Show Summary Details
Page of

(p. 205) Persecutory delusions and theory of mind: longstanding debates and emerging issues 

(p. 205) Persecutory delusions and theory of mind: longstanding debates and emerging issues
(p. 205) Persecutory delusions and theory of mind: longstanding debates and emerging issues

Rhiannon Corcoran

and Suzanne Kaiser

Page of

Subscriber: null; date: 16 January 2019


The rapidity with which interest grew in the topic of ‘theory of mind’ (ToM) in schizophrenia during the latter half of the 1990s took those of us who were involved in the earliest studies of this socio-cognitive skill by surprise. The popularity of the topic probably comes down to two factors. The first is the elegance of Frith’s (1992) meta-representational model of schizophrenia into which the ToM studies fell. Using his highly convincing overarching model, he was able to conduct symptom-based studies which he could ally to diagnosis. Like no other in the field, this model managed to provide a synthesis between the controversial diagnosis on the one hand and the heterogeneity of the symptoms associated with it on the other. Furthermore, in doing so it tackled the polarity between advocates of the biological perspective and those whose understanding of psychosis came from a more psychological perspective. The second reason for the popularity of the ToM proposal was the intuitively compelling suggestion that a misunderstanding of other people’s intentions lay at the heart of paranoia. While this suggestion teeters towards tautology, it was nevertheless open to scientific scrutiny by the adoption of carefully designed empirical tests, some of which were borrowed from the child development literature and others which emerged expressly for the purpose of studying mentalizing in the adult brain. The popularity of this approach to understanding the signs and symptoms of schizophrenia has carried over into genetic and evolutionary accounts of schizophrenia through proposals such as those of Brüne (2001) and Burns (2004) which argue that schizophrenia is the price that must be paid for the significant and rapid advancement in social cognition of Homonids.

(p. 206) We will not here present an exhaustive review of the ToM literature in this area but we would suggest that readers look at the two excellent reviews of the wider area presented by Harrington et al. (2005) and Brüne (2005). We begin with a definition of the term ‘theory of mind’ as it was originally used in association with psychosis, and go on to summarize and assess the arguments that Frith put forward about how and why this socio-cognitive skill may be linked with persecutory delusions (PD). We will follow this with a presentation of key studies which have led to the present state of knowledge within what are now relatively longstanding debates about the relationship of ToM to PD. This will emphasize both what we know and what we still do not know. Some suggestions about how to take current debates forward will be offered. We will proceed with an investigation of some of the emerging issues in the literature including a brief review of studies of the remediation of ToM, transdiagnostic issues, studies of implicit ToM and the fit of the mentalizing model to the situation.

Longstanding debates

The definition and origins of ToM in the context of persecutory delusions

According to Frith’s model, problems with ToM result from a more general problem with meta-representation, the super-ordinate skill of thinking about the contents of the mind. Thus, meta-representation involves the secondary, as opposed to the primary, representation of the world. Indeed, Brüne (2005) has suggested that it may be the failure to distinguish secondary representations from primary representations that underlies the firmly held beliefs that we refer to as delusions. The ability to meta-represent enables introspection and self-awareness as well as reflection on the mental states of others. It is the ability to think about the thoughts, intentions and beliefs of others, that will be used as the working definition of ToM in this chapter because it is this precise ability that was originally argued to be relevant to persecutory delusions (PD).

The possession of a functioning ToM enables one to mentalize or ‘mind-read’, in other words, to infer the mental states of other people. Experience tells us that individual differences exist in the efficiency of this skill. We all know people who have a tendency to ‘put their foot in it’, who fail to ‘take a hint’, who lack tact or who tend to ‘get the wrong end of the stick’. These are familiar terms which people use to reflect a relative lack of aptitude in this key socio-cognitive skill. Some studies exploring mentalizing in healthy adult samples have linked a relative impoverishment of this ability to levels of schizotypal personality wherein people endorse sub-clinical psychotic-like beliefs and (p. 207) experiences (Langdon and Coltheart 1999; Pickup 2006). It is significant that the particular beliefs and experiences sampled in these questionnaires resemble the positive symptoms of psychosis of which PD are probably the most common (Garety et al. 1988; Jorgensen and Jensen 1994; Stompe et al. 1999). According to Wing et al. (1974) PD arise when ‘the subject believes that someone, or some organization, or some force or power, is trying to harm him/her in some way: to damage his/her reputation, to cause him/her bodily injury, to drive him/her mad or to bring about his/her death’ (p. 170). This definition alongside the DSM-IV criteria which asserts that a delusion is ‘a false personal belief based on incorrect inference about external reality …’, makes the proposal that PD stem from incorrect inferences about another’s intention (i.e. a ToM difficulty) very obvious.

