How Cannabis Causes Paranoia: Using the Intravenous Administration of Δ9-Tetrahydrocannabinol (THC) to Identify Key Cognitive Mechanisms Leading to Paranoia

Daniel Freeman; Graham Dunn; Robin M. Murray; Nicole Evans; Rachel Lister; Angus Antley; Mel Slater; Beata Godlewska; Robert Cornish; Jonathan Williams; Martina Di Simplicio; Artemis Igoumenou; Rudolf Brenneisen; Elizabeth M. Tunbridge; Paul J. Harrison; Catherine J. Harmer; Philip Cowen; Paul D. Morrison


Schizophr Bull. 2015;41(2):391-399. 

In This Article

Abstract and Introduction


Paranoia is receiving increasing attention in its own right, since it is a central experience of psychotic disorders and a marker of the health of a society. Paranoia is associated with use of the most commonly taken illicit drug, cannabis. The objective was to determine whether the principal psychoactive ingredient of cannabis—Δ9-tetrahydrocannabinol (THC)—causes paranoia and to use the drug as a probe to identify key cognitive mechanisms underlying paranoia. A randomized, placebo-controlled, between-groups test of the effects of intravenous THC was conducted. A total of 121 individuals with paranoid ideation were randomized to receive placebo, THC, or THC preceded by a cognitive awareness condition. Paranoia was assessed extensively via a real social situation, an immersive virtual reality experiment, and standard self-report and interviewer measures. Putative causal factors were assessed. Principal components analysis was used to create a composite paranoia score and composite causal variables to be tested in a mediation analysis. THC significantly increased paranoia, negative affect (anxiety, worry, depression, negative thoughts about the self), and a range of anomalous experiences, and reduced working memory capacity. The increase in negative affect and in anomalous experiences fully accounted for the increase in paranoia. Working memory changes did not lead to paranoia. Making participants aware of the effects of THC had little impact. In this largest study of intravenous THC, it was definitively demonstrated that the drug triggers paranoid thoughts in vulnerable individuals. The most likely mechanism of action causing paranoia was the generation of negative affect and anomalous experiences.


Paranoia—unfounded fears that others intend harm to the individual—is a central experience of psychotic disorders such as schizophrenia. Factor analytic studies indicate that it is an independent experience that requires explanation in its own right.[1] Many people have a few paranoid ideas, and a few people have many.[2] Paranoia is associated with youth, poverty, poor physical health, suicidal ideation, and the use of cannabis.[3] We took this latter association to carry out a unique experimental investigation into the causes of paranoia. First, we set out to establish that cannabis causes paranoia, using the most comprehensive battery ever used to assess such fears. Second, cannabis was used as a probe to identify the key cognitive mechanisms causing paranoid fears. Determination of the immediate causes can be used to advance the treatment of delusions.[4]

A Cognitive Account of Paranoia

Our cognitive account of paranoia identifies multiple causes (see figure 1).[5] It is hypothesized that the individual experiences an anomalous internal state. These anomalies may include, across the different senses, changes in sensory intensity, distorted sensory experience, sensory flooding, unusual sensory experience, thought echo, and hallucinations.[6] They may also include feelings of unusual salience.[7] The anomalies may be triggered by, eg, life events, poor sleep, or illicit drugs such as cannabis. In essence, the person feels different and this requires an explanation. Odd experiences encourage unusual thoughts. Importantly, a negative affective state makes a paranoid interpretation likely: anxiety leads to the threat content; negative self beliefs highlight the person's vulnerability to harm; and engagement in worry results in negative, implausible ideas. The fears reach a delusional level of conviction when reasoning biases, such as jumping to conclusions, are present. Working memory performance moderates the effect of the reasoning biases.[8,9]

Figure 1.

Outline of factors involved in persecutory delusion development.5

Using Δ9-Tetrahydrocannabinol to Determine the Causes of Paranoia

Establishment of causal roles requires manipulation of the factors of interest.[10] We saw the administration of cannabis as a method of manipulating key putative causal factors in paranoia, while also providing an important opportunity to learn about a substance seen by many as a contributory cause of psychosis (eg, Casadio et al[11]).

The principal psychoactive constituent of cannabis is Δ9-tetrahydrocannabinol (THC). THC administered intravenously is characterized by the appearance of psychopharmocological effects within 5 minutes, which continue for at least 90 minutes, providing an excellent experimental window.[12] In within-subjects tests with nonclinical volunteers, D'Souza, in particular, has shown that intravenous administration of THC causes schizophrenia-like symptoms, perceptual disturbances, anxiety, and impaired working memory (eg, Morrison et al[13] and D'Souza et al[14,15]). Similar but more pronounced results were found in patients with schizophrenia.[15]

Principally, the psychosis-inducing effects of cannabis have been linked to an abnormal salience theory,[16,17] in which THC leads to an increase in anomalous experience which is then misinterpreted. However, THC can also induce anxiety via cannabinoid receptors in the amygdala.[18] Therefore, the prediction from the cognitive model of paranoia is that THC leads to such fears via 2 routes: the generation of confusing anomalous experiences and an increase in negative affect and related processes. Working memory is clearly affected by THC, but since such cognitive problems have not been linked to the positive symptoms of psychosis,[19] a direct memory performance route to paranoia would not be expected.

The Current Study

In the current study, we set out to identify the causes of paranoia via the administration of THC. This requires a between-groups test of the administration of THC against placebo. A third condition tested a further causal factor: misinterpretation. The cognitive perspective considers paranoia to be a misinterpretation of events. If individuals are made sufficiently aware of the potential subjective effects of THC, then this may alter the interpretation made and hence the occurrence of paranoia. Therefore, the third condition was a cognitive awareness manipulation in which the potential effects of THC were explained before drug administration. The participants to be tested were selected from the general population on the key criterion of having had recent paranoid ideation. This was therefore an analogue population, in order that the results are applicable to understanding the clinical phenomenon; testing individuals without signs of vulnerability to disorder would be much less informative about clinical paranoia, while testing patients in such large numbers would have ethical and practical difficulties. We were mindful of "the paranoia problem," the difficulty of determining whether persecutory ideation is unfounded, so the battery of tests included the most extensive range of paranoia assessments yet used in a study. There were 3 hypotheses: (a) THC increases the occurrence of paranoia, (b) the occurrence of anomalous experiences and negative affect (but not alterations in working memory) explains (ie, mediates) the increase in paranoia, and (c) that paranoid interpretations can be partially blocked by cognitive awareness. The second hypothesis concerning underling mechanisms is the hardest to establish—we note the advice of Bullock and colleagues[20] "to think of mediation analysis as a cumulative enterprise."