Due to the COVID-19 crisis, the transition into subscription mode of the Oxford Research Encyclopedia of Neuroscience has been temporarily postponed. Please watch this space for updates as we work toward launching in the near future. Visit About to learn more, meet the editorial board, or learn how to subscribe.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, NEUROSCIENCE (oxfordre.com/neuroscience). (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 04 April 2020

Neurobiology of Auditory Hallucinations

Summary and Keywords

Auditory verbal hallucinations (AVH), also referred to as “hearing voices,” are vivid perceptions of speech that occur in the absence of any corresponding external stimulus but seem very real to the voice hearer. They are experienced by the majority of people with schizophrenia, less frequently in other psychiatric and neurological conditions, and are relatively rare in the general population. Because antipsychotic medications are not always successful in reducing the severity or frequency of AVH, a better understanding is needed of their neurobiological basis, which may ultimately lead to more precise treatment targets.

What voices say and how the voices sound, or their phenomenology, varies widely within and across groups of people who hear them. In help-seeking populations, such as people with schizophrenia, the voices tend to be threatening and menacing, typically spoken in a non-self-voice, often commenting and sometimes commanding the voice hearers to do things they would not otherwise do. In psychotic populations, voices differ from normal inner speech by being unbidden and unintended, co-opting the voice hearer’s attention. In healthy voice-hearing populations, voices are not typically distressing nor disabling, and are sometimes comforting and reassuring. Regardless of content and valence, voices tend to activate some speech and language areas of the brain. Efforts to silence these brain areas with neurostimulation have had mixed success in reducing the frequency and salience of voices. Progress with this treatment approach would likely benefit from more precise anatomical targets and more precisely dosed neurostimulation.

Neural mechanisms that may underpin the experience of voices are being actively investigated and include mechanisms enabling context-based predictions and distinctions between experiences coming from self and other. Both these mechanisms can be studied in non-human animal “models” and both can provide new anatomical targets for neurostimulation.

Keywords: auditory verbal hallucinations, voice hearing, psychosis, wellness spectrum, neurobiology

What Do Voice Hearers Hear?

It is important to define the phenomenon clinical neuroscientists are trying to understand. As Jones (2010) so aptly summarized it: “The term AVH encapsulates a diverse phenomenological experience, which may involve single and/or multiple voices, who may be known and/or unknown, speaking sequentially and/or simultaneously, in the first, second, and/or third person and which may give commands, comments, insults, or encouragement.”

It is also important to note that voice hearers themselves refer to these experiences as “voices” because the word “hallucination” implies they are not real, which to them, they are. “Hallucinations” needlessly pathologizes this experience, which many across the wellness spectrum experience. Both labels are used here.

Although they are referred to as “voices” or “auditory verbal hallucinations” (AVH), we should also consider if they are always an auditory experience. What starts out as a thought or a wisp of meaning may take on auditory qualities, based on whether a voice hearer is expecting to hear something (Wilkinson, 2014). Frith (1992) pointed out that voices can involve “an experience of receiving a communication without any sensory component.” A recent in-depth phenomenological analysis of predominant patterns of AVH (Jones & Luhrmann, 2015) found that fewer than 20% of voice hearers reported hallucinations that were experienced as literally auditory, while about 10% experienced them as only partially auditory, 28.8% experienced a combination of distinctly auditory and thought-like voices, and 15% reported AVH as unambiguously thought-like. This quote from that paper provides some understanding of how difficult it is for people with schizophrenia (and even the non-voice hearing population) to describe inner phenomena:

it becomes really, really difficult for me to distinguish (or explain the difference) between real voices, “hallucinated voices,” thoughts in my own head, other people’s thoughts, thoughts that might be circulating in the air, and then even the extent to which anything exists other than thoughts. Once everything starts to seem like a thought, then it’s really difficult because voices are thoughts, but so are visuals.

(Jones & Luhrmann, 2015)

This quote resonates with the inconsistent findings from the neurobiological literature about the involvement of primary auditory cortex in the experience of hearing voices (Allen, Laroi, McGuire, & Aleman, 2008), described in detail in “Where in the Brain Are Voices Instantiated?”.

Who Hears Voices?

The biological basis of “voices” has been traditionally studied in people with schizophrenia, a population characterized by hearing voices, although “voices” alone are not diagnostic of schizophrenia (Waters & Fernyhough, 2017). While expedient, studying voices in schizophrenia is fraught with confounds; people with schizophrenia are typically medicated, and experience social-economic and cognitive decline, making the assessment of this symptom in isolation difficult. Nevertheless, because of the morbidity and mortality associated with psychosis, it can be argued that people with schizophrenia should be the target population to study.

People with other psychiatric and neurological diagnoses hear voices, with approximately 50% of individuals with substance use disorders, 32% with medical/neurological conditions, and 46–57% with other psychiatric diagnoses specifically experiencing AVH (Waters & Fernyhough, 2017). As in the case of people with schizophrenia, AVH would not be an isolated symptom in these medical and psychiatric groups. These illnesses have distinct but significant clinical and functional sequelae like schizophrenia, and may have distinct phenomenologies that are themselves worthy of study.

In recent years, it has become increasingly apparent that a small but significant percentage of the general population hears voices. Non-clinical populations describe voices similarly to people with schizophrenia, in so far as they seem to originate in external space and speak in a running commentary, which are common features of AVH in schizophrenia. Unlike clinical voice hearers, healthy voice hearers tend to not experience commanding or negative and threatening voices, and the voices do not interfere with their lives. Moreover, they feel that they have control over the voices. The concept of “equi-finality” is relevant here (Cicchetti & Rogosch, 1996): if the symptom (e.g., a running commentary) is the same (“equi-final”) in two different groups, but the diagnosis is different (e.g., schizophrenia vs. healthy), can it be assumed the same neurobiological mechanism is involved (McCarthy-Jones et al., 2014)? As Waters and Fernyhough (2017) point out, “there are dangers in inferring continuity in mechanism from continuity in phenomenology, including a risk of incorrect treatment.”

Important in this decision is what particular aspect of the AVH experience clinical neuroscientists want to study. If it is important to understand the neurobiology of the negative content and command features of the voices, or the lack of control people have over the voices, then studying people with schizophrenia who hear threatening and hostile voices would be a good strategy. However, if the aim is to understand how perception is constructed from expectations, then a powerful strategy might be to study “conditioned hallucinations” in healthy people for whom hearing voices has been incorporated into their lives and livelihoods. An excellent example of the latter is healthy clairaudients who report receiving auditory messages from other realms, much like clairvoyants receive visions (Powers, Kelley, & Corlett, 2017). By studying this population, Powers and colleagues got traction on understanding how people can be conditioned to hear sounds that are not present, but it is unknown whether these voices are on a continuum with voices experienced by people with psychosis. Clearly, there is not one ideal population, one cluster of symptoms, or one mechanism to study in the effort to understand the neurobiology of AVH.

Where in the Brain Are Voices Instantiated?

All of the studies reviewed here involve human functional brain imaging, specifically functional magnetic resonance imaging (fMRI)- and electro-encephalography (EEG)-based methods. EEG is recorded from electrodes attached to the scalp and is a direct measure of underlying neural activity in the sulci and cortical gyri, which can result indirectly from activity in distant cortical and subcortical sources. Individual EEG trials are averaged to produce an event-related potential (ERP), a small voltage change in brain activity resulting from a specific event or stimulus, whose components develop and resolve within milliseconds. Time-frequency decomposition of the EEG signal is also possible, in which power and synchrony of neural oscillations in specific frequencies can be extracted. fMRI data are acquired in an MR scanner and reflect the blood oxygen level dependent (BOLD) response. Because it depends on the slow flow of blood in the brain, fMRI operates on a delayed timescale, taking about four to six seconds to develop and another eight to 10 seconds to resolve (Buckner, 1998). Compared with EEG, fMRI provides superior anatomical, but poor temporal, resolution. Compared with fMRI, EEG provides superior temporal, but poor anatomical, resolution.

Symptom Capture Paradigms

There are several paradigms that have been used to study the neurobiology of voices. The most intuitive and direct is referred to as “symptom capture,” which is a naturalist approach where neurobiological data are collected as study participants experience voices. With this method, investigators are hoping to find neurobiological evidence that the voices are perceived as auditory and activate primary auditory cortex, are perceived as language and involve Broca’s and Wernicke’s areas, and are associated with intrusive memories and involve the hippocampus.

While the symptom capture approach is conceptually simple, it is extremely difficult in practice because it relies not only on the timely occurrence of an elusive subjective experience, but also on the ability to reliably report its initiation and completion. Ideally, they would also be able to report periods when no voices are heard so that periods with and without “voices” can be contrasted. This approach requires patience from the research team and cooperation and insight from the participant. Furthermore, because of the heterogeneous phenomenology of voices, the results of these studies are inconsistent (Jardri, Pouchet, Pins, & Thomas, 2011).