Frith suggested that people who hold beliefs that others mean them harm do so as a result of a temporary malfunction of the meta-representational mechanism when it is engaged in inferring the intentions of other people. The specificity of this malfunction implies that for people who suffer from PD, in the absence of any other psychotic sign or symptom, the stages of meta-representation prior to mentalizing will be intact (Fig. 10.1). In other words, the abilities to introspect, engage in wilful acts and to self-monitor, will be intact in such patients. However, in practice, PD do not occur in isolation but instead tend to present alongside auditory hallucinations as well as other delusions and sometimes thought disorder (e.g. Liddle 1987). Also integral in the original formulation was the assumption that the ToM problem is associated with the acute state and that when symptoms remit this is because the temporary meta-represetational fault has been rectified. It therefore follows that (p. 208) patients who have had PD in the past but who are now in remission should have intact ToM skills. Again, because of the presence of other signs and symptoms, it is difficult to assess adequately the veracity of this claim, and the debate about state and trait continues in the literature as discussed below. The assumption that the difficulty is linked to the deluded state explained, in part, by the content of the incorrect inferences that are made in the context of PD. The misinterpretations are characteristically malign because patients are used to drawing inferences about others’ intentions efficiently and accurately, and they anticipate being able to do so consistently in the future. Consequently when they begin to have difficulties, they make an external attribution to account for this and conclude that the reason for their difficulty is that the other person is trying to conceal his/her intentions. Following from that conclusion, one likely default assumption to account for the concealment of intention is that the other person’s intentions towards one are not benign. This series of speculations could be scrutinized by examining the ability of patients with PD to understand the intention to deceive when the motive is more clearly benign, for example when one tries to conceal a surprise party or a present. To our knowledge no studies of this sort have been conducted.

Fig. 10.1 The stages within Frith’s (1992) meta-representational model.

Fig. 10.1
The stages within Frith’s (1992) meta-representational model.

More recently other researchers have tried to account for the content of PD by suggesting that the ToM difficulty is not a deficit but rather a bias which leads to them holding a default ‘nasty ToM’. This is an idea drawn originally from Happé and Frith’s (1996) work on children with conduct disorder but which recent work by J. Taylor (personal communication) has shown to relate to PD in forensic populations. The idea of a bias rather than a deficit in ToM gives the socio-cognitive problems a further aetiological perspective where it fits with the growing literature showing a link between positive symptoms and histories of abuse (e.g. Read et al. 2005). If mentalizing is like other skills in having a learned element, then one can deduce that a history characterized by abuse may lead to the development of a ‘ToM store’ which is rich in malign intentions. A further neglected issue that comes through clearly in the growing literature on the psychology of violent behaviour in the context of PD is the importance of empathic as opposed to mental state inferences in this group of patients. Abu-Akel and Khalid (2004) found intact ToM in combination with deficient empathic inferences and high levels of hostility towards others in participants who had a history of violence towards others. The extent to which the ability to infer the cognitive states of others should be regarded as distinct from the ability to infer others’ emotional states has arisen quite often in the wider ToM literature (e.g. Stone 2000). Recently this issue has been clarified by functional magnetic resonance imaging showing both shared and separable components in these two types of social inference, (p. 209) with the amygdala in particular being more active when participants think about others’ emotions (Völlm et al. 2006). The pattern of activity is consistent with the arguments of LeDoux (1996) who proposed that both cortical and limbic circuits exist for decision-making (i.e. ‘hot’ and ‘cold’ cognition). It would seem that those with a tendency to act in a violent manner on the basis of their PD are, possibly as a result of increased levels of hostility arising from negative emotional experiences with others now and in the past, less capable of accurately inferring the emotions of other people. It would be valuable to examine the corresponding amygdala activity associated with the emotional inferences that such patients draw during empathic inference tests focusing on strangers versus people known to them, a distinction which has come to a head very recently in the mentalizing literature and to which we return in the section on ‘Theory-theory or simulation theory: the role of different contexts’ (Mitchell et al. 2006).

Another aspect of the original model that has remained untested to date is the specificity of the ToM difficulty to others’ intentions. Much of the work carried out so far has focused on false belief understanding as opposed to the understanding of intentions (e.g. Langdon et al. 1997; Sarfati et al., 1999). Some of the early work did look specifically at intentions, for example when they were hidden behind indirect speech acts, and found difficulties in patients with paranoid symptoms (Corcoran et al. 1995). More recently Corcoran and Frith (2003) have suggested that it is intention to deceive which appears to be most robustly associated with the ToM difficulty in schizophrenia. However, a real test of whether belief, knowledge or intention attributions are differentially affected in the context of PD remains to be conducted.

So far we have looked at questions deriving directly from Frith’s original model. While all of them are still relevant, perhaps the most challenging of these remains the specific link of ToM problems to PD. As we shall see, this link is far from clearly established.

Are ToM difficulties related to PD at all?

The literature on schizophrenia consistently shows that neuropsychological deficits are not associated with positive symptoms (e.g. Keefe et al. 2006). In the context of these very robust negative findings, the evidence associating ToM problems to PD looks quite convincing on the face of it. However, a proper review of the literature demonstrates that the situation is far from straightforward.

