Research to date

Early research on psychotic experiences focused on a continuum with, and prediction of, psychotic disorders (Johns & Van Os, Reference Johns and Van Os2001; Poulton et al., Reference Poulton, Caspi, Moffitt, Cannon, Murray and Harrington2000; Van Os, Linscott, Myin-Germeys, Delespaul, & Krabbendam, Reference Van Os, Linscott, Myin-Germeys, Delespaul and Krabbendam2009). As with a depression continuum from mild sadness to pervasively depressed mood, it was proposed that psychotic experiences could be conceptualised as existing on a continuum from minor perceptual abnormalities and overvalued unusual thought content to the fully psychotic hallucinations and delusions of schizophrenia. Meta-analyses suggested 5–7% of the general adult population experienced at least occasional psychotic experiences (Linscott & Van Os, Reference Linscott and Van Os2013; Van Os et al., Reference Van Os, Linscott, Myin-Germeys, Delespaul and Krabbendam2009) and follow-up studies showed that these individuals were at elevated risk of developing psychotic disorders (Poulton et al., Reference Poulton, Caspi, Moffitt, Cannon, Murray and Harrington2000; Welham et al., Reference Welham, Scott, Williams, Najman, Bor, O'callaghan and McGrath2009).

Research that followed highlighted that, as opposed to a homotypic relationship with psychotic disorder, psychotic experiences are predictive of a wide range of mental disorders (Fisher et al., Reference Fisher, Caspi, Poulton, Meier, Houts, Harrington and Moffitt2013; McGrath et al., Reference McGrath, Saha, Al-Hamzawi, Andrade, Benjet, Bromet and Demyttenaere2016; Werbeloff et al., Reference Werbeloff, Drukker, Dohrenwend, Levav, Yoffe, van Os and Weiser2012; Wigman et al., Reference Wigman, van Nierop, Vollebergh, Lieb, Beesdo-Baum, Wittchen and van Os2012). Indeed, although only a small minority of individuals with psychotic experiences go on to develop a psychotic disorder, a high proportion develop other, more common mental disorders, including depressive, anxiety, behavioural and substance use disorders (Healy & Cannon, Reference Healy and Cannon2020; Kelleher et al., Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012a; Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin and Cannonb; Reference Kelleher, Lynch, Harley, Molloy, Roddy, Fitzpatrick and Cannonc; Rimvall et al., Reference Rimvall, van Os, Verhulst, Wolf, Larsen, Clemmensen and Jeppesen2020; Scott, Chant, Andrews, & McGrath, Reference Scott, Chant, Andrews and McGrath2006; Trotta et al., Reference Trotta, Arseneault, Caspi, Moffitt, Danese, Pariante and Fisher2020; Varghese et al., Reference Varghese, Scott, Welham, Bor, Najman, O'Callaghan and McGrath2011; Wigman et al., Reference Wigman, van Nierop, Vollebergh, Lieb, Beesdo-Baum, Wittchen and van Os2012).

A wide variety of factors have been shown to be associated with psychotic experiences, including lower IQ (Cannon et al., Reference Cannon, Caspi, Moffitt, Harrington, Taylor, Murray and Poulton2002; Johns et al., Reference Johns, Cannon, Singleton, Murray, Farrell, Brugha and Meltzer2004), adversity (Croft et al., Reference Croft, Heron, Teufel, Cannon, Wolke, Thompson and Zammit2019; Crush, Arseneault, & Fisher, Reference Crush, Arseneault and Fisher2018; McGrath et al., Reference McGrath, Saha, Lim, Aguilar-Gaxiola, Alonso, Andrade and Cardoso2017; Morgan et al., Reference Morgan, Reininghaus, Reichenberg, Frissa, Hotopf, Hatch and Team2014; Newbury et al., Reference Newbury, Arseneault, Caspi, Moffitt, Odgers and Fisher2018; Oh, Cogburn, Anglin, Lukens, & DeVylder, Reference Oh, Cogburn, Anglin, Lukens and DeVylder2016), poorer coping skills (Lin et al., Reference Lin, Wigman, Nelson, Vollebergh, Van Os, Baksheev and Yung2011; Wigman et al., Reference Wigman, Devlin, Kelleher, Murtagh, Harley, Kehoe and Cannon2014a), migration (Laurens, West, Murray, & Hodgins, Reference Laurens, West, Murray and Hodgins2008; Scott et al., Reference Scott, Chant, Andrews and McGrath2006), ethnicity (El Bouhaddani, van Domburgh, Schaefer, Doreleijers, & Veling, Reference El Bouhaddani, van Domburgh, Schaefer, Doreleijers and Veling2019; Oh, Yang, Anglin, & DeVylder, Reference Oh, Yang, Anglin and DeVylder2014) and socioeconomic factors (Scott et al., Reference Scott, Chant, Andrews and McGrath2006), as well as genetic and molecular variation (Niculescu et al., Reference Niculescu, Levey, Phalen, Le-Niculescu, Dainton, Jain and Weber2015; Pain et al., Reference Pain, Dudbridge, Cardno, Freeman, Lu, Lundstrom and Ronald2018; Ramsay et al., Reference Ramsay, Kelleher, Flannery, Clarke, Lynch, Harley and Cannon2013).

