Understanding the Causes of Hallucinations in Psychosis
Research type
Research Study
Full title
Aberrant Salience of Thoughts as a Causal Factor of Hallucinations in Psychosis
IRAS ID
328306
Contact name
Liam Myles
Contact email
Sponsor organisation
University of Oxford, Research Governance, Ethics and Assurance
Clinicaltrials.gov Identifier
NA, NA
Duration of Study in the UK
1 years, 10 months, 1 days
Research summary
Summary of Research
The causes of hallucinations in people with psychosis is unclear. The Salience Hypothesis argues that people with psychosis have very attention-grabbing thoughts. This theory has garnered support from studies of ‘latent inhibition.’ Latent inhibition refers to the finding that people without psychosis will reduce their attention to a stimulus after being repeatedly exposed to it. In contrast, people with psychosis direct their attention towards stimuli for longer than healthy controls. However, research studies have exclusively examined attention to stimuli in the external environment, such as pictures on a computer screen. The aim of this study is to evaluate the hypothesis that people with psychosis who hallucinate pay more attention to their thoughts.
Patients with psychosis who hallucinate (hallucinators) and who do not hallucinate (non-hallucinators) will be recruited, in addition to people without psychosis (healthy controls). Participants will be trained to associate two stimuli together (stimuli ‘A’ and ‘X’), resulting in them learning that A is linked to X. This training should result in presentation of one stimulus (e.g., A) making participants think of the other (X). Following this, they will receive repeated exposure to stimulus A alone; repeatedly presenting stimuli reduces the extent to which they grab attention. Therefore, healthy controls should pay less attention to A and, critically, the thought of X, as participants will also be repeatedly exposed to the thought of X when A is presented. During the test phase, the attention that participants are paying to X will be examined. If more attention is paid to X, they will learn that X predicts a novel stimulus more quickly, and vice versa.
As the Salience Hypothesis predicts that hallucinators will maintain greater attention to the thought of X, we predict that hallucinators will associate ‘X’ with a novel stimulus more rapidly than do healthy controls and non-hallucinators.
Summary of Results
Psychosis is a severe mental health difficulty that affects approximately 3% of people (Kirkbride et al., 2012; Moreno-Küstner et al., 2018; Morgan et al., 2014; Perälä et al., 2007). It involves different types of symptoms, which are grouped into so-called ‘positive,’ ‘negative’ and ‘disorganised’ symptoms (APA, 2013). Positive symptoms refer to hallucinations and unusual beliefs. Negative symptoms include social withdrawal and reduced motivation, emotional affect, pleasure and speech. Disorganised symptoms pertain to problems with attention, decision making, thinking, memory and planning.The salience hypothesis is an influential theory about the causes of psychosis (Howes & Kapur, 2009; Kapur, 2003; Kapur et al., 2005). This theory suggests that people with psychosis pay higher attention than people without psychosis to stimuli of otherwise low significance. It suggests that unusual beliefs arise because people with psychosis try to make sense of how these salient stimuli grab their attention. For example, if an individual repeatedly notices a car parked outside their home, they may conclude they are being monitored. The salience hypothesis also suggests that people with psychosis hallucinate because they also pay higher attention to their thoughts than people without psychosis. For example, an individual with psychosis may pay so much attention to thoughts about a parent’s critical comments that these thoughts may be perceived as an external voice.
However, there is mixed evidence for the salience hypothesis. A recent review found conflicting evidence for the theory that people with psychosis pay relatively high attention to stimuli of otherwise low significance (Myles et al., 2023). Also, there is very limited evidence that people with psychosis hallucinate because they pay relatively high attention to their own thoughts. Therefore, this study investigated two questions:
1) Do people with psychosis who hallucinate pay more attention to their thoughts?
2) Do people with people with psychosis, irrespective of whether they hallucinate, pay more attention to stimuli of otherwise low significance?To answer these questions, this study recruited 111 adults. These were 37 people without a diagnosis/history of mental health difficulties (‘controls’), 37 people with psychosis and ongoing hallucinations (‘hallucinators’) and 37 people with psychosis without hallucinations (‘non-hallucinators’).
Participants were asked to provide demographic information and were asked about how frequently they hallucinate. Next, participants completed a questionnaire assessing the extent to which they experience paranoia, hallucinations, cognitive disorganisation, grandiosity and anhedonia. Then, participants completed an experiment which examined how much attention they paid to environmental stimuli and their own thoughts. In this experiment, participants were presented with fictitious words, which we termed ‘pseudowords.’ They were trained to associated two pseudowords together (e.g., ‘lonc’ and ‘cazs’), such that presentation of one pseudoword (i.e., ‘lonc’) would make them think of the other (i.e., ‘casz’). Participants were then repeatedly presented with one of these pseudowords (i.e., ‘lonc’). Repeatedly presenting participants with ‘lonc’ should reduce their attention to both ‘lonc’ and the thought of the pseudoword with which it is associated (i.e., ‘cazs’). Finally, to evaluate how much attention participants were paying to each of the pseudowords, they were required to learn that presentation of each pseudoword predicted the subsequent presentation of a novel pseudoword (i.e., ‘dax’). Participants were expected to learn the relationship between the pseudowords and ‘dax’ faster if they were paying more attention to the pseudowords.The results suggested that controls, hallucinators and non-hallucinators did not differ in the extent to which they paid attention to environmental stimuli or their own thoughts. These findings contradict the salience hypothesis in two ways. First, they suggest that unusual believes are not caused by paying relatively high attention to stimuli of otherwise low significance. Second, they suggest that hallucinations are not caused by paying relatively high attention to one’s own thoughts.
However, this study had some limitations. These included methodological issues, which could mean that the experiment did not accurately measure how much attention participants paid to environmental stimuli and/or their thoughts. Also, this study had a relatively small sample, and some participants had unconfirmed psychosis diagnoses. This may limit how well the findings apply to other individuals with psychosis. Furthermore, it is possible that participants’ use of medications may have made it harder to determine whether psychosis is linked with differences in the way clients pay attention. Finally, this study had limited ethnic diversity, which may affect how well the findings apply to people from different cultural backgrounds.
It is too early to derive clinical implications, as further research is needed to examine the salience hypothesis. For example, studies should investigate whether hallucinations stem from paying large amounts of attention to particular types of internal experiences, such as emotive thoughts or memories. Also, future research should adapt this experiment to overcome the aforementioned limitations.In conclusion, these findings contradict the salience hypothesis. Indeed, the results suggest that unusual beliefs are not caused by paying relatively high attention to stimuli of otherwise low significance and that hallucinations are not caused by paying relatively high attention to one’s own thoughts.
REC name
South East Scotland REC 02
REC reference
24/SS/0016
Date of REC Opinion
22 Mar 2024
REC opinion
Further Information Favourable Opinion