The two excellent reviews of the literature on ToM in schizophrenia by Harrington et al. (2005) and Brüne (2005) provide us with enough information to assess whether or not the weight of evidence supports the existence of a (p. 210) ToM difficulty in relation to PD. Having assessed the methodological issues which make the assessment of this question difficult—for example, how the presence of PD are established; whether or not other symptoms were present; the chronicity of the disorder; the extent of the contribution made to any ToM difficulty by problems with other ‘domain-general’ cognitive skills and the nature of the ToM task itself—Harrington et al. concluded ‘it is clear that current evidence points to thought disorder and paranoid symptoms as being most consistently associated with the well-established ToM deficit present in schizophrenia.’ (p. 275). Brüne is a little less firm in his assessment of the evidence overall but is nevertheless happy to conclude that: ‘It is therefore conceivable that patients with persecutory delusions are particularly compromised when they have to “mentalize on the spot” but may perform quite normally on standard ToM tests when not under time pressure.’ (p. 34).

One study by Walston et al. (2000) indicated that no clear problems are associated with ToM when circumscribed PD exist in isolation of other signs and symptoms of psychosis. These authors examined a small number of patients with ‘pure’ PD in the context of delusional disorder using a range of ToM tasks including stories to second-order, hints and jokes. The good performance on ToM tasks by this group led the authors to suggest that PD should be distinguished from other paranoid-type symptoms when assessing ToM. However, these patients did have difficulties when they were asked to try to understand the mental states of their ‘persecutors’. This may support the presence of a circumscribed ToM deficit existing alongside the circumscribed delusion.

Compelling evidence for the association between ToM and PD comes from recent studies taking a transdiagnostic approach to the psychology of PD (see ‘Transdiagnostic issues’ below) and on the basis of these findings and the conclusions of the reviews referred to above, we think it is reasonable to conclude that ToM difficulties are associated with the presence of PD specifically.

Abu-Akel (1999) has suggested that some people with schizophrenia may over-attribute the intentions of others and thus show hyper-ToM as opposed to a deficit or bias. Interestingly, the results of a recent study by Russell et al. (2006) support this proposal to some extent. Russell et al. used animations which showed geometric shapes moving in a random fashion, in an apparently goal-directed way (e.g. resembling foraging) or finally in a way that looked like they were interacting on the basis of a knowledge of mental states (e.g. resembling teasing). What these authors noticed was that with the lower-level stimuli (i.e. random and goal-directed), some patients in their paranoid group (which included PD and other symptoms) had a tendency to attribute mental states to these animations. On the other hand, when the interaction (p. 211) animations were considered, there was good evidence that the paranoid group tended not to attribute mental states as much as controls. Perhaps the association between ToM and paranoia is more complex than early studies imply. Russell et al. suggested that people with paranoid-type delusions can attribute mental states on-line but do so unreliably and rather inappropriately—which was effectively the same argument as originally proposed by Frith (1992). A slightly different interpretation of the diversity of the ToM difficulty seen in schizophrenia was offered by Abu-Akel and Bailey (2000) who suggested that there may be three types of ToM malfunction; impaired ToM, intact ToM with compromised capacity to apply the information, and hyper-ToM. This tripartite model is eminently testable and we await further work by this group.

State or trait?

The debate about whether the ToM difficulty associated with the diagnosis of schizophrenia is state- or trait-related is ongoing, as both Harrington et al. and Brüne point out. However, recent evidence suggests that perhaps the best way to summarize the literature with respect to PD is to propose that the ToM difficulty is present during remission of these beliefs, and is seen in relation to high schizotypy (Pickup 2006) and in unaffected relatives of people with schizophrenia (Janssen et al. 2003). However, the level of difficulty associated with the skill appears to increase during episodes of acute illness. Amongst the studies that would be consistent with such an interpretation would be the cross-sectional study conducted by Randall et al. (2003) which looked specifically at PD and the findings of Drury et al. (1998) which included a short follow-up period. These authors concluded that the case for the ToM difficulty being a consistent trait-like feature for those showing a propensity for PD was strong.