Our own research has highlighted psychotic experiences as indicators of psychopathology severity, demonstrating that, compared to individuals with mental disorders without psychotic experiences, individuals with psychotic experiences have a higher psychiatric symptom burden (Kelleher, Clarke, Rawdon, Murphy, & Cannon, Reference Kelleher, Clarke, Rawdon, Murphy and Cannon2013a), greater multimorbidity (Kelleher, Cederlof, & Lichtenstein, Reference Kelleher, Cederlof and Lichtenstein2014a; Kelleher et al., Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin and Cannon2012b), more suicidal behaviour (Kelleher et al., Reference Kelleher, Cederlof and Lichtenstein2014a; Kelleher et al., Reference Kelleher, Corcoran, Keeley, Wigman, Devlin, Ramsay and Cannon2013b; Kelleher et al., Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin and Cannon2012b), cognitive deficits (Blanchard et al., Reference Blanchard, Jacobson, Clarke, Connor, Kelleher, Garavan and Cannon2010; Kelleher et al., Reference Kelleher, Clarke, Rawdon, Murphy and Cannon2013a) and poorer socio-occupational functioning (Healy et al., Reference Healy, Campbell, Coughlan, Clarke, Kelleher and Cannon2018; Kelleher et al., Reference Kelleher, Wigman, Harley, O'Hanlon, Coughlan, Rawdon and Cannon2015; Wigman et al., Reference Wigman, Devlin, Kelleher, Murtagh, Harley, Kehoe and Cannon2014a). Researchers have also demonstrated that psychotic experiences are associated with high levels of psychological distress (Armando et al., Reference Armando, Nelson, Yung, Ross, Birchwood, Girardi and Nastro2010), more mental health service use (Bhavsar et al., Reference Bhavsar, Maccabe, Hatch, Hotopf, Boydell and McGuire2017) and poorer treatment responses to both pharmacotherapy (Perlis et al., Reference Perlis, Uher, Ostacher, Goldberg, Trivedi, Rush and Fava2011) and psychotherapy (Wigman et al., Reference Wigman, van Os, Abidi, Huibers, Roelofs, Arntz and Peeters2014b).

A broader biopsychosocial framework

We believe it is now timely to consider psychotic experiences in a broader framework of neurobiological and psychosocial functioning, with illness severity representing just one aspect. Specifically, we believe that psychotic experiences can be conceptualised as markers of a clinically significant mental threshold – the threshold being where mental distress transitions to dysfunction in neural systems integration and communication, which we term the neural efficiency threshold model. One of the earliest clinical signs of crossing this threshold, we believe, is the emergence of problems with sensory perception skills and, as a result, hallucinatory experiences.

Supporting evidence of neural systems dysfunction in individuals with psychotic experiences comes from a number of neurocognitive and neuromotor studies. From the perspective of neurocognition, a wide range of deficits have been demonstrated in individuals with formal psychotic disorders (Barch & Ceaser, Reference Barch and Ceaser2012; Bowie & Harvey, Reference Bowie and Harvey2005; Green et al., Reference Green, Bearden, Cannon, Fiske, Hellemann, Horan and Sergi2011; Szöke et al., Reference Szöke, Trandafir, Dupont, Meary, Schürhoff and Leboyer2008). Lesser degrees of cognitive impairment have also been observed in more common mental disorders – notably depression (Rock, Roiser, Riedel, & Blackwell, Reference Rock, Roiser, Riedel and Blackwell2014; Service et al., Reference Service, Upegui, Ramírez, Port, Moore, Umanes and López2020; Weiser et al., Reference Weiser, Reichenberg, Rabinowitz, Knobler, Lubin, Yazvitzky and Davidson2004) – which might suggest that neural systems dysfunction extends more broadly to general psychopathology. However, research on cognitive function in common mental disorders has typically failed to assess or account for co-occurring psychotic experiences. Our model suggests that common mental disorders that do not have co-occurring psychotic experiences would not be associated with significant cognitive impairment but that we would see progressive cognitive effects as psychotic experiences emerge and become more severe.

Preliminary support for this comes from Barnett et al., who looked at childhood cognitive function as it longitudinally related to psychopathology and psychotic experiences (Barnett et al., Reference Barnett, McDougall, Xu, Croudace, Richards and Jones2012). Although initial analyses suggested there may be a relationship between psychopathology and poorer cognitive performance, this relationship was, in fact, entirely accounted for by co-occurring psychotic experiences. That is, individuals with psychopathology who did not have psychotic experiences did not demonstrate poorer cognitive performance.

Similarly, in our own research, we have found that young people with psychotic experiences performed more poorly than their peers on multiple measures of cognitive processing speed, as well as in non-verbal working memory (Kelleher et al., Reference Kelleher, Clarke, Rawdon, Murphy and Cannon2013a), findings that were not accounted for by general psychopathology. Individuals with psychotic experiences have, furthermore, also been shown to score more poorly on tests of intellectual ability (Cannon et al., Reference Cannon, Caspi, Moffitt, Harrington, Taylor, Murray and Poulton2002; Johns et al., Reference Johns, Cannon, Singleton, Murray, Farrell, Brugha and Meltzer2004).

Other evidence of neural systems dysfunction associated with psychotic experiences comes from neuromotor studies. In the Dunedin study, poorer motor skills development in infancy and early childhood was predictive of psychotic experiences at age 11 years (Cannon et al., Reference Cannon, Caspi, Moffitt, Harrington, Taylor, Murray and Poulton2002). We have also found evidence of neuromotor skills deficits in older children with psychotic experiences in terms of poorer scores on the Purdue pegboard task (Blanchard et al., Reference Blanchard, Jacobson, Clarke, Connor, Kelleher, Garavan and Cannon2010). What is more, we found that poorer childhood pegboard motor performance was predictive of psychotic experiences into adulthood (Carey et al., Reference Carey, Dooley, Gillan, Healy, Coughlan, Clarke and Cannon2019).

Because of the high degree of regional specialisation within the brain, complex mental processes are dependent not just upon the efficiency of relevant brain regions but also upon high-speed communication between the different hubs that must cooperate to faithfully execute these processes. This high-speed cooperation is dependent upon the integrity and efficiency of tracts connecting diffuse neural networks, referred to as the connectome. The connectome is the matrix of neurons connecting and facilitating communication between different (specialised) brain regions.

The development of an efficient connectome allowed the evolution of a high degree of regional specialisation in the human brain because it facilitated rapid communication and coordination between the many specialised neural hubs involved in executing complex mental processes. The connectome might be likened to a complex road network, with the major white matter tracts representing the motorways, and a whole series of progressively smaller ‘road networks’ that connect diffuse brain regions. Each ‘road network’ has a limited capacity beyond which ‘traffic’ can no longer efficiently travel, resulting in poorer interaction and coordination between hubs.

Complex mental processes depend on rapid, coordinated transmission of information between multiple hubs. If delays arise in parts of this information transmission process then the complex mental processes carried out by these hubs will be disturbed. For example, information transmission problems in neural circuits involved in sensory perception will mean that smooth integration of sensory perceptual experiences will breakdown, leading to hallucinatory experiences. The greater the latency in information transmission, the more severe the hallucinations. This applies to many complex mental processes that require coordination between neural hubs, such as sensory-perception, motor, cognitive, social and language skills – all processes that are implicated in psychosis. Psychosis, then, might be considered the evolutionary price for developing highly specialised brains – asynchrony or poor integration of neural activity from one hub to the other may give rise to psychotic experiences.

Why are some individuals more sensitive to information transmission latencies than others? Developmental issues and individual differences

The synaptic pruning process that defines adolescent brain development leads to more efficient integration and coordination between brain hubs (‘neurons that fire together, wire together’). We suggest that this is the reason why the prevalence of psychotic experiences markedly decreases over the course of adolescence (Kelleher et al., Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012a): where this maturational process proceeds smoothly, neural efficiency increases, meaning that the threshold for disordered integration and coordination (which precipitates psychotic experiences) is less likely to occur.

Individuals for whom the synaptic pruning process is less successful, however, will have lower thresholds at which their key information transmission systems break down, meaning that psychotic experiences are more likely – the more disordered the synaptic pruning process, the lower the neural efficiency threshold.

As well as between person differences, there may also be within person differences in terms of vulnerability to deficits in one domain over another. For example, an individual may have relatively poor connectivity in certain brain regions, such as in brain hubs involved in auditory sensory processing, but relatively better connectivity in other regions, such as in brain hubs involved in language or in limb movements, meaning that deficits will be more marked in terms of auditory hallucinations, with fewer problems with disorganised speech or behaviour. Individual variation in risk for psychotic experiences may be dependent upon an interaction between the integrity of the connectome and the levels of demand or mental stress placed upon the individual (and, therein, their available neural resources), with each individual having their own personal threshold, beyond which communication cannot continue efficiently, leading to the emergence of psychotic experiences. That is, the more efficient the connectome, the greater the demand needed to overwhelm neural resources and cause psychotic experiences.

Connectome efficiency (or the speed of transfer of information facilitated by the connectome), we believe, may also be the reason why psychotic experiences are far more prevalent in childhood than in adolescence and adulthood. Adolescence represents the key developmental stage during which consolidation of the hubs of the connectome occurs (Whitaker et al., Reference Whitaker, Vértes, Romero-Garcia, Váša, Moutoussis, Prabhu and Rittman2016). A more efficient connectome, resulting from a healthy consolidation process, will, we suggest, result in reduced vulnerability to psychotic experiences.

We suggest that the relative hypoconnectivity of the pre-adolescent brain may be the reason that psychotic experiences are for more prevalent in children. We hypothesise that, given their stage of brain development, even relatively low levels of demand may overwhelm neural resources in otherwise healthily developing younger children and precipitate psychotic experiences.

In keeping with the above, emerging neuroimaging research in individuals with psychotic experiences has shown reduced integrity of white matter tracts that are important in coordinating high-speed information processing between distributed neural networks (Amico et al., Reference Amico, O'Hanlon, Kraft, Oertel-Knochel, Clarke, Kelleher and Cannon2017; Drakesmith et al., Reference Drakesmith, Caeyenberghs, Dutt, Zammit, Evans, Reichenberg and Jones2015; Drakesmith et al., Reference Drakesmith, Dutt, Fonville, Zammit, Reichenberg, Evans and David2016; Jacobson et al., Reference Jacobson, Kelleher, Harley, Murtagh, Clarke, Blanchard and Cannon2010; Jacobson McEwen et al., Reference Jacobson McEwen, Connolly, Kelly, Kelleher, O'hanlon, Clarke and Sheehan2014; O'Hanlon et al., Reference O'Hanlon, Leemans, Kelleher, Clarke, Roddy, Coughlan and Cannon2015; Orr, Turner, & Mittal, Reference Orr, Turner and Mittal2014). Most recently, neuroimaging research has demonstrated reduced functional connectivity in young people with psychotic experiences, including relative hypoconnectivity affecting the default mode network, central executive network, salience network and motor network (O'Neill et al., Reference O'Neill, Carey, Dooley, Healy, Coughlan, Kelly and Cannon2020). Importantly, recent research suggests that hypoconnectivity associated with psychotic experiences may be dynamic and state-dependent (Mennigen et al., Reference Mennigen, Jolles, Hegarty, Gupta, Jalbrzikowski, Olde Loohuis and Bearden2020).

Once an individual passes their neural efficiency threshold, effects on a wide range of mental processes may begin to emerge: impaired communication between hubs involved in logical thought, for example, may give rise to delusional beliefs and thought disorder. Impaired communication between hubs involved in social cognition may give rise to social skills deficits. Impaired communication between hubs involved in language may give rise to disorganised speech. Impaired communication between hubs involved in complex motor skills may give rise to disorganised behaviour. Crucially, problems integrating neural activity from hubs involved in sensory perception may give rise to hallucinations.

