Skip to Content

Belief Formation Program

The Belief Formation Program aims to advance understanding of the disorders that are associated with higher-order cognition, i.e., delusions and other psychiatric symptoms. To meet this aim, we use a range of methodologies (e.g., hypnosis, cognitive neuropsychiatry, experimental psychology) and encourage cross-disciplinary perspectives that bring together cognitive scientists, philosophers and psychiatrists. We investigate the cognitive and brain systems underlying our ability to formulate and test beliefs about the world, with special reference to disorders of belief formation, such as delusion in schizophrenia and in dementia. Our approach to delusional belief is based on the two-deficit account that has been developed over the past decade by Coltheart, Langdon and colleagues. According to this account, for delusional beliefs to arise, the delusional person must be suffering from two distinct neuropsychological deficits. The first is an impairment of perceptual or cognitive processing which leads to the delusional idea first occurring to the person. The second factor is an impairment of the belief evaluation process which we normally use to reject bizarre or unsubstantiated beliefs, but which the deluded person fails to do. We collaborate with clinicians who use cognitive remediation, in the design, execution and evaluation of psychological treatments for different psychiatric symptoms.

Program Leader

Associate Professor Robyn Langdon

Program Leader Associate Professor Robyn Langdon



The two-factor theory of delusional belief

Robyn Langdon, Max Coltheart, Hannah Morgan, Emily Connaughton, Robert Ross, Anne Jäeger, Naomi Oliver, Jakob Hohwy (Monash University), Philip Gerrans (The University of Adelaide), Dominic Murphy (The University of Sydney), Matthew Large (University of New South Wales) and Olav Nielssen (The University of Sydney)

For the past 15 years, researchers in the Belief Formation Program have been developing a cognitive-neuropsychiatric theory intended to explain how various types of delusional belief arise. According to their theory, any delusion can be explained by discovering the answers to two questions. First: what caused the delusional idea to occur in the first place? Second: why does the patient nevertheless persist in believing that the delusional idea is true when the patient's family, friends and clinicians insist that it is false? The two-factor theory has been applied successfully to the explanation of a number of delusions, including Capgras delusion, which is the delusional belief that a loved one, typically a spouse, has been replaced by a look-alike stranger; and mirrored-self misidentification, which is the belief that the person you see when you look in the mirror is not yourself, but some stranger who looks like you. These are monothematic delusions; that is, delusions concerning a single idea. However, many delusional patients are delusional about many different ideas: this condition is known as polythematic delusion, and we have begun work on whether our two-factor theory of delusional belief, hitherto applied only to monothematic delusions, might be used to explain how polythematic delusional conditions arise.

Hypnosis research

Rochelle Cox, Amanda Barnier, Michael Connors, Vince Polito, Robyn Langdon, Max Coltheart, Emily Connaughton, Martha Turner (University College London, UK), Nora Breen (Royal Prince Alfred Hospital) and Lisa Bortolotti (University of Birmingham, UK)

This research uses hypnosis to model pathological symptoms such as delusions and hallucinations. These symptoms occur in neuropsychological and psychiatric conditions, but they are difficult to investigate experimentally because they often occur with other impairments. To overcome this challenge, researchers at the CCD Hypnosis Laboratory have been developing hypnotic models of delusions and hallucinations. Hypnotic suggestions can create compelling, unusual experiences that are believed with conviction, similar to clinical cases. As such, hypnosis is a useful technique for producing temporary, reversible 'virtual patients'.

We were interested in clinical reports suggesting that delusions may wax and wane under different circumstances. We used our hypnotic analogue of mirrored-self misidentification delusion to investigate whether we could switch this delusion on and off. The success of this study highlights the ability of hypnosis to model features of delusions that are otherwise difficult to investigate.

We also are investigating olfactory hallucinations, which involve smelling odours that are not present. We gave high and low hypnotisable individuals a hypnotic suggestion to experience a sniffing hallucination and tested whether they smelled any odours as a consequence. We found that high hypnotisable subjects not only experienced sniffing hallucinations but many of them detected odours during these hallucinations. These findings confirm that hypnosis can be useful for studying less prevalent psychiatric symptoms.

Prediction error processing and delusion formation

Mike Le Pelley (The University of New South Wales), Melissa Green, Richard Morris (The University of Sydney) and Simon Killcross (The University of New South Wales)

Many of our beliefs regarding the world are formed as a result of learning about the relationships that we experience between events in our environment. For example, an illness incurred after eating watermelon might result in the belief that watermelon caused the illness, or a traumatic event experienced during driving might result in a belief that driving is dangerous. Given that associative learning is an important source of our beliefs about the world, it seems possible that a dysfunction of fundamental associative learning processes might contribute to the formation and maintenance of unusual beliefs (delusions) that are a characteristic feature of psychosis, most notably as a "positive" symptom of schizophrenia. This idea has a long history and, as a result of advances in techniques for studying associative learning processes, it has recently come under concerted empirical scrutiny. We used newly developed procedures to provide an empirical test of the relationship between delusional beliefs and prediction error signalling.

First we investigated individual differences in prediction errors by healthy people as a function of variations in levels of schizotypal traits (measured using self-report questionnaires). We also validated new tasks. Second we piloted these tasks with schizophrenia patient groups and healthy controls recruited via the Australian Schizophrenia Research Bank.

A novel social cognitive training program for people with schizophrenia

Pamela Marsh, Robyn Langdon, Vince Polito, Max Coltheart, Melissa Green and Anthony Harris (Westmead Hospital)

'SoCog' is a novel psychosocial group treatment to help people with schizophrenia overcome their profound social difficulties. Social cognition refers to the abilities that sustain our understanding of the actions, intentions, thoughts and feelings of other people. These abilities underpin successful social interactions, which are reliant on social cognitive abilities to understand what others might be thinking or feeling. Social impairments are identified by people with schizophrenia, carers, and clinicians as a significant unmet treatment need. SoCog is comprised of two programs that use a suite of games and activities focusing on the specific social cognitive problems experienced by people with schizophrenia. One program is emotion recognition training, to improve the recognition of others' facial expressions of emotion. The second is mental-state reasoning training, to encourage flexible thinking about others' likely thoughts, the tolerance of ambiguity, and thoughtful consideration of other people's perspectives. We also were awarded funding from Schizophrenia Fellowship NSW (2013-2014) to develop an internet version of SoCog (eSoCog) to supplement our original SoCog Program. This internet-administered treatment program has the potential to reach more people, including people in rural locations, with cheaper administration costs than face-to-face treatments, and to serve as an important bridge between early intervention services for young people, which typically last for 18-24 months, into standard care thus preventing loss of treatment benefits and increasing the likelihood of ongoing engagement with mental health services.

Aberrant beliefs, including those involving unusual experiences of one's own body

Glenn Carruthers, Max Coltheart, Regine Zopf (Macquarie University), Xioaqing Gao (York University, Canada), Rachel Robbins (University of Western Sydney) and Kristina Musholt (University of Magdeburg, Germany)

This research focuses on beliefs about and experiences of one's own body, and reports of spontaneous bizarre experiences such as out of body experiences. The study uses a well-known technique (the rubber hand illusion) in which an artificial hand is touched in synchrony with one's own hand, such that people come to experience the artificial hand as if it were their own real hand. In recent work we have focussed on explaining the occurrence of this odd experience in terms of changes in the perception of similarity between the artificial hand and one's real hand. This experience of something being a part of one's body is also at play in a variety of cognitive disorders. For example, a delusion called somatoparaphrenia arises when stroke patients come to believe that a part of their body (e.g., their left arm), is not their own arm but, in fact, someone else's. Our work on out of body experiences has highlighted difficulties in distinguishing subjects' experience from confabulations (i.e., narratives told in an attempt to make sense of the bizarre experience). Although it seems as though it should be possible to distinguish experience and confabulation, it is impossible to do so. This surprising state of affairs, we argue, is predicted by a particular theory of consciousness known as 'functionalism'.

Further Information

Contact Details

Telephone: +61 2 9850 4127
Email :
Web :

Sign up to receive CCD event and research announcements

Upcoming CCD Seminars

Full Seminar List

Macquarie Univeristy Logo University of New South Wales Logo University of Western Australia Logo