Resolving the longstanding debates

In order to bring the debates addressed above to firmer conclusions, longitudinal studies of first-onset or ‘at-risk’ cohorts are required. Such studies should include several ToM tasks assessing intentions and false beliefs, testing explicit and implicit forms of mental state inference (see section on ‘Measuring implicit or ‘on-line’ ToM’ below) and using both verbal and non-verbal paradigms to do so. Prior to this, however, groups conducting this type of research should endeavour to establish the psychometric properties of the measures they use (e.g. see Shryane et al. 2007). They should explore the construct validity of their measures and the test–retest and inter-rater reliability of the measures. Furthermore, the measurement of PD needs to be explicit and based upon accepted criteria for the presence of delusions with clear (p. 212) evidence of their persecutory nature established by appropriate standardized assessments which also assess the presence, severity and chronicity of other signs and symptoms of psychosis. Studies should continue to assess the potential impact of ‘domain-general skills’ on ToM as the specificity of this skill is far from well-established (e.g. Russell 1998); but the search for candidate cognitive sub-components or contributors to social cognition should be widened to include not only IQ, but also aspects of higher order cognition which could theoretically underpin mentalizing. Such skills are generally not included in standard neuropsychological batteries. Some suggestions as to what these might be are presented in ‘How to mentalize when ToM malfunctions: the use of domain-general skills’ below, but assessments of the general ability to simulate per se (i.e. the ability to run through scenarios in the head with a view to generating solutions to a problem) would be important. Finally, attempts need to be made to establish the impact of the duration and chronicity of the illness on social cognition as well as antipsychotic drug treatment, a factor which has been conspicuously absent in previous studies of ToM.

Emerging issues

Transdiagnostic issues

We would like now to move away from debates which have been in the forefront of research for some time and instead consider themes that are beginning to emerge in the more recent literature. The first of these is the issue of whether the ToM impairment associated with PD is seen across diagnoses. As the majority of published studies on ToM in psychosis have been on samples of patients with schizophrenia, this issue has been largely avoided. Early work by Doody et al. (1998) reported that ToM was intact in a group of patients with affective disorder in spite of a shared symptom profile with a group of patients with schizophrenia whose ToM was impaired. Kerr et al. (2003) went on to demonstrate poor ToM in patients with bipolar disorder. However, these studies did not specifically focus on PD. Moore et al. (2006) did look exclusively at PD in patients with very-late-onset schizophrenia-like psychosis and found evidence of impairments in the understanding of deception on a story task which tested both deception and false belief to second-order level. This study agrees with the suggestion of Corcoran and Frith (2003) that it may be the understanding of the intention to deceive that is core to PD. Other work of a transdiagnostic nature conducted by R. Corcoran et al. (2007) has demonstrated that the presence of PD is associated with ToM difficulty in the context of both schizophrenia spectrum disorders and affective disorders. Evidence is therefore beginning to accrue to support the idea that, although ToM problems are robustly associated with a (p. 213) diagnosis of schizophrenia (and that within that diagnosis, several signs and symptoms appear to be related to ToM problems), these same difficulties are central to PD wherever they originate.

How to mentalize when ToM malfunctions: the use of domain-general skills

According to Dennett (1987), humans take the intentional stance by default. In other words we cannot help but interpret others’ behaviours as resulting from their thoughts and intentions. While it may very well be the case that people with autistic spectrum disorders and possibly people with the negative features of schizophrenia are exceptions to this rule in that they may not be driven, by default, to think about others (or themselves) in this way (Frith 1992; Baron-Cohen 1995), it would seem that people with PD do take the intentional stance. The weight of evidence (from formal empirical studies and clinical experience) indicates that people with PD do have a ToM, in the sense that they understand that others are driven by mental states, but that their inferences about these mental states are compromised. If we assume that the default adoption of the intentional stance indicates the working of an evolved mechanism (an assumption that not everyone would agree with, of course) then it follows that this mechanism fails in the presence of PD but that patients are still driven to understand the intentions of others. How do they do this? Some work proposes that people rely on domain-general skills to answer mentalizing questions when their ToM modules ‘crash’, and this, alongside the fact that these domain-general skills may themselves be compromised, results in the drawing of unreliable ToM inferences. Three studies have demonstrated that skills needed to draw inductive inferences are compromised in patients with schizophrenia who show ToM problems. The idea is that because the ToM ‘module’ is malfunctioning, people with schizophrenia have to rely on a poor or biased autobiographical memory in conjunction with impaired social conditional reasoning to draw ToM conclusions (Corcoran 2003; Corcoran and Frith 2003, 2005). It is possible that if these domain-general skills were intact, then the ToM output resulting from them would be reliable and there would be no need for a devoted ToM module at all. Some later work has looked at heuristic reasoning, a form of fast experience-based reasoning thought to be relied on for everyday problem-solving situations, in patients with PD. Corcoran et al. (2006) reported that patients with PD tended not to use information about themselves as a basis for inferences about the experiences of other people. This result is consis tent with there being poor simulation-based mentalizing in patients with PD (see ‘Theory-theory or simulation theory: the role of different contexts’ below).

(p. 214) Amongst other work focusing on less-well-established cognitive correlates or models of ToM is that of Charlton et al. who propose that ToM is intact in patients with PD (in the sense of understanding that others behave according to the contents of their minds) but their ToM judgments go wrong because these judgements are arrived at using the somatic marker mechanism (Damasio 1995). This means that people work backwards from their emotional body states (e.g. fear) to guess which intentions of others would cause such body states (e.g. intention to harm). It is the bias towards particular emotional states that informs mentalizing in people with PD. Although not directly tested, this hypothesis is consistent with the general findings of Corcoran et al. (2006) of biased heuristic reasoning in the context of threatening situations. Charlton’s (2001) somatic marker proposal could be more directly tested in patients with PD using the tasks derived by Damasio et al. in their studies of patients with ventromedial lesions such as the Iowa Gambling Task (Bechara et al. 1994).

