top of page

Religious Delusions

How do religious delusions differ from normal beliefs? After all there are many people who have religious ideas that others may consider different, but this does not mean that such beliefs are abnormal or delusional in nature. What we need to do is to see if the religious belief has the structure of a delusion, also are their signs and symptoms of mental illness. If we have such a situation than we can probably say the belief is a delusion. Additionally, religious delusions tend to fall under the headings of persecution, grandiose and belittlement (Dein 2004). 


Religious delusions can involve normal objects, e.g., god, the devil (Knabb 2012) and may involve prior religious experiences though not everyone has been religious. Often such beliefs can be quite clear cut, but it is not unheard of for such religious belief and even delusions to shift in their content (Drinnan 2006).


Religious delusions have found with delusions of reference, grandiose ability, identity, and delusions of control. But in those with severe religious delusions, we see less reference / control delusions and more delusions of grandeur (Suhail 2010). About 20% of those with psychosis suffer religious delusions, as well delusions of grandiosity and control and were just as likely to engage with treatment (Iyassu 2013). Religious delusions were found in about a third of those with different psychosis and may be commoner in mood disorder (Noort 2018)


The Ventromedial Prefrontal cortex has been implicated in religious delusions. Those with damage to the vmPFC tended to more authoritarianism and fundamentalism (Asp 2012). Assessing religious beliefs tends to activate the vmPFC, as well as other areas (Harris 2009). Right dorsolateral prefrontal cortex is also involved, e.g., when religious people read religious passages (Azari 2001). There may be right orbitofrontal, superior frontal, parietal cortex, and cerebellum (Azari 2005). A comparison of patients with mystical experiences found that mysticism was related to lesions in the dorsolateral prefrontal cortex (Cristofori 2016).


Neuroimaging has implicated various brain areas in religious beliefs. Belief might involve activity in the left vmPFC. Fear of god may relate to reduced volume of the left precuneus and orbitofrontal cortex, but non-threatening god beliefs were associated with an increase in volumes. Interestingly the pattern of neural activation differs across cultures (Gaw 2019). A diffuse pattern of brain activation was seen in Carmelite nuns whilst ‘in union with god’ (Beauregard 2006). Spirituality may lead to a reduced activation in the left inferior parietal lobe and decreased activity in the medial thalamus and striatum (Miller 2018)


Religious content of delusions was investigated in a series of Pakistani schizophrenics, as this is a country with a traditional conservative religious basis. Grandiose delusions were commonest, then religious delusions, then persecution and jealousy. Those who were more religious tended to show grandiose delusions of ability / identity as well as power / being prominent. While those who rated lower for religious beliefs were more likely to show delusions of reference or external influence (Suhail 2013). Amongst 200 psychiatric patients in Chandigarh, India, it was found that 20% had religious beliefs with most of these being over 30 and married. They were split evenly between rural and urban areas, and over socio-economic levels and 18 were illiterate (Kala 1982)


A case review of found some of the religious themes included grandiosity (having god given powers) or paranoia (being controlled by evil, being possessed or the evil eye). Such patients would talk, pray, preach, attend religious places or alter their practises. Those with religious symptoms tended to be younger (Atallah 2001). In Poland incidence of religious content related to the delusion has varied (Dudek 2019). Within Malaysia delusions of paranoia, grandiosity and reference were found. There were fewer religious delusions in less developed areas (Azhar 1995). Religious delusions have been found in all religious (Ndetei 1985). Out of 60 Xhosa patients with delusions, 70% had religious delusions. Their beliefs spread over Christian and traditional systems (Connell 2014)


What about the relationship between religious beliefs and religious delusions? A small case series found that those with delusions parental conflict (Drinnan 2006). In schizophrenic patients in Lithuania, most had some religiosity. But religious delusions were commoner in the divorced, the educated and those of an urban birth. Though in a multi-logistic model self-determined religiosity was not related to religious delusions (Rudaleviciene 2008). Religious delusions in protestants and Catholics were compared to other psychotic patients who had no religious allegiance. Protestants tended to have higher scores for religious delusions, as measured by the SAPS religious item (Getz 2001, Clarke 1980).


Looking at the prevalence of delusions in a non-religious group, christian group, a new religious movement (NRM) and a deluded group. The deluded group and NRM had higher scores for the Peters Delusional Inventory (PDI-21). Though the NRM group were less distressed and preoccupied by their beliefs (Peters 1999b). Another study found the NRM tended to show higher scores and were more convinced of their beliefs (Smith 2009).


In a longitudinal study, women who believed in god often felt their TV was sending them messages, they had sinned more, that the world was going to end and has thought echo. Those whose belief in god was uncertain tended to feel that others were not what they seemed to be, that they will be important, believed in telepathy and the occult. The authors interpreted this as showing that alternative beliefs were associated with higher levels of delusional thinking (Aird 2010).


Religious Delusions and Mental Illness


Those with religious delusions may describe themselves as more religious than other groups (Cothran 1986). Another group of schizophrenics found about 60% of men with religious delusions (Rudalevicine 2008). In a series of patients admitted in Rome, in those with religious delusions, there was higher odds of having schizophrenia but lower levels of mood disorder (Raja 2000). Religious delusions in schizophrenics maybe associated with a worse prognosis. Though some studies have found no difference in psychopathology in those with and without religious delusions (Siddle 2004). 


Positive effects from religious delusions have been reported also, such as feeling that religion is important in their life and were less likely to engage in religious practises or psychiatric care (Mohr 2010). A UK sample found religious delusions were commoner in single people, the lower social classes and the educated (Lucas 1962). Following American psychiatric patients over 100 years found no change in prevalence of religious delusions with time (Cannon 2011). A review of chronic psychotic patients that did and did not have self-reported beliefs in possession found that there may have been higher scores of dissociations, sexual abuse, thought control and auditory hallucinations (Goff 1991)


Self-harm of has been found in those with religious delusions. Many religious often have rituals that involve self-harm (Schweroske 2012). Some remove their testicles (Kushner 1967), their penis (Ozan 2010) or castrated themselves due to masturbation (Waugh 1986). review of chronic psychotic patients that did and did not have self-reported beliefs in possession found that there may have been higher scores of dissociations, sexual abuse, thought control and auditory hallucinations, but they did not control for multiple comparisons (Goff 1991)


Religious delusions and Treatment - Some have suggested religious delusions are linked to better outcomes may help in dealing with delusions or hallucinations (Mohr 2010). Amongst Indian schizophrenic treatment effects were independent of religiosity and religious delusions (Siddle 2004). In an outpatient series, there were no differences in demographics in those with religious and non-religious delusions. Those with religious content had more problems with mental health treatment, were less likely to engage in religious practises and to be excluded from the parent religious group (Mohr 2010).


In a group of Xhosa patients with religious delusions, treatment caused a reduction in thinking and behaviour (Connell 2014). The cultural beliefs of relatives have been shown to impact on the treatments. In an Indian out-patient clinic about a quarter believed sorcery could cause mental illness, or that god was punishing people. About a third of relatives would perform rituals to cure the individuals (Kulhara 2000).


Analysis of Case Reports of Religious Delusions


I had collected 81 case reports on religious delusions. 31 were female.

  • 19 had a job.

  • 17 were in relationships.

  • 23 had medical problems, 53 had substance abuse, 54 had past psychiatric issues, 12 had family mental health issues, 34 hallucinated.

  • 81 had other delusional types; 4 had reference,1 had bizarre, 9 had thought interference, 14 had somatic, 19 had grandiose, 27 had paranoia, 4 had jealousy, 5 had depressive delusions and 5 had folie delusions.

  • 23 had children, 24 had siblings, 24 were aggressive.

  • 36 were on treatment of whom 16 responded.

  • 7 engaged in greater religious study, 6 preached more, 23 felt they had a special identity, 9 prayed more often, 2 felt the world was special, 5 began to speak in tongues and 11 felt the devil was present.

  • 41 were Christian, 2 were Buddhists, 2 were Hindu, 6 were Muslim, 6 were Jewish and 4 were classed as other.

  • In terms of the dominant religion in sufferers’ country

    • Buddhism for 1 case, Christianity in 52, Hinduism in 3, Islam in 8 and Jewish in 2 cases

  • 27 lived in the United States, 12 in Romania with the rest from around the world.

Delusions of Reference

Delusions of reference has been defined as when neutral events are felt to refer to the self (Menon 2011). They in essence mean that environmental occurrences are taken to have a special meaning for individuals and only for them and not others. It has occurred via the television or the radio for example. It also seems that delusions of reference can be divided into two groups; delusions of observation and delusions of communication (Startup 2009; Bucci 2008; Bucci 2008b) with delusions of observations possibly being related to delusions of paranoia. 

The original study that found this division, used scores on the Scale for the Assessment of Positive Symptoms (with the reference question being replaced by several questions concerning this phenomena). After a factor analysis, there were high loadings for verbal, non-verbal, public media and inanimate which was taken to represent aspects of communication, while the second was loaded with gossip and being watched and was taken to represent being observed (Startup 2005). There are scales that have been developed to measure levels of ideas and delusions of reference. One of these is the Referential Thinking Scale which may tell apart patients with delusions of reference from health controls and patients without delusions of reference (Startup 2010).

An indication of the prevalence of ideas / delusions of reference comes from a study looking at a Hong Kong Chinese group of patients, with a point prevalence of 31.4% and prevalence of 65.7%. it was associated with auditory hallucinations, persecutory delusions, circumstantiality, impersistence, being unable to feel close to others and depression. Some complaints were being followed, or looked at, etc. and was also found in those in clinical remission (Wong 2008). 

In delusions of communication, the communication relates to the self but the patient attributes origin to some external agency. Hence, it might be that there are abnormalities in self-monitoring of non-verbal aspects of the self, e.g., they may be more likely to misinterpret incidental movements as gestures. To investigate this in more detail psychotic patients were compared to normal controls and they were asked to view clips of people on a computer screen and to determine whether movements made by the people in the clips were deliberate or incidental. Those with delusions of communication tended to make more errors than those without such delusions. They also were more likely to view gestures as signs of rejection, to view incidental movements as gestures (Bucci 2008) and more likely to detect gestures (Bucci 2008b).

Now the dominant theoretical framework for delusions of reference so far is that of Frith’s theory of mind model, but deficits here cannot account for delusions of reference, especially when we consider that delusions of persecution tend to be related to delusions of observation, rather than communication. A relatively new theory describes the genesis of delusions of reference as follows (Startup 2009).

1. Inappropriate attention is attached to irrelevant stimuli.
2. Non-verbal systems are activated.
3. Non-verbal signals are detected.
4. There is a search for meaning.
5. Formation of delusion.
6. Previous non-verbal communication is remembered.


Patients with social anxiety may have more ideas of reference (Wong 2008b). In terms of psychological styles, there may also be a link with internal attribution (Auckenthaler 2008). Also bringing in prediction error and the abnormal attribution of salience to environmental stimuli, those with delusions of reference may be more susceptible to having a higher signal to noise ratio, i.e., more likely to detect spurious signals that controls (Wong 2012). Such patients may also display a greater ability to detect their names spoken out aloud, presented with other distracting stimuli (Lawson 2016). 

Now when it comes to cognitive biases, we have evidence that there might be what is called a jumping to conclusions style in those with delusions. But most of these studies were done on the paranoid and there is evidence that such a deficit may not be found in those with delusions of reference. When 18 patients with schizophrenia all with prominent delusions of reference were compared to healthy controls on a jumping to conclusions task, a theory of mind task and attribution task, there was NO difference between the groups, i.e., there was no difference in the information required to come to a conclusion, there was no evidence of a personalising bias, but patients might have shown less externalising bias (Menon 2013). 

In studies done on the neurological substrates of referential activities, it seems that cortical midline structures (e.g., ventro / dorsomedial prefrontal cortex, anterior / posterior cingulate and precuneus) are involve. A ventral pathway (ventromedial prefrontal – ACC – subcortex) might be involved in attaching salience to stimuli, while a dorsal (dorsomedial prefrontal – cortical – subcortical pathway) may be involved in introspective activities and can interact with the more reflexive ventral pathway. This was looked at in patients with delusions of reference and compared with controls in terms of deciding on whether a sentence was about them or not. Those with delusions of reference were more likely to feel that the sentences were about them, especially sentences that did not refer to them in some way and they took longer to respond to stimuli that were not about felt to be self-referent. Again, there seemed to be activity in the central midline structures, with activity in the insula and ventral striatum being positively correlated with intensity of delusions of reference (Menon 2011).

Emotional / salience networks might in involved (Girard 2017). When schizophrenics with or without delusions of reference were compared to controls in terms of whether statements referred to them or not, found a posterior cortical midline structure was deactivated in controls, and an anterior cortical midline network that was hyperactive in delusional patients [the medial prefrontal].

bottom of page