrus (see Jones & Fernyhough Reference Jones and Fernyhough2007a and studies cited therein). Given that the parietal and cingulate cortices subserve attention to internal and external bodily space and the attribution of significance to sensory information, they provide a plausible neural substrate for the misattribution of self-generated inner speech to other (see Spence et al. Reference Spence, Brooks, Hirsch, Liddle, Meehan and Grasby1997).