The remediation of ToM

It has taken a relatively long time for studies aimed at improving ToM in people with schizophrenia to emerge. It is not clear what led to this relative inertia as researchers were interested in the topic from an early stage. The lack of publications may reflect a lack of success at establishing these trials or a lack of positive findings. To date there have been three published attempts to remediate ToM in people with schizophrenia spectrum disorders (Roncone et al. 2004; Penn et al. 2005; Kayser et al. 2006). All three of these studies reported some success, though they all suffer from methodological problems, casting doubt on the validity of their findings. The problems include failure to take account of regression to the mean (Penn et al. 2005), use of inappropriate psychiatric control groups (Roncone et al. 2004; Kayser et al. 2006), lack of a control group against which to compare the ToM therapy (Penn et al. 2005), confounding of general and directed therapies (Roncone et al. 2004), and weak analyses (Kayser et al. 2006). Of the three studies Roncone et al.’s is perhaps the best designed. It offers some evidence that ToM may be improved. However, the fact that Roncone et al. embedded a directed therapy for ToM within a more general cognitive therapy means that it is impossible to determine what aspect of their remediation programme was responsible for the positive results. These studies have used heterogeneous groups of patients with various signs and symptoms of psychosis. Frith (1992) argued, and empirical evidence indicates, that patients with negative signs have a severe and enduring ToM impairment similar in extent to that seen in people with autism (e.g. Corcoran et al. 1995). Attempts at remediation including such (p. 215) severely impaired participants may jeopardize findings which may be more fruitful if they were to focus on patients with PD.

Measuring implicit or ‘on-line’ ToM

Implicit ToM is more automatic, more moment-to-moment, and less effortful than explicit ToM, which requires overt reasoning and effort. Implicit ToM is the type of mentalizing employed in everyday social situations, particularly conversation, and it is possible that implicit ToM might be more driven by fast emotionally based reasoning. The significant neglect of this type of ToM reasoning in favour of the ecologically less sound use of empirical tests has, in our opinion, hindered advance in the understanding of how ToM really functions and relates to PD and psychosis more generally. The few studies that have been conducted have used diverse methods and reached differing conclusions. Quantitative work by Langdon et al. (1997) showed that people with schizophrenia used fewer mental state terms when describing their responses to a picture-sequencing task than healthy controls. This finding was replicated by Russell et al. (2006) in the context of descriptions of scenarios involving animated triangles. However, both Langdon et al. and Russell et al. examined mental state references in the context of monologue, even though mental state referents would most likely occur during dialogue. Conducting an examination of dialogue, McCabe et al. (2004) used a qualitative analysis of conversations between people with positive symptoms of psychosis and mental health professionals. They concluded that implicit ToM was intact since patients could maintain conversation and because they recognized that others did not share their delusional beliefs. However, an alternative explanation to account for McCabe et al.’s observations is presented below.

As part of the work conducted for her PhD, Kaiser (2007) attempted to improve on the published studies in the area by using a measure of implicit ToM and empathic inference based on dialogues between the experimenter and the participant probing brief stories written by the participant in response to verbal or pictorial prompts. Controlling for total speech produced, Kaiser compared ratings of mental state and emotional references about story characters, positive references to own mental state (e.g. ‘I think’, ‘In my opinion’), and references to the conversational partner’s mental states. Eighteen people with schizophrenia spectrum disorders with chronic histories and often persistent positive symptoms including PD participated in this study along with nine healthy adults with similar age, sex and IQ profiles. Kaiser found that the patients made significantly fewer mental state and emotional references and references to the conversational partner’s mental state than the healthy controls. However, there was no difference between the groups for (p. 216) references to own mental states. According to this more rigorous study of implicit mentalizing, operationalized as references to both the story characters’ mental states and the conversational partner’s mental state, this type of mentalizing is impaired, as is implicit empathizing in patients with schizophrenia spectrum disorders. These results support the findings of Langdon et al. (1997) and Russell et al. (2006) but not the conclusions of McCabe et al. (2004).

We believe that the conclusions drawn by McCabe et al. on the basis of transcribed therapy session of patients with positive symptoms are incorrect. We propose that the ability of patients to converse and exchange views with their therapists, and indeed with other people more generally, in meaningful ways can be accounted for by their intact processes of alignment (Garrod and Pickering 2004). Alignment is an automatic process that begins with the first utterance of a dialogue and continues over the course of the interaction to make the interlocutors’ representations at several levels become more similar. It is relevant here that, according to Garrod and Pickering, the ultimate form of alignment takes place at the level of ‘situation models’, global representations of a situation that take into account information about sequence, space, time, contributing individuals, and causality. Each interlocutor creates and updates his or her own situation model based on contributions to the interaction, and the extent of the similarity between interlocutors’ situation models influences the success of the conversation. Garrod and Pickering also illustrate clearly how alignment can produce meaningful conversation in the presence of impoverished ToM.