Although early motor, language, social or thought process problems may be subtle enough to be overlooked, unusual perceptual experiences such as perceiving one's thoughts as though they are aloud represents a qualitative difference that is difficult to ignore. Therefore, hallucinations may emerge as one of the earliest signs of neural systems dysfunction and could represent a clinical bellwether of an individual passing their neural efficiency threshold. Although problems in neurocognition, social cognition, language and motor skills might emerge at the same time, these deficits may need to be significantly more advanced before they become readily apparent.

Psychotic experiences and trauma

Beyond neurocognitive and neuromotor effects, we suggest that dysfunction associated with passing a neural efficiency threshold is also a major factor contributing to the strong relationship between psychotic experiences and trauma (Arseneault et al., Reference Arseneault, Cannon, Fisher, Polanczyk, Moffitt and Caspi2011; Kelleher et al., Reference Kelleher, Keeley, Corcoran, Ramsay, Wasserman, Carli and Cannon2013c; McGrath et al., Reference McGrath, Saha, Lim, Aguilar-Gaxiola, Alonso, Andrade and Cardoso2017; Morrison, Frame, & Larkin, Reference Morrison, Frame and Larkin2003; Schäfer & Fisher, Reference Schäfer and Fisher2011). Psychological stress, arising from exposure to adverse life events, will drive increased interaction between relevant neural hubs, including those involved in threat and stress responses. This increased neural activity ('traffic’) in the connectome reduces capacity for interaction and high-speed coordination between other hubs. The greater the degree of distress, the greater the network capacity demand that is taken up responding to this stressor, which results in reduced network capacity for other interacting hubs. Therefore, coordination between neural hubs that are dependent on high-speed interaction may be delayed, including coordination between hubs involved in sensory perception, which may give rise to hallucinatory experiences.

Psychotic experiences and suicidal behaviour

We suggest that reduced efficiency of cooperation between hubs involved in complex mental processes may serve to increase the risk of suicidal behaviour at the same time as increasing risk of psychotic experiences (Bromet et al., Reference Bromet, Nock, Saha, Lim, Aguilar-Gaxiola, Al-Hamzawi and Degenhardt2017; DeVylder, Lukens, Link, & Lieberman, Reference DeVylder, Lukens, Link and Lieberman2015; Fisher et al., Reference Fisher, Caspi, Poulton, Meier, Houts, Harrington and Moffitt2013; Hielscher et al., Reference Hielscher, Connell, Lawrence, Zubrick, Hafekost and Scott2019; Kelleher et al., Reference Kelleher, Cederlof and Lichtenstein2014a, Reference Kelleher, Devlin, Wigman, Kehoe, Murtagh, Fitzpatrick and Cannonb; Yates et al., Reference Yates, Lang, Cederlof, Boland, Taylor, Cannon and Kelleher2019). This includes, for example, cooperating hubs involved in the execution of complex problem-solving skills (Barzilay et al., Reference Barzilay, Calkins, Moore, Wolf, Satterthwaite, Scott and Gur2019), hubs involved in language skills that allow one to verbalise one's challenges, and hubs involved in complex social skills that allow one to effectively elicit help from others. Taken together, these problems will result in difficulties generating sophisticated solutions in the face of complex problems and, as a result, may contribute to the risk of suicide.

Testable hypotheses

We propose that the neural efficiency-threshold model allows the generation and testing of many different hypotheses with regard to psychotic experiences. For example, from a cognitive perspective, the model suggests that neurocognitive deficits are a feature of the emergence of psychotic experiences rather than of the underlying mental disorder per se. From a developmental perspective, the model suggests that stressful situations are more likely to elicit psychotic experiences in pre-adolescent children than in adolescents and adults. From a functional perspective, the model suggests that, within individuals with mental disorders, the presence of psychotic experiences would be better predictors of functioning than information on the severity of symptoms of the underlying disorder itself.

One particular important testable hypothesis, which arises from this model, is that individuals exposed to adverse life events who develop psychotic experiences should have a higher risk of suicidal behaviour than individuals who experience the same adverse life events but who do not develop psychotic experiences. In that way, psychotic experiences might act as a measurable marker of the subjective impact of adverse life events, including in terms of suicidal behaviour risk.

We hope this neural efficiency threshold model will provide a useful framework for a broader bio-psycho-social understanding of psychotic experiences to guide and inspire future research.