Despite the difficulty inherent in this approach, there have been numerous successes. Among the first was a positron emission tomography (PET) study of five people with psychosis that compared activity during periods when voices are present to periods when they are absent (Silbersweig et al., 1995). The onset of AVHs was associated with activity in hippocampus and parahippocampus, thalamus, ventral striatum, and anterior cingulate, but not primary auditory cortex. This led others to start using less invasive and more accessible methods, such as MRI. Lennox, Park, Jones, and Morris (1999) studied one person whose hallucinations activated right middle temporal gyrus (MTG), but not primary auditory cortex. Dierks et al. (1999) studied three people whose hallucinations activated Broca’s area and primary auditory cortex. Moreover, they subsequently confirmed that the same area in primary auditory cortex activated during AVH was activated by external tones as well as verbal auditory stimuli, suggesting that voices in these people had a distinct auditory perceptual quality. In a larger study of 24 people with psychosis, Sommer et al. (2008) compared periods of hearing voices to periods of repeating single words silently. The principal difference between these two conditions was in the lateralization of the effects: during voice hearing, the right homologue of Broca’s area and the right superior temporal gyrus were active, while during inner speech, the left homologues of these areas were active. The authors suggested that the predominant engagement of the right inferior frontal area during voice hearing may be related to the low semantic complexity and negative emotional content typical of voices in people with psychosis.

In an extension of this work, Diederen et al. (2010) reported data from 24 psychotic people, many of whom participated in the study by Sommer et al. (2008), and from 26 healthy controls. The content of their AVHs did not parallel their own thoughts, such as commands or derogatory critique. During fMRI scanning, the participants were asked to squeeze a handheld balloon to signal the presence of voices, and their data were compared to data from the healthy controls who were instructed to squeeze the balloon about 10 times at random intervals. AVHs were associated with activation in bilateral language and motor regions. Deactivation preceding AVH was observed in the left parahippocampal gyrus, left superior temporal, right inferior frontal, and left middle frontal gyri as well as in the right insula and left cerebellum. Relevant to the point raised in the section “Who Hears Voices,” Diederen et al. (2012) next compared 21 non-psychotic voice hearers to 21 psychotic voice hearers and found no differences in brain areas activated during voice hearing, suggesting AHVs may result from comparable brain networks in the psychotic and non-clinical groups. Furthermore, these results suggest that the healthy voice hearing population may indeed offer an opportunity to study the neurobiology of AVH in the absence of other symptoms extant in people with psychosis, but with the caution that continuity in mechanism may not reflect continuity in phenomenology (Waters & Fernyhough, 2017).

These studies all relied on a motor response to indicate the onset of hallucinations. Importantly, pressing a button to signal a salient auditory event is well known to activate the same brain structures as “voices” (van Lutterveld, Diederen, Koops, Begemann, & Sommer, 2013), so care must be taken to avoid this interpretative confound. Shergill, Brammer, Williams, Murray, and McGuire (2000) acquired fMRI scans at unpredictable intervals while people with schizophrenia were intermittently hearing voices, and after each acquisition, the participants signaled whether they had been hallucinating during that scan. They found a distributed network of cortical and subcortical activity associated with hearing voices: inferior frontal/insular, anterior cingulate, temporal cortex bilaterally, right thalamus and inferior colliculus, left hippocampus, para-hippocampal cortex. Of note: no activation was observed in the primary auditory cortex, possibly because data were specifically acquired during a silent period in the MRI. Six to nine seconds before hearing a voice, left inferior frontal and right middle temporal gyri were active. This supports the hypothesis that during hallucinations, activation in cortical regions mediating the generation of inner speech may precede the engagement of areas implicated in the perception of auditory verbal material (Shergill, Samson, Bays, Frith, & Wolpert, 2005).

Finally, a recent quantitative meta-analysis of fMRI studies of AVH reported activation of Broca’s area in the frontal cortex, Wernicke’s area in the superior temporal gyrus, and hippocampus during hallucinations (Zmigrod, Garrison, Carr, & Simons, 2016). They additionally noted insula activation, which may relate to awareness and self-representation characteristic of voices, and the paracingulate region of medial prefrontal cortex and cerebellum, which may be involved in misattribution of self-generated percepts as coming from external sources.

In spite of the problems associated with symptom capture studies, the neuroanatomical data from these studies have been used to inform treatment. For example, clinicians have targeted areas of the brain known to be active during the experience of AVH with neurostimulation and have had some success reducing AVH frequency and salience (Hoffman et al., 2003). However, success with neurostimulation has been mixed (Slotema, Aleman, Daskalakis, & Sommer, 2012) and might be improved with individualized identification of areas of the brain that are active as an individual is experiencing voices.

Although lacking anatomical specificity, EEG has been used successfully in a symptom capture paradigm. Kindler and colleagues found that transiently stable “microstates” associated with error monitoring were shorter during hallucinatory episodes compared with non-hallucinatory episodes in people with schizophrenia (Kindler, Hubl, Strik, Dierks, & Koenig, 2010). The authors speculated that the early termination of this microstate facilitated the erroneous misattribution of self-generated inner speech to external sources during hallucinations.

Resting State Functional Connectivity

Resting state fMRI has revealed that activities in functionally related brain regions correlate with each other even in the absence of external stimuli (Beckmann, DeLuca, Devlin, & Smith, 2005; Biswal, Yetkin, Haughton, & Hyde, 1995; De Luca, Beckmann, De Stefano, Matthews, & Smith, 2006; Fox et al., 2005; Fransson, 2005; Greicius, Krasnow, Reiss, & Menon, 2003). These coherent, very low (< 1 Hz) frequency fluctuations during rest are thought to reflect the intrinsic functional and structural architecture of the brain (see Buckner, Andrews-Hanna, & Schacter, 2008; Fox & Raichle, 2007), but they may also reveal connectivity associated with spontaneous thoughts or “inner speech” in people who do and do not tend to hear voices. In fact, the application of functional connectivity analyses to the study of voices is motivated by a range of studies demonstrating that inter-region functional coordination plays an important role in determining whether neural activity is experienced consciously as percepts (Cosmelli et al., 2004; John, 2002; Melloni et al., 2007; Sergent & Dehaene, 2004), or perhaps voices.

Alderson-Day et al. (2016) reviewed the resting state connectivity literature and concluded that people who hear voices have abnormal resting connectivity in left hemisphere auditory and language regions, as well as atypical interactions between the default mode network (which is active when no task is assigned) and cognitive control and salience networks. This might suggest that voices are salient, co-opt conscious control, and distract the voice hearer from important activities. They did not find evidence for the most intuitive of them all: hyper-connectivity between the default mode network and auditory cortex, which would have given neurobiological support for the intuition that when the wandering mind interacts with auditory cortex, contents of the thoughts would become audible, as predicted by Northoff and Qin (2010).

In one of the first resting state connectivity studies of voice hearing, Hoffman and colleagues examined functional connectivity between components of the speech perception network during rest (Hoffman, Fernandez, Pittman, & Hampson, 2011). They predicted that hyper-connectivity within the speech perception network would facilitate the emergence of neural activity as conscious percepts. As a seed, they used the activity in Wernicke’s area, known to be involved in speech perception, and the right hemisphere homologue of Wernicke’s area, which they had found in a previous symptom capture study to activate during voice hearing (Hoffman, Anderson, Varanko, Gore, & Hampson, 2008). Compared with people with schizophrenia who did not hear voices, voice hearers showed greater connectivity between the bilateral temporal lobe seed and activity in language areas in the left inferior frontal gyrus, associated with language production and speech. In a follow-up analysis, they estimated connectivity between Wernicke’s and inferior frontal gyrus, between inferior frontal gyrus and putamen, and between Wernicke’s and putamen. This value was robustly greater for people with schizophrenia who heard voices compared with those who did not and compared with healthy control subjects.

In trying to understand this finding, it is important to note that the putamen appears to play a critical role in the generation of spontaneous language (Price, 2010) and in determining whether auditory stimuli are registered consciously as percepts (Mhuircheartaigh et al., 2010). Higher levels of functional coordination in this loop may be a causal factor leading to the experience of hearing voices. As a result, the threshold for consciousness of spontaneous activity arising in these cortical regions would be lowered to produce high rates of unbidden verbal representations. Moreover, a lowered threshold for consciousness may allow verbal content to emerge that is uncharacteristic of inner speech. These experiences may be cast in the speaking voice of other people since this network prominently incorporates right hemisphere temporal regions that play a critical role in the perception and speaker identification of non-self speech (Lattner, Meyer, & Friederici, 2005; Mathiak et al., 2007). Thus, the hyper-coupled network co-opts neural resources in an unbidden manner, and the percepts that result may not be suppressed due to failures of the efference copy in tagging the experiences as coming from “self” (see “Mechanistic Models of Voice Hearing”). In a further discussion of this finding, the authors noted that a weakening in any link in this hyper-connected chain is likely to be protective against hearing voices (Hoffman & Hampson, 2011).