S.L. Kaiser et al. (unpublished) examined the process of alignment in a sample of 59 patients with schizophrenia and 38 age-, sex- and IQ-matched controls. The task examined the ability to align responses to requests for level of politeness while also establishing the ability of participants to infer the knowledge state of one of the characters involved in the scenarios used. Using this quantitative task, she showed that patients were able to align for politeness to the same degree as controls while demonstrating relatively impoverished knowledge attribution (i.e. ToM). While the sample used here was a mixed group of patients with DSM-IV-defined schizophrenia, some (~40%) had active PD alongside other signs and symptoms. This process, which appears to enable interaction with others despite poor ToM, requires more investigation in those with delusions as well as in patients with formal thought disorder where it is possible that processes of alignment might very well fail.

Theory-theory or simulation theory: the role of different contexts

Recent imaging work by Mitchell et al. (2006) has led to the proposal that the ToM system contains both theory-based and simulation-based elements, (p. 217) depending on the target of the mentalizing process. This has implications for the debate over the relationship between ToM and PD. Mitchell et al.’s findings showed that different regions of the medial prefrontal cortex were activated in healthy controls depending on the subjective degree of similarity of the target to the participant. These authors argue that nearly all traditional ToM tasks used in imaging studies consist of characters who will be seen as dissimilar to the participant, triggering theory-based mentalizing (e.g. Russell et al. 2000; Brunet et al. 2003; Calarge et al. 2003; Völlm et al. 2006). This is also true of ToM studies in people with PD where, to our knowledge, patients have never been asked to infer the mental states of people known to them or judged to be similar to them. Thus, the findings relating ToM to PD refer only to theory-based mentalizing. If faults in the theory-based mentalizing system play a role in the development and maintenance of PD, it would mean that people with PD have learned mentalizing rules that are incorrect. If these incorrect rules are applied during the mentalizing process, the person will be unable accurately to discern others’ intentions, leading them to conclude that the target is hiding (presumably malevolent) intentions. However, if Mitchell et al. are correct, this process will only occur when targets are perceived as dissimilar to the participant. Currently it is not known whether simulation-based mentalizing is impaired in psychosis, but on the basis of the evidence available it seems that PD in part results from a combined process of judgements of dissimilarity and inaccurate theory-based mentalizing. These speculations raise several questions for future research. It is unknown whether judgements of similarity (which would trigger theory-based or simulation-based mentalizing) are affected by psychosis in general or differ among people with different symptom profiles. If these judgements are affected, people with psychosis may employ theory-based versus simulation-based mentalizing differently from healthy populations. Clearly, suitable tests of similarity/familiarity judgments, simulation per se and simulation-based mentalizing should be developed and applied to samples with psychosis.

The future

In this chapter we have presented an overview of the findings which concern the nature of the relationship between PD and ToM. Many questions deriving from Frith’s (1992) original proposal still remain to be answered by future studies with improved methodological rigour. However, the emerging issues in this area are clearly relevant to the association of ToM with PD. These extend the field of future research outward to include different aspects of ToM and related processes into various diagnoses where PD present.


Abu-Akel, A. (1999) Impaired theory of mind in schizophrenia. Pragmatics and Cognition, 7, 247–282.Find this resource:

Abu-Akel, A. and Bailey, A.L. (2000) Letter to the editor. Psychological Medicine, 30, 735–738.Find this resource:

Abu-Akel, A. and Khalid, A. (2004) ‘Theory of mind’ in violent and non-violent patients with paranoid schizophrenia. Schizophrenia Research, 69, 46–53.Find this resource:

Baron-Cohen, S. (1995) Mindblindness. An Essay on Autism and Theory of Mind. Cambridge, MA/London: MIT Press.Find this resource:

    Bechara, A., Damasio, A.R., Damasio, H. and Anderson, S. (1994) Insensitivity to future consequences following damage to the human prefrontal cortex. Cognition, 50, 7–12.Find this resource:

    Brüne, M. (2001) Social cognition and psychopathology in an evolutionary perspective. Psychopathology, 34, 85–94.Find this resource:

    Brüne, M. (2005) ‘Theory of mind’ in schizophrenia: a review of the literature. Schizophrenia Bulletin, 31, 21–42.Find this resource:

    Brunet, E., Sarfati, Y., Hardy-Baylé, M.-C. and Decety, J. (2003) Abnormalities of brain function during a nonverbal theory of mind task in schizophrenia. Neuropsychologia, 41, 1574–1582.Find this resource:

    Burns, J.K. (2004) An evolutionary theory of schizophrenia: cortical connectivity, metarepresentation, and the social brain. Behavioral and Brain Sciences, 27, 831–855.Find this resource:

    Calarge, C., Andreasen, N.C. and O’Leary, D.S. (2003) Visualizing how one brain understands another: a PET study of theory of mind. American Journal of Psychiatry, 160, 1954–1964.Find this resource:

    Charlton, B.G. (2001) Theory of mind and the “somatic marker mechanism” (SMM). Behavioral and Brain Sciences, 24, 1141–1142.Find this resource:

    Corcoran, R. (2003) Inductive reasoning and the understanding of intention in schizophrenia. Cognitive Neuropsychiatry, 8, 223–235.Find this resource:

    Corcoran, R. and Frith, C. D. (2003) Autobiographical memory and theory of mind: evidence of a relationship in schizophrenia. Psychological Medicine, 33, 897–905.Find this resource:

    Corcoran, R. and Frith C.D. (2005) Thematic reasoning and theory of mind. Accounting for social inference difficulties in schizophrenia. Evolutionary Psychology, 3, 1–19.Find this resource:

    Corcoran, R., Mercer, G. and Frith, C.D. (1995) Schizophrenia, symptomatology and social inference: investigating ‘theory of mind’ in people with schizophrenia. Schizophrenia Research, 17, 5–13.Find this resource:

    Corcoran, R., Cummins, S., Rowse, G. et al. (2006) Reasoning under uncertainty: heuristic judgments in patients with persecutory delusions or depression. Psychological Medicine, 36, 1109–1118.Find this resource:

    Corcoran, R., Rowse, G., Moore, R. et al. (2007) A transdiagnostic investigation of ‘theory of mind’ and ‘jumping to conclusions’ in patients with persecutory delusions. Psychological Medicine doi: 10.1017/S0033291707002152.Find this resource:

      Damasio, A.R. (1995) Towards a neurobiology of emotion and feeling: operational concepts and hypotheses. The Neuroscientist, 1, 19–25.Find this resource:

      Dennett, D.C (1987) The Intentional Stance. Cambridge, MA: MIT Press.Find this resource:

        Doody, G.A., Götz, M., Johnstone, E.C., Frith, C.D. and Cunningham Owens, D.G. (1998) Theory of mind and psychoses. Psychological Medicine, 28, 397–405.Find this resource:

        (p. 219) Drury, V.M., Robinson, E.J. and Birchwood, M. (1998) ‘Theory of mind’ skills during an acute episode of psychosis and following recovery. Psychological Medicine, 28, 1101–1112.Find this resource:

        Frith, C.D. (1992) The Cognitive Neuropsychology of Schizophrenia. Hove: Erlbaum.Find this resource:

          Garety, P.A., Everitt, B.S. and Hemsley, D.R. (1988) The characteristics of delusions: a cluster analysis of deluded subjects. European Archives of Psychiatry and Neurological Sciences, 237, 112–114.Find this resource:

          Garrod, S. and Pickering, M.J. (2004) Why is conversation so easy? Trends in Cognitive Sciences, 8, 8–11.Find this resource:

          Happé, F.G.E. and Frith, U. (1996) theory of mind and social impairment in children with conduct disorder. British Journal of Developmental Psychology, 14, 385–398.Find this resource:

          Harrington, L., Siegart, R.J. and McClure, J. (2005) Theory of mind in schizophrenia: a critical review. Cognitive Neuropsychiatry, 10, 249–286.Find this resource:

          Janssen, I., Krabbendam, L., Jolles, J. and van Os, J. (2003) Alterations in theory of mind in patients with schizophrenia and non-psychotic relatives. Acta Psychiatrica Scandinavica, 108, 110–117.Find this resource:

          Jorgensen, P. and Jensen, J. (1994) Delusional beliefs in first admitters. Psychopathology, 27, 100–112.Find this resource:

          Kaiser, S.L. (2007) Theory of mind and cognitive processes in psychosis. Unpublished doctoral dissertation, University of Manchester, Manchester, UK.Find this resource:

            Kayser, N., Sarfati, Y., Besche, C., and Hardy-Baylé, M.-C. (2006) Elaboration of a rehabilitation method based on a pathogenetic hypothesis of “theory of mind” impairment in schizophrenia. Neuropsychological Rehabilitation, 16, 83–95.Find this resource:

            Keefe, R.S., Bilder, R.M., Harvey, P.D. et al. (2006) Baseline neurocognitive deficits in the CATIE schizophrenia trial. Neuropsychopharmacology, 31, 2033–2046.Find this resource:

            Kerr, N., Dunbar, R.I.M., Bentall, R.P. (2003) Theory of mind deficits in bipolar affective disorder. Journal of Affective Disorders, 73, 253–259.Find this resource:

            Langdon, R. and Coltheart, M. (1999) Mentalising, schizotypy, and schizophrenia. Cognition, 71, 43–71.Find this resource:

            Langdon, R., Michie, P.T., Ward, P.B., McConaghy, N., Catts, S.V. and Coltheart, M. (1997) Defective self and/or other mentalising in schizophrenia: a cognitive neuropsychological approach. Cognitive Neuropsychiatry, 2, 167–193.Find this resource:

            LeDoux, J. (1996) The Emotional Brain. Simon & Schuster, New York.Find this resource:

              Liddle, P.F. (1987) The symptoms of chronic schizophrenia: a re-examination of the positive–negative dichotomy. British Journal of Psychiatry, 151, 145–151.Find this resource:

              McCabe, R., Leudar, I. and Antaki, C. (2004) Do people with schizophrenia display theory of mind deficits in clinical interactions? Psychological Medicine, 34, 1–12.Find this resource:

              Mitchell, J.P., Macrae, C.N. and Banaji, M.R. (2006) Dissociable medial prefrontal contributions to judgments of similar and dissimilar others. Neuron, 50, 655–663.Find this resource:

              Moore, R., Blackwood, N., Corcoran, R. et al. (2006) Misunderstanding the intentions of others: an exploratory study of the cognitive etiology of persecutory delusions in very late-onset schizophrenia-like psychosis. American Journal of Geriatric Psychiatry, 14, 410–418.Find this resource:

              Penn, D., Roberts, D.L., Munt, E.D., Silverstein, E., Jones, N. and Sheitman, B. (2005) A pilot study of social cognition and interaction training (SCIT) for schizophrenia. Schizophrenia Research, 80, 357–359.Find this resource:

              (p. 220) Pickup, G. (2006) theory of mind and its relation to schizotypy. Cognitive Neuropsychiatry, 11, 177–192.Find this resource:

              Randall, F., Corcoran, R., Day, J.C. and Bentall, R.P. (2003) Attention, theory of mind, and causal attributions in people with persecutory delusions: a preliminary investigation. Cognitive Neuropsychiatry, 8, 287–294.Find this resource:

              Read, J., van Os, J., Morrison, A.P. and Ross, C. (2005) Childhood trauma, psychosis and schizophrenia: a literature review with theoretical and clinical implications. Acta Psychiatrica Scandinavica, 112, 330–350.Find this resource:

              Roncone, R., Mazza, M., Frangou, I. et al. (2004) Rehabilitation of theory of mind deficit in schizophrenia: a pilot study of metacognitive strategies in group treatment. Neuropsychological Rehabilitation, 14, 421–435.Find this resource:

              Russell, J. (1998) How executive disorders can bring about an inadequate theory of mind. In: Russell, J. (ed.), Autism as an Executive Disorder, pp. 256–299. Oxford: Oxford University Press.Find this resource:

              Russell, T.A., Rubia, K., Bullmore, E.T. et al. (2000) Exploring the social brain in schizophrenia: left prefrontal underactivation during mental state attribution. American Journal of Psychiatry, 157, 2040–2042.Find this resource:

              Russell, T.A., Reynaud, E., Herba, C., Morris, R. and Corcoran, R. (2006) Do you see what I see? Interpretations of intentional movement in schizophrenia. Schizophrenia Research, 81, 101–111.Find this resource:

              Sarfati, Y., Hardy-Baylé, M.-C., Brunet, E. and Widlöcher, D. (1999) Investigating theory of mind in schizophrenia: influence of verbalizationin disorganized and non-disorganized patients. Schizophrenia Research, 37, 183–190.Find this resource:

              Shryane, N.M., Corcoran, R., Moore, R., Rowse, G., Cummins, S., Kinderman, P., Blackwood, N., Howard, R. and Bentall, R.P. (2007) Deception and false belief in paranoia: modeling theory of mind stones. Cognitive Neuropsychiatry, 13, 8–32.Find this resource:

              Stompe, T., Friedman, A., Ortwein, G. et al. (1999) Comparisons of delusions among schizophrenics in Austria and Pakistan. Psychopathology, 32, 225–234.Find this resource:

              Stone, V.E. (2000) The role of the frontal lobes and amygdala in theory of mind. In: Baron-Cohen S., Tager-Flusberg H. and Cohen, D. (eds), Understanding Other Minds: Perspectives from Autism and Developmental Cognitive Neuroscience, 2nd edn, pp. 253–273. Oxford: Oxford University Press.Find this resource:

                Völlm, B., Taylor, A., Richardson, P. et al. (2006) Neural correlates of theory of mind and empathy: an fMRI study in a nonverbal task. NeuroImage, 29, 90–98.Find this resource:

                Walston, F., Blennerhassett, R.C. and Charlton, B.G. (2000) ‘Theory of mind,’ persecutory delusions and the somatic marker mechanism. Cognitive Neuropsychiatry, 5, 161–174.Find this resource:

                Wing, J.K., Cooper, J.E. and Sartorius, N. (1974) Measurement and Classification of Psychiatric Symptoms: An Instruction Manual for the PSE and Catego Program. Cambridge: Cambridge University Press.Find this resource: