The belief that others intend to harm us

“Describe the evidence supporting the cognitive model of persecutory paranoia.”

Now for the final essay in this series. This essay felt a little easier than the others. Partly because, by this point, my technique had improved. My reading and note-taking were ruthlessly prioritised. Instead of drowning in notes, I had simple spider diagrams. But there was more to it than that.

Reading for this essay was, genuinely, a humbling experience. Learning about the experiences of people suffering from severe persecutory paranoia certainly gave me some perspective on my usual feelings of essay-anxiety. Rather than letting it take over, I was able to remind myself that the fears I had (not finishing on time, or producing something terrible) were not only unrealistic (I feel like that every time, and things tend to turn out ok), but also just not that bad. If those things actually happened, my relatively comfortable and happy life would go on.

I’ve not yet had a mark back for this essay, but my sense of perspective persists. Regardless of how it turns out, I’m glad I had the opportunity to learn more about this topic. So this essay, at least, feels worth sharing.


Essay Summary

(If you fancy jumping straight into the real thing, just keep scrolling…)

  • Persecutory paranoia exists as a spectrum across the general population. Persecutory paranoia is the incorrect idea that others intend to harm us. Many of us experience this from time to time. Perhaps we imagine our colleagues are being deliberately irritating, or that people on public transport are trying to p*ss us off. At the more extreme end of the spectrum, some individuals hold strong beliefs that someone intends to cause them serious harm, or even to kill them. As you can imagine, this can be extremely debilitating.
  • To help those suffering with these beliefs, it’s important to understand how they begin, and how they’re maintained. Cognitive models have been described in an attempt to do this. A key model is the threat anticipation model of persecutory delusions.
  • The model suggests that persecutory beliefs develop in an attempt to explain strange, or ‘anomalous’, events. Certain emotional processes, such as those involved in anxiety, and reasoning biases, such as the tendency to jump to conclusions, can make the chosen explanation more likely to be of a threatening nature, and strengthen the level of belief in this explanation.
  • The essay discusses the specific emotional processes and reasoning biases that seem to be involved in the formation and maintenance of persecutory paranoia, and reviews the evidence for them having causal roles.
  • Some exciting new research is helping to pinpoint the specific factors that cause the formation and maintenance of extreme persecutory paranoia, in order to develop more targeted treatments.
  • Overall, the message is clear: more research is needed to continue this progress, and bring relief to many.

If you’re interested in paranoia and anxiety, you might be interested in these:

Read on for the full essay…

Describe the evidence supporting the cognitive model of persecutory paranoia.


Humans make daily judgements of whether to trust other people. The decision not to trust can protect us from harm. But these judgements are often based on limited information, and errors are made. Holding incorrect ideas that another person is deliberately trying to cause us harm is termed persecutory ideation. The frequency, content and strength of conviction of these persecutory ideas – the level of persecutory paranoia – forms a spectrum across the general population (Freeman et al., 2005; Bebbington et al., 2013). At the lower end of the spectrum, individuals might, for example, experience fleeting thoughts that others are trying to upset them. At the higher end, individuals might hold strong beliefs of severe threat from others, for example that others intend to kill them. Experiences at this end of the spectrum are termed persecutory delusions, and are one of the most common psychotic experiences (Sartorius et al., 1986), associated with extremely low psychological wellbeing (Freeman et al., 2014a). This essay will discuss the evidence supporting the major cognitive model of persecutory paranoia: the threat anticipation model (Freeman, Garety, Kuipers, Fowler, & Bebbington, 2002).

Screenshot 2018-06-06 12.01.59
The paranoia hierarchy. All except ‘ideas of reference’ are examples of the various levels of persecutory paranoia found across the general population. From Freeman, 2007.

The threat anticipation cognitive model of persecutory paranoia

At the heart of the model is the conceptualisation of persecutory paranoia in terms of threat beliefs (Freeman, 2007). The model proposes that threat beliefs form in an attempt to explain anomalous events (Maher, 1988), and that the explanations considered and chosen are influenced by multiple factors (Freeman et al., 2002). Anomalous experiences, emotional processes and reasoning biases are proposed to interact in the formation of persecutory delusions, and to support mechanisms of belief maintenance: confirmatory evidence being obtained and disconfirmatory evidence being discarded (Freeman et al., 2002).

The threat anticipation model of persecutory delusion formation (left) and maintenance (right). From Freeman, 2007.

Anomalous experiences

The cognitive model proposes that persecutory delusions arise from the need to explain anomalous experiences, and that anomalous experiences are perceived as confirmatory evidence of the delusion (Freeman et al., 2002). Such experiences are suggested to include clearly anomalous internal events such as hallucinations, subtler perceptual abnormalities such as sounds seeming louder, and feelings of heightened physiological arousal (Freeman, 2007).

Many patients cite anomalous internal experiences as evidence for their delusional beliefs (e.g. Buchanan et al., 1993), and those who do so are less able to provide alternative, non-delusionary, explanations for this evidence, than those who cite other types of evidence (Freeman et al., 2004). A large-scale longitudinal study in a non-clinical population found the risk of future psychosis was increased if delusional ideas followed hallucinations, but not if hallucinations followed delusional ideas (Krabbendam et al., 2004), suggesting explanations of anomalous experiences are important in delusion development.

The main psychoactive ingredient of cannabis, THC, is a known risk factor for developing psychosis. This seems to be due, at least in part, to its role in increasing negative emotions and anomalous experiences.

However, these studies do not distinguish between persecutory delusions and other sub-types (such as delusions of reference and grandiose delusions), and do not explicitly demonstrate a causal role. More recently, a randomised controlled trial (RCT) suggested a causal role for anomalous experiences and negative affect in persecutory delusions. Freeman and colleagues (2014b) demonstrated that the main psychoactive ingredient of cannabis, a known risk factor for psychosis, increased negative affect and anomalous experiences, which together accounted for increases in persecutory paranoia. More manipulation studies are needed to support a causal role of anomalous experiences in persecutory paranoia.

Emotional processes

The findings of Freeman et al., (2014b) are consistent with the model’s assertion that triggers of psychosis, such as illicit drugs, can lead to anomalous experiences directly, or indirectly through activation of emotional processes or cognitive biases (Freeman et al., 2002). The model proposes that emotional processes also influence the nature of the explanation selected (Freeman, 2007).

For example, persecutory delusions are proposed to reflect negative beliefs about the self, such as being vulnerable (Freeman et al., 2002; c.f. Bentall, Kinderman, & Kaney, 1994). Negative self-beliefs are prevalent in individuals with persecutory paranoia (e.g. Kesting & Lincoln, 2013), and evidence for a causal role of negative self-beliefs is growing (Freeman, 2016). A recent study experimentally manipulated self-confidence in paranoid individuals, and demonstrated that reducing self-confidence increased persecutory thoughts (Atherton et al., 2016). This is consistent with other manipulation studies in vulnerable individuals (e.g. Freeman et al., 2014c; Lincoln, Hohenhaus, & Hartmann, 2013).

Worry is proposed to be a key emotion involved in persecutory delusions, since it involves a persistent concern with threat (Matthews 1990; Freeman et al., 2002). Individuals with persecutory delusions have a worry thinking style (Freeman & Garety, 1999), and associations with specific worry mechanisms, such as intolerance of uncertainty, have recently been shown (Startup et al., 2016). Such evidence has prompted development of interventions, the effects of which support a causal role of worry. In an RCT with 150 individuals with persecutory delusions, a form of cognitive behavioural therapy (CBT) that reduced worry significantly reduced delusions compared to a standard care control, which was maintained after 6 months (Freeman et al., 2015).

Safety behaviours are actions performed to avoid the threat from being realised. This can include avoiding specific situations, for example by not leaving the house. Or it can include using certain behaviours within a situation, such as being aggressive to those who are believed to intend harm.

Anxiety is proposed to maintain persecutory paranoia through encouraging safety behaviours – actions that individuals perform to avoid the threat being realised (Salkovskis, 1996; Freeman et al., 2002). Safety behaviours are suggested to maintain threat beliefs by limiting access to disconfirmatory evidence, such that individuals attribute their lack of harm to these behaviours, rather than to inaccuracy of their threat belief (Freeman, 2007). One study found 96% of patients with persecutory delusions had performed safety behaviours during the previous month (Freeman et al., 2007), and evidence for a causal role is provided by a recent small-scale RCT, in which patients with persecutory delusions were exposed to their threatening situation in virtual reality (Freeman et al., 2016). Dropping the use of safety behaviours accounted for a 22% reduction in delusional conviction compared to exposure alone.

Reasoning biases

The cognitive model hypothesises that the explanations considered by an individual for their anomalous experiences are also influenced by reasoning biases (Freeman et al., 2002). One such bias is jumping to conclusions (JTC), in which individuals gather insufficient data before drawing conclusions (Garety, Hemsley, & Wessely, 1991). Many studies demonstrate the bias in patients with delusions (e.g. Falcone et al., 2015; Garety et al., 2013), and a meta-analysis of 71 studies link the bias to increased probability of delusions (Dudley, Taylor, Wickham, & Hutton, 2015). However, such studies rarely discriminate between delusion sub-types, therefore provide only limited support for a role of JTC in persecutory delusions, specifically. Specific prevalence, however, has been shown by a study which found the bias was significantly more prevalent in individuals experiencing persecutory delusions (50%) compared to a non-clinical group (10%), and that this could not be explained by co-existence of other delusion sub-types (Startup, Freeman, & Garety, 2008).

Early articulations of the model suggested reduced ability to infer the mental states of others (ToM), associated with schizophrenia (Frith, 1992), might contribute to persecutory delusion formation (Freeman et al., 2002). However, this account has theoretical shortcomings (Freeman, 2007; Walston, Blennerhassett, & Charlton, 2000), and empirical support for associations between ToM and paranoia is mixed (Craig, Hatton, Craig, & Bentall, 2004; Greig, Bryson, & Bell, 2004). Methodological limitations could contribute to this lack of replication (Greig et al., 2004), but ToM difficulties seem to be more closely associated with negative psychotic symptoms than with persecutory delusions (Brüne, 2005).

An attributional bias for blaming others for negative events was also suggested as a candidate factor in threat belief formation (Freeman et al., 2002). Again, there is mixed empirical support, with some studies finding an association with persecutory delusions (Fear, Sharp, & Healy, 1996) and others failing to (Martin & Penn, 2002). Methodological limitations, such as failure to control for grandiosity and depression (Freeman, 2007), which often co-occur with persecutory delusions and are thought to be associated with attributional style (Jolley et al., 2006), suggest further research into this bias may be warranted.

In recent articulations of the model, belief inflexibility, the failure to consider alternative explanations, is highlighted as an important process in the maintenance of threat beliefs (Freeman, 2016). This is consistent with evidence from several studies. In terms of prevalence, 75% of individuals experiencing delusions, the majority of which were persecutory, could not report any possible alternative explanation for their experience (Freeman et al., 2004). A recent proof-of-concept RCT in individuals with persistent persecutory delusions employed an intervention targeted at reducing belief inflexibility and JTC, which significantly improved state paranoia, belief flexibility and distress measures, compared to a usual care control (Waller et al., 2015). This is consistent with the finding that belief inflexibility mediates the relationship between JTC and delusional conviction (Garety et al., 2005). The model’s hypothesis that reasoning biases play a role in persecutory delusion formation and maintenance appears to be valid.

“It was quite simple. I learnt to slow down and think carefully about the situation. In the future I will be very hesitant about coming to a fixed conclusion”

Feedback from a participant of the ‘Thinking Well’ trial, which was targeted at reducing belief inflexibility and JTC. From Waller et al., 2015.


Despite numerous limitations, substantial empirical evidence supports the core tenet of the cognitive model of persecutory paranoia: that multiple factors contribute to delusion formation and maintenance, including anomalous experiences, emotional processes and reasoning biases. Efforts to isolate delusion sub-types, experimentally manipulate potential factors, and test proof-of-concept interventions have strengthened the evidence for specific causal factors such as internal anomalous experiences, worry, anxiety, JTC and belief inflexibility, offering promise for improved understanding of this complex phenomenon, and life-changing relief for many.



One thought on “The belief that others intend to harm us

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s