Recently, a large number of resting state functional connectivity studies have reported thalamic hypo-connectivity with the cerebellum and prefrontal cortex as well as thalamic hyper-connectivity with sensory cortical regions in people with schizophrenia (Anticevic et al., 2014; Cheng et al., 2015; Damaraju et al., 2014; Klingner et al., 2014; Li et al., 2016; Woodward, Karbasforoushan, & Heckers, 2012). This pattern was recently replicated in 180 people with schizophrenia, which extended the literature by finding that hyper-connectivity between thalamus and middle temporal gyrus was related to AVH and delusions, while hypo-connectivity between thalamus to cerebellum was correlated with delusions and bizarre behavior (Ferri et al., 2018). This is illustrated in Figure 1. It is worth commenting on the overlap between delusions and hallucinations; while each can occur independently, they often co-occur, with delusions influencing a person’s understanding of why he or she is hearing voices (Maher, 1974). In fact, some of the features of AVH are themselves delusional; for example, believing that the voices can be heard by other people, that they are personifications of magical identities, and that they have special significance (Waters & Fernyhough, 2017).

Neurobiology of Auditory Hallucinations

Figure 1. Data from Ferri et al. (2018) are illustrated in this cartoon to highlight hyper-connectivity (wide arrow) between thalamus and right auditory cortex in people with schizophrenia (SZ) compared with healthy controls (HC), and hypo-connectivity (narrow arrow) between thalamus (gray oval) and cerebellum. Hyper-connectivity to right middle temporal gyrus (x = 63, y = -1, z = -23) is associated with Total Positive Symptoms, Delusions, and Hallucinations from the Scale for the Assessment of Positive Symptoms (SAPS) (Andreasen & Flaum, 1991). Hypo-connectivity to right posterior cerebellum (x = 39, y = -46, z = -47) is also associated with SAPS Total Positive Symptoms and Delusions, as well as Bizarre Behavior.

Probe Experiments—State and Trait

Due to the hypothesized involvement of auditory cortex in the experience of hearing voices, cortical responses to auditory stimuli have been used as a probe of auditory cortical activity during periods with and without hallucinations within a person (“state” studies) or between people who do and do not experience hallucinations (“trait” studies). The assumption in this approach is that the brain is tonically “tuned” to the internal channels in which hallucinatory stimuli are typically broadcast and is in a state of readiness to process internally generated auditory signals whenever they are spontaneously emitted (Kompus, Westerhausen, & Hugdahl, 2011). The auditory processing “bandwidth” devoted to internal verbal dialogues is posited to be particularly costly from the standpoint of auditory processing resources, diminishing the capacity of the auditory cortex to process external sounds. The N100 component of the EEG-based ERP, which is generated in the auditory cortex (Ford, Roach, Palzes, & Mathalon, 2016) and peaks within 100 milliseconds of the onset of a probe sound, as well as the BOLD response of the auditory cortex in response to a probe during fMRI, have been used to understand the effects of current AVH and the trait to experience them. In general, EEG and fMRI findings consistently report involvement of auditory areas of the brain in both AVH state and trait studies, but are less consistent in whether primary or secondary auditory cortical regions are involved, or left or right hemisphere.


In an early study of two people with psychosis, Tiihonen et al. (1992) reported abnormalities in the N100 to pure tones during hallucinations. These abnormalities included a latency delay in both and an amplitude reduction in one. Hubl, Koenig, Strik, Garcia, and Dierks (2007) used a hybrid symptom capture and probe approach in seven people with psychosis, who were asked to signal periods of hallucinations while simultaneously being presented with a series of tones. They found N100 responses to the tones were reduced, especially on the left, during periods of hallucinations compared with periods without hallucinations. David et al. (1996) used a similar hybrid approach with fMRI to study one person with psychosis, who signaled periods of hallucinations while simultaneously being probed with speech sounds and visual stimuli. Voices were associated with reduced responsiveness to sounds but not visual events. Another fMRI study by Woodruff et al. (1997) scanned seven people with psychosis during a period of severe AVHs and again about three months later when AVHs were diminished. They were not specifically asked to signal periods when they were hearing voices, but rather to rate the severity of any hallucinations that occurred before or during the scan. The authors reported less responsivity to external speech probes in right middle temporal gyrus during hallucinations. There is a consistent theme of involvement of auditory cortex, primary or secondary, left or right hemisphere, while people with psychosis are hearing voices.


In trait studies, neurobiological responses to probes are typically compared in people with psychosis who tend to hear voices to either those who do not tend to hear voices or to healthy controls. Alternatively, the neurobiological response can be regressed on the severity of the voices they experience. Using the latter approach, Plaze et al. (2006) assessed the state of auditory cortex in 16 people with schizophrenia while they listened to sentences spoken aloud. They reported a strong negative relationship between activation in left posterior superior temporal sulcus and clinician-rated severity of auditory hallucinations. A large multi-site study used a tone to probe the state of auditory cortex in people with psychosis and found that hallucinators (n = 66) had less activation to probe tones in left primary auditory cortex than non-hallucinators (n = 40) (Ford et al., 2009). These two studies converge on deficits in left primary auditory cortex in processing sounds in people who tend to hear voices, suggesting voices have a distinct auditory perceptual quality.

The mismatch negativity (MMN) and the positive P300 component of the auditory ERP might also be considered assays of the state of auditory cortex and higher cortical regions. Both are elicited by a deviant auditory event in a sequence of regular, repetitive auditory events; P300 is typically task relevant and the focus of attention, while MMN can be elicited even when attention is directed away from the sequence. The P300 reflects attention to the deviant sound, elicited effortfully or automatically, and MMN is considered pre-attentive. MMN occurs about 100–150 ms, and P300 about 300 ms, after the onset of a deviant event. Both MMN (see Erickson, Ruffle, & Gold, 2016; Umbricht & Krljes, 2005) and P300 (see Jeon & Polich, 2003; Ford, 1999; Bramon et al., 2004) are reduced in people with schizophrenia. It is conceivable that these reductions could be accounted for by the majority of people with schizophrenia who hear voices, reflecting insufficient cortical resources to switch attention, either automatically or effortfully, to an external stimulus. There are more than 100 reports of P300 amplitude reduction in schizophrenia; however, it is rarely reported to be correlated with hearing voices, although there are exceptions to this rule (Fisher, Labelle, & Knott, 2010; Havermans et al., 1999; Turetsky, Colbath, & Gur, 1998). Similarly, relationships between MMN and hearing voices are rarely reported, again with some exceptions (Fisher, Labelle, & Knott, 2008a, 2008b; Youn, Park, Kim, Kim, & Kwon, 2003). This could be due to a number of reasons. First, lacking a strong hypothesis, investigators may not have tried to find relationships between symptoms and MMN or P300. Second, investigators may have tested relationships but could not find them and did not report their negative results. A failure to find relationships could be due to antipsychotic medications reducing the range of symptom severity; symptom severity being rated over the last month but not at the time of ERP testing; and other reasons discussed in “Barriers.”

Mechanistic Models of Voice Hearing

In this section, two of the prominent models of voices are described. Jones (2010) notes that although “models are created to fit the phenomenology of AVHs, due to the diverse nature of the experience they inevitably end up only fitting specific types of AVHs. Whilst it is possible a grand unified theory of AVHs and AHs may be developed, possibly involving a neurological mechanism cutting across many cognitive, metacognitive, linguistic and/or perceptual capacities, at present no such model exists (p. 573).” Nevertheless, both the predictive coding and the self-monitoring theories have provided the field with some traction on understanding the mechanistic underpinnings of psychosis and psychotic symptoms.

Faulty Predictions Affecting Perception

Powers, Mathys, and Corlett, (2017) have had success testing the predictive coding and active inference model of AVH. As described in “Who Hears Voices?,” they recruited self-proclaimed clairaudient psychics through discussions held at meetings of spiritualists and referrals from other psychics. The voice-hearing experiences of clairaudient psychics were similar in many ways to help-seeking voice hearers: they shared low-level acoustic characteristics, content, and frequency. However, the clairaudients had a more positive affective response to the voices and felt they could control them. In a brain imaging study of this sample, Powers, Mathys, and Corlett (2017) paired near-threshold-level sounds to a light to condition participants to hear sounds even when none were presented. They found that people with schizophrenia who experience AVH were more susceptible to conditioning than people with schizophrenia who did not experience AVH. Importantly, clairaudients (by definition, with AVH) were also more susceptible than healthy people without AVH. This view of how AVH may result from strong, but erroneous, expectations is supported by this quote from Wilkinson (2014):

conscious percept is not determined by incoming stimuli, but by the brain’s best hypothesis as selected on the basis of how well it minimises prediction error. The question is not: Where does this erroneous stimulus come from? Rather the question is: Why does the brain adopt such a strange hypothesis? The answer to this question is (at least partially): erroneous precision weighting on prediction error. (p. 153)

In both voice hearing groups, areas previously shown to be active in symptom capture studies, such as superior temporal sulcus, were active when voice hearers mistakenly said they heard a tone when none was present. In addition, people who were more sensitive to their errors in perception activated the vermis of the cerebellum and were less likely to erroneously endorse hearing a sound when none was presented (Powers, Mathys, & Corlett, 2017). This is consistent with the role of the cerebellum in the production and updating of predictive models (Blakemore, Wolpert, & Frith, 1999) and the Powers et al. suggestion that “interventions to mollify psychosis more broadly, such as cerebellar transcranial magnetic stimulation (p. 600)” should be tried. Although it is difficult to argue from null findings, the relative lack of cerebellum involvement reported in many symptom capture studies suggests cerebellum activity contributes to normal predictive coding and the lack of AVH. It is worth noting that cerebellum was no longer active when controlling for motor activity associated with signaling the occurrence of AVH (Shergill et al., 2000).

Self-Monitoring Gone Awry

Many have argued that “voices” are comprised of inner speech that is misperceived or misattributed to external sources. As intuitive as this may seem, it has its detractors who point out that psychotic people who hear voices can also have normal inner speech (Langdon, Jones, Connaughton, Fernyhough, 2009) and are able to distinguish between their inner speech and their “voices” (Hoffman, Varanko, Gilmore, & Mishara, 2008). Furthermore, these experiences generally—although not universally—seem to be cast in one or more specific non-self-speaking voices (Hoffman, Varanko, et al., 2008; Nayani & David, 1996) with characteristic timbre, accent, and gender—features that are not associated with inner speech. Raij and Riekki (2012) directly compared imagined voices to AVH in people with schizophrenia. They found that the language areas of the brain were active in both conditions, but only supplementary motor area (SMA) was active during the imagery condition. Because SMA is involved in intentional motor acts (Engbert, Wohlschlager, & Haggard, 2008), they suggested that voices do not have intentionality.

Expanding “inner speech” to include any inner experience bubbling up from the subconscious and co-opting consciousness, with or without auditory or linguistic qualities, is closer to the real phenomenology. These experiences arise spontaneously, are unbidden, and are difficult to control. This is the normal experience of everyone; as minds wander, thoughts are encountered, as are recent and planned conversations. The tonal quality of the other person’s voice or prosody can often be “heard.” What converts this normal experience into a hallucination may be the sense that the experience is not self-generated and cannot be controlled. That is, regardless of content (commands vs. comments), tone (hostile vs. supportive), linguistic quality (words vs. sentences), or familiarity (plans vs. memories), these experiences are typically unbidden, yet they seem to be coming from ourselves, rather than from an external source. As said so eloquently by Gallagher (2000) about his own wandering thoughts, “Not only do they appear to be part of my stream of consciousness, but, despite the fact that I am not willing them, and may even be resisting them, they still seem to be generated within my own cognitive experience.”

Given their amorphous nature, these spontaneous experiences probably do not violate the assumptions of AVH being “inner speech.” How these inner musings/thoughts/ruminations become audible and take on different personalities may be due to strong prior (and erroneous) expectations of the voice hearers (Corlett et al., 2019; Wilkinson, 2014), discussed in “Faulty Predictions Affecting Perception.” Alternatively, as discussed in “Agency,” these inner experiences may become audible because they are missing the “agency tag” that identifies these experiences as coming from self.


Regardless of whether AVHs are pulled from old memories, have a linguistic quality, or are distinctly audible, they are typically experienced as separate from one’s own mental processes and lack “self” attributes. This suggests there may be deficits in recognizing “self” as the agent of the voices, regardless of their content. Deficits in agency might, thus, be considered a “super-construct.” To wit, failure to monitor inner (self) speech is a leading explanatory construct of AVH (Allen et al., 2008). Accordingly, assays of agency may provide insight into “how” inner experiences—thoughts, inner speech, ruminations—are experienced as external.

A dominant view about agency is that it “involves a generalisable relation between actions and their consequences, and is triggered by efferent motor commands” (Engbert et al., 2008). It is posited that every move is accompanied by transmission of an efference copy of the motor plan from motor cortex to cerebellum via pons, generating a corollary discharge of the expected sensory consequences of the motor act. The expected sensations resulting from the movement are posited to be compared with, and effectively subtracted from, the real sensations, resulting in net suppression or cancellation. This is illustrated in Figure 2. A poor match would send an error signal to cortex via thalamus. It is through this comparison process that errors are detected and corrected, before the proprioceptive system has a chance to respond and implement corrective action. This is likely carried out by automatic side-loops in the cerebellum, outside of conscious awareness. This system allows all species to fluidly and flexibly navigate the environment and discount sensations resulting from the movement. An excellent review describes this system in more detail (Ramnani, 2006).

Neurobiology of Auditory Hallucinations

Figure 2. In solid arrows, this cartoon shows a sequence of events launched by an intention to say “ah” and the consequent motor command to speak. The spoken sound is picked up by the microphone, routed through the audio system to the ear inserts. From the ear, the sensation is processed by the primary auditory system to the thalamus (gray oval), as the re-afference. Simultaneously, in dashed arrows, an efference copy of the motor plan is sent from frontal lobe areas to the cerebellum (gray striated structure) via the pons, where a corollary discharge of the expected sensation is registered. A recursive process between the thalamus and cerebellum rapidly compares the re-afference and the corollary discharge. When they match, the re-afference is relatively suppressed, as shown in a dotted arrow from thalamus to auditory cortex. When they do not match, the sound is more thoroughly processed in auditory cortex, as shown in the solid arrow from thalamus to auditory cortex. This illustrates the rapid, relatively inflexible thalamo-cerebellar loops that allow for parallel processing of motor actions and consequent sensations needed for executing the over-learned skills of everyday life (Ramnani, 2006).

Agency and the Cortical–Cerebellar–Thalamic Path

Fifty years ago, Marr (1969) suggested that high-level commands from the cerebral cortex could access low-level cerebellar representations, so that after motor skills are acquired, “a simple or incomplete message from the cerebrum will suffice to provoke their execution.” That is, cerebellum could be trained to run routine operations that result in skillfully executed movements, with little effort expended from cortex. Koziol et al. (2014) suggested that the internal modeling abilities that evolved for flexible motor control “were exapted during evolution, and that the development of increasingly richer and far-reaching anticipatory control loops bootstrapped cognition and thought . . . ” (p. 165).

More specifically, information from motor cortex reaches the cerebellar cortex via monosynaptic pathways in the pontine nuclei that form part of a closed loop system, in which the cerebellum returns projections to the cerebral cortex via the thalamus (Schmahmann, 1996). Given functional connectivity between somatosensory cortex and thalamus, cerebellum may play a substantial role in mediating this communication, and together with primary motor cortex, it may alter the perception of sensations coming from one’s actions (Blakemore, Wolpert, & Frith, 1999). It may be that a corollary discharge of the expected sounds during vocalization is deposited in cerebellum for immediate comparison against actual sounds, with the outcome of this comparison being registered in auditory cortex and language learning areas, enabling fine-tuning of local accents of conspecifics (Sitek et al., 2013).

In a consensus paper on the cerebellum by Koziol et al. (2014), Parker and Andreasen further suggested that fluidly coordinating sequences of thought and action depends on very rapid on-line processing and feedback between the cerebral cortex and the cerebellum, again, mediated through the thalamus. Deficits in this circuit may cause misconnections between percepts and their meanings, and in turn, cause errors in perceptual binding and misinterpretations of many kinds (e.g., delusions and hallucinations). Indeed, cerebellar activation has been noted during symptom capture studies of hallucinations (Shergill et al., 2000; Silbersweig et al., 1995). As shown in Figure 1, hypo-connectivity between thalamus and cerebellum is negatively correlated with delusions and bizarre behavior, adding further converging evidence of the role of cerebellum in psychotic symptoms (Ferri et al., 2018).

Invasive techniques have been used to study the neural basis of agency during vocalization in a variety of species, from crickets (Poulet & Hedwig, 2006) and bats (Suga & Schlegel, 1972), to human (Chen et al., 2011; Creutzfeldt, Ojeman, & Lettich, 1989) and non-human primates (Eliades & Wang, 2003, 2005, 2008). In all cases, auditory responsiveness is relatively suppressed when the animal is vocalizing compared with when the sound is coming from external sources. This suppression suggests the animal “knows” it is the agent of the sound and further processing is unnecessary. Accordingly, processing is minimized, resources are conserved, and the sound is tagged as coming from self.

These data are consistent with human neurophysiological studies using EEG or magneto-encephalographic (MEG) potentials synchronized to the onset of vocalization (Chen et al., 2011; Curio, Neuloh, Numminen, Jousmaki, & Hari, 2000; Ford, Mathalon, Heinks, et al., 2001; Heinks-Maldonado, Mathalon, Gray, & Ford, 2005; Houde, Nagarajan, Sekihara, & Merzenich, 2002; Wang et al., 2014). The N100 ERP component, or the corresponding M100 of the MEG-based response, shows that activity in auditory areas is suppressed during vocalization, compared with when the same speech sound is played back. Importantly, similar suppression is seen with carefully timed imagined or inner speech (Whitford et al., 2017). Suppression of auditory cortical responses may reflect the result of the comparison process between the actual or even imagined sound and the corollary discharge of the expected sound. Suggesting sensitivity to psychosis and psychosis vulnerability, N100 suppression is reduced in chronic (Ford, Gray, Faustman, Roach, & Mathalon, 2007; Ford, Mathalon, Kalba, et al., 2001) and early illness schizophrenia (Perez et al., 2012), psychotic bipolar and schizoaffective disorders (Ford et al., 2013), schizotypy (Oestreich et al., 2015), and youth at clinical high risk for psychosis (Mathalon et al., 2018).

Agency and Hallucinations

Feinberg (1978) suggested that dysfunction of the efference copy/corollary discharge mechanisms may contribute to the positive symptoms of schizophrenia, such as hallucinations and delusions. These mechanisms evolved for motor control, and the link between thinking (the presumed raw material of AVH) and motor mechanisms has been described by psychologists for a century or more (Jackson, 1958; Washburn, 1916). Feinberg linked thinking and the corollary discharge mechanism, when he said:

These discharges may assist in the distinction between self-generated and externally produced movements; they also allow (or represent) monitoring of the motor commands before the effector response has occurred. Here, it is hypothesized that this mechanism of control and integration is also present in thinking, which as Hughlings Jackson pointed out, may be considered the highest and most complex form of motor activity. I speculate that if corollary discharges are normally part of the motor mechanisms of thought, their derangement could produce many of the symptoms of schizophrenia. (1978, p. 639)

Thus, thinking and overt speech may utilize the same action-based motor system mechanisms to tag self-generated sensations and distinguish them from sensations with an external source. If this mechanism is dysfunctional, auditory sensations resulting from these inner experiences may not be tagged as “self” and may be experienced as AVH. Frith (1987) expanded this concept and prompted a series of experiments confirming failures of agency or self-monitoring in people with schizophrenia who tend to experience AVH (e.g., Brebion et al., 2000; Shergill et al., 2005; Stirling, Hellewell, & Quraishi, 1998) and delusions (e.g. Lindner, Thier, Kircher, Haarmeier, & Leube, 2005; Stirling et al., 1998).

While appealing and intuitive, the reality of finding strong relationships between AVH and the assay of corollary discharge has been disappointing. While a deficit in N100 suppression during talking appears to be a vulnerability marker of psychosis, it is not related to the severity of psychotic symptoms in any of these populations (but see Heinks-Maldonado et al. [2007]). However, by looking at the data in the time-frequency domain, relationships emerge with AVH and delusions. The dominant frequency of N100 is in the theta (4–7 Hz) range, and its amplitude is an admixture of theta power and trial-to-trial consistency of the phase of theta oscillations elicited by an auditory stimulus. Trial-to-trial phase consistency is sometimes called “phase resetting.” Using time-frequency analytic tools, N100 can be decomposed into theta power and phase consistency. Suppression of trial-to-trial theta phase consistency is more sensitive to psychotic symptoms, specifically delusions, and more sensitive to the diagnosis of schizophrenia than is N100 amplitude suppression during talking (Ford, Roach, Loewy, Stuart, & Mathalon, 2019). Using similar analytic tools, neural activity preceding vocalization has also been decomposed into beta band (13–20 Hz) phase consistency, the exact frequency band being limited by the rapid speed of vocalizations in this case (Ford et al., 2007). Ford et al. (2007) reported a relationship between AVH severity and beta band (13–20 Hz) phase consistency as people were preparing to speak. Since it was highly correlated with subsequent suppression of N100 amplitude, this precisely timed signal was posited to reflect the efference copy being sent from speech production areas, heralding the arrival of the imminent sound. Furthermore, coherence between frontal and temporal lobes in the theta band, which was greater during speaking than listening, was reduced in people with schizophrenia, especially in those who heard voices (Ford, Mathalon, Whitfield, Faustman, & Roth, 2002). These latter two studies suggest that the additional precision afforded by time-frequency analysis of EEG signals during the motor act of speaking is sensitive to both the diagnosis of schizophrenia and the symptoms experienced.

In light of these data, and with reference to Figure 2, the agency deficit related to psychotic symptoms may be in the efference copy being sent from cortex to cerebellum. N100 suppression may reveal the comparison between the expected and observed sensations executed in cerebellum, with the outcome being relayed through thalamus to auditory cortex. Important details about where in the cerebellum the efference copy might land, where the calculation might be made, and from where in the cerebellum the input to thalamus might be launched remains to be determined. As more is learned about the precise area of the cerebellum involved in the rapid processing of expected outcomes of actions, these regions might then be the target of focused neurostimulation to boost their function.


In the last half-century, human neuroscience methods have been refined to provide a way to study the brain non-invasively, and neuroscientists have established that hearing voices is associated with subtle abnormalities in brain structure and function. However, efforts to understand the neurobiological mechanism of voices are still under way. Obstacles are many and include medication confounds, failure to distinguish state and trait effects, correlation–causation ambiguity, and the absence of compelling animal models of specific symptoms to test mechanistic hypotheses (see Mathalon & Ford [2012] for a full discussion).

At the core is the questionable validity of our assessment of the voices people are experiencing. It is dependent on their ability to describe the experiences and the clinicians’ ability to understand and interpret them. For the voice hearer, describing the voices is akin to a non-voice hearer attempting to describe the content of a wandering mind, but with the possible overlay of hostile, threatening content. Moreover, voice hearers have the additional burden of needing to provide accurate information to the interviewer about the voices, including their frequency, duration, severity, and impact on their lives. Further difficulties may arise from a lack of insight, deficits in self-reflection and self-monitoring, poor memory when trying to recount the frequency and severity of voices over prior weeks or months, with guardedness and paranoia leading to under-reporting of voices, and with desire to please the interviewer leading to over- or under-reporting of symptoms.

Can the findings described here be generalized to all psychotic people who hear voices? Can it be assumed that the voices experienced by more severely ill psychotic people, who do not volunteer for brain imaging studies, will be underpinned by the same neurobiology? Similarly, can it be assumed that psychotic people who tolerate hours in a noisy tube have the same neurobiological underpinnings as those who cannot tolerate the MR scanner? EEG-based studies at least circumvent claustrophobia and concerns about metal in the body.

A significant limitation of clinical research in general, and schizophrenia research in particular, is that preexisting symptoms and brain abnormalities limit what can be studied. The inability to experimentally manipulate brain mechanisms to modulate specific symptoms leads to the exclusive reliance on correlational data to evaluate mechanistic models of specific symptoms. Since correlations cannot prove causation, even the demonstration of robust and replicable correlations between a neurobiological measure and symptom severity cannot definitively establish that there is a causal relationship. Animal models of the constructs that may underlie AVH, such as agency, are being studied by basic neuroscientists (Rummell, Klee, & Sigurdsson, 2016). If perturbations of these mechanisms can duplicate the abnormalities seen in people with AVH, the field can move forward with testable animal models focused on treatments. Clinical neuroscientists are encouraged to include the cerebellum in their fMRI studies of action-based phenomena, and basic physiologists are encouraged to record from cerebellum as others are doing (see Schneider & Mooney [2018]). The cerebellum, after all, contains more than 80% of all the neurons in the brain.


The neurobiology of auditory verbal hallucinations or “voices” is generally consistent with the phenomenology. But, because the phenomenology is inconsistent, so is the underlying neurobiology! Nevertheless, the existing evidence from symptom capture studies converges on the involvement of temporal lobe structures, cortical and subcortical, and perhaps cerebellum. Importantly, data do not necessarily point to the involvement of primary auditory cortex. Instead, they point to the involvement of middle temporal gyri and Wernicke’s area, involved in speech comprehension. Consistent with the involvement of memories in the content of the voices, hippocampus in the medial temporal lobe is sometimes active during the experience of voice hearing. Resting state connectivity studies, which look for differences between people who do and do not hear voices, find atypical interactions between auditory/language areas and the salience network, suggesting that pathways in the brains of voice hearers promote auditory-like experiences coupled with distraction from important activities. One of the dominant mechanistic models of AVH is that people who hear voices fail to tag the internal experience as being truly external, and do not have the sense that they, themselves, are the agent of the experience. Another dominant model of AVH is that perceptions of voices are constructed from strong prior beliefs that voices will be heard. Both models are being actively investigated and provide a hopeful path forward in trying to understand how voices can be heard when no one is talking.

Given the inconsistencies in the brain imaging literature, a clear conclusion is difficult to formulate. The inconsistency in brain imaging data might be underpinned by the heterogeneous phenomenology of AVHs, ranging from clear speech coming from outside a person’s head to wisps of meaning an entity is attempting to communicate. Added to this is the limited ability to capture an individual’s phenomenology in clinical interviews. Matching phenomenology to neurobiology, and neurobiology to targeted treatment, is one path forward.


Alderson-Day, B., Diederen, K., Fernyhough, C., Ford, J. M., Horga, G., Margulies, D. S., . . . & Jardri, R. (2016). Auditory hallucinations and the brain’s resting-state networks: Findings and methodological observations. Schizophrenia Bulletin, 42(5), 1110–1123.Find this resource:

Allen, P., Laroi, F., McGuire, P. K., & Aleman, A. (2008). The hallucinating brain: A review of structural and functional neuroimaging studies of hallucinations. Neuroscience & Biobehavioral Reviews, 32(1), 175–191.Find this resource:

Andreasen, N. C., & Flaum, M. (1991). Schizophrenia: The characteristic symptoms. Schizophrenia Bulletin, 17(1), 27–49.Find this resource:

Anticevic, A., Cole, M. W., Repovs, G., Murray, J. D., Brumbaugh, M. S., Winkler, A. M., . . . & Glahn, D. C. (2014). Characterizing thalamo-cortical disturbances in schizophrenia and bipolar illness. Cerebral Cortex, 2, 3116–3130.Find this resource:

Beckmann, C. F., DeLuca, M., Devlin, J. T., & Smith, S. M. (2005). Investigations into resting-state connectivity using independent component analysis. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 360, 1001–1013.Find this resource:

Biswal, B., Yetkin, F. Z., Haughton, V. M., & Hyde, J. S. (1995). Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magnetic Resonance in Medicine, 3, 537–541.Find this resource:

Blakemore, S. J., Wolpert, D. M., & Frith, C. D. (1999).The cerebellum contributes to somatosensory cortical activity during self-produced tactile stimulation. Neuroimage, 10, 448–459.Find this resource:

Bramon, E., Rabe-Hesketh, S., Sham, P., Murray, R. M., & Frangou, S. (2004). “Meta-analysis of the P300 and P50 waveforms in schizophrenia,” Schizophrenia Research, 70, 315–329.Find this resource:

Brebion, G., Amador, X., David, A., Malaspina, D., Sharif, Z., & and Gorman, J. M. (2000). Positive symptomatology and source-monitoring failure in schizophreni: An analysis of symptom-specific effects. Psychiatry Research, 95, 119–131.Find this resource:

Buckner, R. L. (1998). Event-related fMRI and the hemodynamic response. Human Brain Mapping, 6, 373–377.Find this resource:

Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain’s default network: Anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences, 1124, 1–38.Find this resource:

Chen, C. M., Mathalon, D. H., Roach, B. J., Cavus, I., Spencer, D. D., & Ford, J. M. (2011). The corollary discharge in humans is related to synchronous neural oscillations. Journal of Cognitive Neuroscience, 23, 2892–2904.Find this resource:

Cheng, W., Palaniyappazn, L., Li, M., Kendrick, K. M., Zhang, J., Luo, Q., . . . & Wang, Q. (2015). Voxel-based, brain-wide association study of aberrant functional connectivity in schizophrenia implicates thalamocortical circuitry. npj Schizophrenia, 1.Find this resource:

Cicchetti, D., & Rogosch, F. A. (1996). Equifinality and multifinality in developmental psychopathology. Development and Psychopathology, 8, 597–600.Find this resource:

Corlett, P. R., Horga, G., Fletcher, P. C., Alderson-Day, B., Schmack, K., & Powers, A. R. 3rd. (2019). Hallucinations and strong priors. Trends in Cognitive Science, 23, 114–127.Find this resource:

Cosmelli, D., David, O., Lachaux, J. P., Martinerie, J., Garnero, L., Renault, B., & Varela, F. (2004). Waves of consciousness: Ongoing cortical patterns during binocular rivalry. Neuroimage, 23, 128–140.Find this resource:

Creutzfeldt, O., Ojeman, G., & Lettich, E. (1989). Neuronal activity in the human lateral temporal lobe. II Responses to the subject’s own voice. Experimental Brain Research, 77, 476–489.Find this resource:

Curio, G., Neuloh, G., Numminen, J., Jousmaki, V., & Hari, R. (2000). Speaking modifies voice-evoked activity in the human auditory cortex. Human Brain Mapping, 9, 183–191.Find this resource:

Daalman, K., Boks, M. P., Diederen, K. M., de Weijer, A. D., Blom, J. D., Kahn, R. S., & Sommer, I. E. (2011). The same or different? A phenomenological comparison of auditory verbal hallucinations in healthy and psychotic individuals. Journal of Clinical Psychiatry, 72, 320–325.Find this resource:

Damaraju, E., Allen, E. A., Belger, A., Ford, J. M., McEwen, S., Mathalon, D. H., . . . & Preda, A. (2014). Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia. NeuroImage: Clinical, 5, 298–308.Find this resource:

David, A. S., Woodruff, P. W. R., Howard, R., Mellers, J. D. C., Brammer, M., Bullmore, E., . . . & Williams, S. C. R. (1996). Auditory hallucinations inhibit exogenous activation of auditory association cortex. Neuroreport, 7, 932–936.Find this resource:

De Luca, M., Beckmann, C. F., De Stefano, N., Matthews, P. M., & Smith, S. M. (2006). fMRI resting state networks define distinct modes of long-distance interactions in the human brain. Neuroimage, 29, 1359–1367.Find this resource:

Diederen, K. M., Daalman, K., de Weijer, A. D., Neggers, S. F., van Gastel, W., Blom, J. D., . . . & Sommer, I. E. (2012). Auditory hallucinations elicit similar brain activation in psychotic and nonpsychotic individuals. Schizophrenia Bulletin, 38, 1074–1082.Find this resource:

Diederen, K. M., Neggers, S. F., Daalman, K., Blom, J. D., Goekoop, R., Kahn, R. S., & Sommer, I. E. (2010). Deactivation of the parahippocampal gyrus preceding auditory hallucinations in schizophrenia. American Journal of Psychiatry, 167, 427–435.Find this resource:

Dierks, T., Linden, D. E. J., Jandl, M., Formisano, E., Goebel, R., Lanfermann, H., & Singer, W. (1999). Activation of Heschl’s Gyrus during auditory hallucinations. Neuron, 22, 615–621.Find this resource:

Eliades, S. J., & Wang, X. (2003). Sensory-motor interaction in the primate auditory cortex during self-initiated vocalizations. Journal of Neurophysiology, 89, 2194–2207.Find this resource:

Eliades, S. J., & Wang, X. (2005). Dynamics of auditory-vocal interaction in monkey auditory cortex. Cerebral Cortex, 15, 1510–1523.Find this resource:

Eliades, S. J., & Wang, X. (2008). Neural substrates of vocalization feedback monitoring in primate auditory cortex. Nature, 453, 1102–1106.Find this resource:

Engbert, K., Wohlschlager, A., & Haggard, P. (2008). Who is causing what? The sense of agency is relational and efferent-triggered. Cognition, 107, 693–704.Find this resource:

Erickson, M. A., Ruffle, A., & Gold, J. M. (2016). A meta-analysis of mismatch negativity in schizophrenia: From clinical risk to disease specificity and progression. Biological Psychiatry, 79, 980–987.Find this resource:

Feinberg, I. (1978). Efference copy and corollary discharge: Implications for thinking and its disorders. Schizophrenia Bulletin, 4, 636–640.Find this resource:

Ferri, J., Ford, J. M., Roach, B. J., Turner, J. A., van Erp, T. G., Voyvodic, J., . . . & Mathalon, D. H. (2018). Resting-state thalamic dysconnectivity in schizophrenia and relationships with symptoms. Psychological Medicine, 1–8.Find this resource:

Fisher, D. J., Labelle, A., & Knott, V. J. (2008a). Auditory hallucinations and the mismatch negativity: Processing speech and non-speech sounds in schizophrenia. International Journal of Psychophysiology, 70, 3–15.Find this resource:

Fisher, D. J., Labelle, A., & Knott, V. J. (2008b). The right profile: Mismatch negativity in schizophrenia with and without auditory hallucinations as measured by a multi-feature paradigm. Clinical Neurophysiology, 119, 909–921.Find this resource:

Fisher, D. J., Labelle, A., & Knott, V. J. (2010). Auditory hallucinations and the P3a: Attention-switching to speech in schizophrenia. Biological Psychology, 85, 417–423.Find this resource:

Ford, J. M., Gray, M., Faustman, W. O., Roach, B. J., & Mathalon, D. H. (2007). Dissecting corollary discharge dysfunction in schizophrenia. Psychophysiology, 44, 522–529.Find this resource:

Ford, J. M., Mathalon, D. H., Heinks, T., Kalba, S., Faustman, W. O., & Roth, W. T. (2001). Neurophysiological evidence of corollary discharge dysfunction in schizophrenia. American Journal of Psychiatry, 158, 2069–2071.Find this resource:

Ford, J. M., Mathalon, D. H., Kalba, S., Whitfield, S., Faustman, W. O., & Roth, W. T. (2001). Cortical responsiveness during talking and listening in schizophrenia: An event-related brain potential study. Biological Psychology, 50, 540–549.Find this resource:

Ford, J. M., Mathalon, D. H., Roach, B. J., Keedy, S. K., Reilly, J. L., Gershon, E. S., & Sweeney, J. A. (2013). Neurophysiological evidence of corollary discharge function during vocalization in psychotic patients and their nonpsychotic first-degree relatives. Schizophrenia Bulletin, 39, 1272–1280.Find this resource:

Ford, J. M., Mathalon, D. H., Whitfield, S., Faustman, W. O., & Roth, W. T. (2002). Reduced communication between frontal and temporal lobes during talking in schizophrenia. Biological Psychology, 51, 485–492.Find this resource:

Ford, J. M. (1999). Schizophrenia: The broken P300 and beyond. Psychophysiology, 36, 667–682.Find this resource:

Ford, J. M., Roach, B. J., Jorgensen, K. W., Turner, J. A., Brown, G. G., Notestine, R., . . . & Mathalon, D. H. (2009). Tuning in to the voices: A multisite FMRI study of auditory hallucinations. Schizophrenia Bulletin, 35, 58–66.Find this resource:

Ford, J. M., Roach, B. J., Loewy, R. L., Stuart, B. K., & Mathalon, D. H. (2019). Theta phase synchrony is sensitive to corollary discharge abnormalities in early illness schizophrenia but not in the clinical high-risk syndrome. BioRxiv.Find this resource:

Ford, J. M., Roach, B. J., Palzes, V. A., & Mathalon, D. H. (2016). Using concurrent EEG and fMRI to probe the state of the brain in schizophrenia. NeuroImage: Clinical, 12, 429–441.Find this resource:

Fox, M. D., & Raichle, M. E. (2007). Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nature Reviews Neuroscience, 8, 700–711.Find this resource:

Fox, M. D., Snyder, A. Z., Vincent, J. L., Corbetta, M., Van Essen, D. C., & Raichle, M. E. (2005). The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proceedings of the National Academy of Sciences of the USA, 102, 9673–9678.Find this resource:

Fransson, P. (2005). Spontaneous low-frequency BOLD signal fluctuations: An fMRI investigation of the resting-state default mode of brain function hypothesis. Human Brain Mapping, 26, 15–29.Find this resource:

Frith, C. D. (1987). The positive and negative symptoms of schizophrenia reflect impairments in the perception and initiation of action. Psychological Medicine, 17, 631–648.Find this resource:

Frith, C. D. (1992). The Cognitive Neuropsychology of Schizophrenia. Hillsdale, NJ: Lawrence Erlbaum Associates.Find this resource:

Gallagher, S. (2000). Self-Reference and schizophrenia: A cognitive model of immunity to error through misidentification. In D. Zahavi (Ed.), In exploring the self: Philosophical and psychopathological perspectives on self-experience. Amsterdam: John Benjamins.Find this resource:

Greicius, M. D., Krasnow, B., Reiss, A. L., & Menon, V. (2003). Functional connectivity in the resting brain: A network analysis of the default mode hypothesis. Proceedings of the National Academy of Sciences of the USA, 100, 253–258.Find this resource:

Havermans, R., Honig, A., Vuurman, E. F. P. M., Krabbendam, L., Wilmink, J., Lamers, T., . . . & vanPraag, H. M. (1999). A controlled study of temporal lobe structure volumes and P300 responses in schizophrenic patients with persistent auditory hallucinations. Schizophrenia Research, 38, 151–158.Find this resource:

Heinks-Maldonado, T. H., Mathalon, D. H., Gray, M., & Ford, J. M. (2005). Fine-tuning of auditory cortex during speech production. Psychophysiology, 42, 180–190.Find this resource:

Heinks-Maldonado, T. H., Mathalon, D. H., Houde, J. F., Gray, M., Faustman, W. O., & Ford, J. M. (2007). Relationship of imprecise corollary discharge in schizophrenia to auditory hallucinations. Archives of General Psychiatry, 64, 286–296.Find this resource:

Hoffman, R. E., Anderson, A. W., Varanko, M., Gore, J. C., & Hampson, M. (2008). Time course of regional brain activation associated with onset of auditory/verbal hallucinations. British Journal of Psychiatry, 193, 424–425.Find this resource:

Hoffman, R.E., Fernandez, T., Pittman, B., & Hampson, M. (2011). Elevated functional connectivity along a corticostriatal loop and the mechanism of auditory/verbal hallucinations in patients with schizophrenia. Biological Psychiatry, 69, 407–414.Find this resource:

Hoffman, R. E., & Hampson, M. (2011). Functional connectivity studies of patients with auditory verbal hallucinations. Frontiers in Human Neuroscience, 6, 6.Find this resource:

Hoffman, R. E., Hawkins, K. A., Gueorguieva, R., Boutros, N. N., Rachid, F., Carroll, K., & Krystal, J. H. (2003). Transcranial magnetic stimulation of left temporoparietal cortex and medication-resistant auditory hallucinations. Archives of General Psychiatry, 60, 49–56.Find this resource:

Hoffman, R. E., Varanko, M., Gilmore, J., & Mishara, A. L. (2008). Experiential features used by patients with schizophrenia to differentiate “voices” from ordinary verbal thought. Psychological Medicine, 38, 1167–1176.Find this resource:

Houde, J. F., Nagarajan, S. S., Sekihara, K., & Merzenich, M. M. (2002). Modulation of the auditory cortex during speech: An MEG study. Journal of Cognitive Neuroscience, 14, 1125–1138.Find this resource:

Hubl, D., Koenig, T., Strik, W. K., Garcia, L. M., & Dierks, T. (2007). Competition for neuronal resources: How hallucinations make themselves heard. British Journal of Psychiatry, 190, 57–62.Find this resource:

Jackson, J. H. (1958). Selected writings of John Hughlings Jackson. New York: Basic Books.Find this resource:

Jardri, R., Pouchet, A., Pins, D., & Thomas, P. (2011). Cortical activations during auditory verbal hallucinations in schizophrenia: A coordinate-based meta-analysis. American Journal of Psychiatry, 168, 73–81.Find this resource:

Jeon, Y-W., & Polich, J. (2003). Meta-analysis of P300 and schizophrenia : patients, paradigms, and practical implicaitons. Psychophysiology, 40, 655–831.Find this resource:

John, E. R. (2002). The neurophysics of consciousness. Brain Research Reviews, 39, 1–28.Find this resource:

Jones, N., & Luhrmann, T. M. (2015). Beyond the sensory: Findings from an in-depth analysis of the phenomenology of “auditory hallucinations” in schizophrenia. Psychosis.Find this resource:

Jones, S. R. (2010). Do we need multiple models of auditory verbal hallucinations? Examining the phenomenological fit of cognitive and neurological models. Schizophrenia Bulletin, 36, 566–575.Find this resource:

Kindler, J., Hubl, D., Strik, W. K., Dierks, T., & Koenig, T. (2010). Resting-state EEG in schizophrenia: Auditory verbal hallucinations are related to shortening of specific microstates. Clinical Neurophysiology, 122, 1179–1182.Find this resource:

Klingner, C. M., Langbein, K., Dietzek, M., Smesny, S., Witte, O. W., Sauer, H., & Nenadic, I. (2014). Thalamocortical connectivity during resting state in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 264, 111–119.Find this resource:

Kompus, K., Westerhausen, R., & Hugdahl, K. (2011). The “paradoxical” engagement of the primary auditory cortex in patients with auditory verbal hallucinations: A meta-analysis of functional neuroimaging studies. Neuropsychologia, 49, 3361–3369.Find this resource:

Koziol, L. F., Budding, D., Andreasen, N., D’Arrigo, S., Bulgheroni, S., Imamizu, H., . . . & Yamazaki, T. (2014). Consensus paper: The cerebellum’s role in movement and cognition. Cerebellum, 13, 151–177.Find this resource:

Langdon, R., Jones, S. R., Connaughton, E., & Fernyhough, C. (2009). The phenomenology of inner speech: Comparison of schizophrenia patients with auditory verbal hallucinations and healthy controls. Psychological Medicine, 39, 655–663.Find this resource:

Lattner, S., Meyer, M. E., & Friederici, A. D. (2005). Voice perception: Sex, pitch, and the right hemisphere. Human Brain Mapping, 24, 11–20.Find this resource:

Lennox, B. R., Park, S. B., Jones, P. B., & Morris, P. G. (1999). Spatial and temporal mapping of neural activity associated with auditory hallucinations. Lancet, 353, 644.Find this resource:

Li, T., Wang, Q., Zhang, J., Rolls, E. T., Yang, W., Palaniyappan, L., . . . & Liu, Z. (2016). Brain-wide analysis of functional connectivity in first-episode and chronic stages of schizophrenia. Schizophrenia Bulletin: sbw099.Find this resource:

Lindner, A., Thier, P., Kircher, T. T., Haarmeier, T., & Leube, D. T. (2005). Disorders of agency in schizophrenia correlate with an inability to compensate for the sensory consequences of actions. Current Biology, 15, 1119–1124.Find this resource:

McCarthy-Jones, S., Thomas, N., Strauss, C., Dodgson, G., Jones, N., Woods, A., . . . & Sommer, I. E. (2014). Better than mermaids and stray dogs? Subtyping auditory verbal hallucinations and its implications for research and practice. Schizophrenia Bulletin, 40 Suppl 4, S275–284.Find this resource:

Maher, B. A. (1974). Delusional thinking and perceptual disorder. Journal of Individual Psychology, 30, 98–113.Find this resource:

Marr, D. (1969). A theory of cerebellar cortex. Journal of Physiology, 202, 437–470.Find this resource:

Mathalon, D. H., & Ford, J. M. (2012). Neurobiology of schizophrenia: Search for the elusive correlation with symptoms. Frontiers in Human Neuroscience, 6, 136.Find this resource:

Mathalon, D. H., Roach, B. J., Ferri, J. M., Loewy, R. L., Stuart, B. K., Perez, V. B., . . . & Ford, J. M. (2018). Deficient auditory predictive coding during vocalization in the psychosis risk syndrome and in early illness schizophrenia: The final expanded sample. Psychological Medicine, 1–8.Find this resource:

Mathiak, K., Menning, H., Hertrich, I., Mathiak, K. A., Zvyagintsev, M., & Ackermann, H. (2007). Who is telling what from where? A functional magnetic resonance imaging study. Neuroreport, 18, 405–409.Find this resource:

Melloni, L., Molina, C., Pena, M., Torres, D., Singer, W., & Rodriguez, E. (2007). Synchronization of neural activity across cortical areas correlates with conscious perception. Journal of Neuroscience, 27, 2858–2865.Find this resource:

Mhuircheartaigh, R. N., Rosenorn-Lanng, D., Wise, R., Jbabdi, S., Rogers, R., & Tracey, I. (2010). Cortical and subcortical connectivity changes during decreasing levels of consciousness in humans: A functional magnetic resonance imaging study using propofol. Journal of Neuroscience, 30, 9095–9102.Find this resource:

Nayani, T. H., & David, A. S. (1996). The auditory hallucination: A phenomenological survey. Psychological Medicine, 26, 177–189.Find this resource:

Northoff, G., & Qin, P. (2010). How can the brain’s resting state activity generate hallucinations? A “resting state hypothesis” of auditory verbal hallucinations. Schizophrenia Research, 127, 202–214.Find this resource:

Oestreich, L. K., Mifsud, N. G., Ford, J. M., Roach, B. J., Mathalon, D. H., & Whitford, T. J. (2015). Subnormal sensory attenuation to self-generated speech in schizotypy: Electrophysiological evidence for a “continuum of psychosis.” International Journal of Psychophysiology, 97, 131–138.Find this resource:

Perez, V. B., Ford, J. M., Roach, B. J., Loewy, R. L., Stuart, B. K., Vinogradov, S., & Mathalon, D. H. (2012). Auditory cortex responsiveness during talking and listening: Early illness schizophrenia and patients at clinical high-risk for psychosis. Schizophrenia Bulletin, 38, 1216–1224.Find this resource:

Plaze, M., Bartres-Faz, D., Martinot, J. L., Januel, D., Bellivier, F., De Beaurepaire, R., . . . & Paillere-Martinot, M. L. (2006). Left superior temporal gyrus activation during sentence perception negatively correlates with auditory hallucination severity in schizophrenia patients. Schizophrenia Research, 87, 109–115.Find this resource:

Poulet, J. F., & Hedwig, B. (2006). The cellular basis of a corollary discharge. Science, 311, 518–522.Find this resource:

Powers, A. R., 3rd, Kelley, M. S., & Corlett, P. R. (2017). Varieties of voice-hearing: Psychics and the psychosis continuum. Schizophrenia Bulletin, 43, 84–98.Find this resource:

Powers, A. R., Mathys, C., & Corlett, P. R. (2017). Pavlovian conditioning-induced hallucinations result from overweighting of perceptual priors. Science, 357, 596–600.Find this resource:

Price, C. J. (2010). The anatomy of language: A review of 100 fMRI studies published in 2009. Annals of the New York Academy of Sciences, 1191, 62–88.Find this resource:

Raij, T. T., & Riekki, T. J. (2012). Poor supplementary motor area activation differentiates auditory verbal hallucination from imagining the hallucination. NeuroImage: Clinical, 1, 75–80.Find this resource:

Ramnani, N. (2006). The primate cortico-cerebellar system: Anatomy and function. Nature Reviews Neuroscience, 7, 511–522.Find this resource:

Rummell, B. P., Klee, J. L., & Sigurdsson, T. (2016). Attenuation of responses to self-generated sounds in auditory cortical neurons. Journal of Neuroscience, 36, 12010–12026.Find this resource:

Schmahmann, J. D. (1996). From movement to thought: Anatomic substrates of the cerebellar contribution to cognitive processing. Human Brain Mapping, 4, 174–198.Find this resource:

Schneider, D. M., & Mooney, R. (2018). How movement modulates hearing. Annual Review of Neuroscience, 41, 553–572.Find this resource:

Sergent, C., & Dehaene, S. (2004). Neural processes underlying conscious perception: Experimental findings and a global neuronal workspace framework. Journal of Physiology–Paris, 98, 374–384.Find this resource:

Shergill, S. S., Brammer, M. J., Williams, S. C., Murray, R. M., & McGuire, P. K. (2000). Mapping auditory hallucinations in schizophrenia using functional magnetic resonance imaging. Archives of General Psychiatry, 57, 1033–1038.Find this resource:

Shergill, S. S., Samson, G., Bays, P. M., Frith, C. D., & Wolpert, D. M. (2005). Evidence for sensory prediction deficits in schizophrenia. American Journal of Psychiatry, 162, 2384–2386.Find this resource:

Silbersweig, D. A., Stern, E., Frith, C. D., Cahill, C., Holmes, A., Grootoonk, S., . . . & Frackowiak, R. S. J. (1995). A functional neuroanatomy of hallucinations in schizophrenia. Nature, 378, 176–179.Find this resource:

Sitek, K. R., Mathalon, D. H., Roach, B. J., Houde, J. F., Niziolek, C. A., & Ford, J. M. (2013). Auditory cortex processes variation in our own speech. PLoS One, 8, e82925.Find this resource:

Slotema, C. W., Aleman, A., Daskalakis, Z. J., & Sommer, I. E. (2012). Meta-analysis of repetitive transcranial magnetic stimulation in the treatment of auditory verbal hallucinations: Update and effects after one month. Schizophrenia Research, 142, 40–45.Find this resource:

Sommer, I. E., Diederen, K. M., Blom, J. D., Willems, A., Kushan, L., Slotema, K., . . . & Kahn, R. S. (2008). Auditory verbal hallucinations predominantly activate the right inferior frontal area. Brain, 131, 3169–3177.Find this resource:

Sommer, I. E., Slotema, C. W., Daskalakis, Z. J., Derks, E. M., Blom, J. D., & van der Gaag, M. (2012). The treatment of hallucinations in schizophrenia spectrum disorders. Schizophrenia Bulletin, 38, 704–714.Find this resource:

Stirling, J. D., Hellewell, J. S., & Quraishi, N. (1998). Self-monitoring dysfunction and the schizophrenic symptoms of alien control. Psychological Medicine, 28, 675–683.Find this resource:

Suga, N., & Schlegel, P. (1972). Neural attenuation of responses to emitted sounds in echolocating bats. Science, 177, 82–84.Find this resource:

Tiihonen, J., Hari, R., Naukkarinen, H., Rimon, R., Jousmaki, V., & Kajola, M. (1992). Modified activity of the human auditory cortex during auditory hallucinations. American Journal of Psychiatry, 149, 255–257.Find this resource:

Turetsky, B., Colbath, E. A., & Gur, R. E. (1998). P300 subcomponent abnormalities in schizophrenia: II. Longitudinal stability and relationship to symptom change. Biological Psychiatry, 43, 31–39.Find this resource:

Umbricht, D., & Krlijes, S. (2005). Mismatch Negativity in schizophrenia: A meta-analysis. Schizophrenia Research, 76, 1–23.Find this resource:

van Lutterveld, R., Diederen, K. M., Koops, S., Begemann, M. J., & Sommer, I. E. (2013). The influence of stimulus detection on activation patterns during auditory hallucinations. Schizophrenia Research, 145, 27–32.Find this resource:

Wang, J., Mathalon, D. H., Roach, B. J., Reilly, J., Keedy, S. K., Sweeney, J. A., & Ford, J. M. (2014). Action planning and predictive coding when speaking. Neuroimage, 91, 91–98.Find this resource:

Washburn, M. F. (1916). Movement and mental imagery: Outlines of a motor theory of the complexer mental processes. Boston, MA: Houghton Mifflin.Find this resource:

Waters, F., & Fernyhough, C. (2017). Hallucinations: A systematic review of points of similarity and difference across diagnostic classes. Schizophrenia Bulletin, 43, 32–43.Find this resource:

Whitford, T. J., Jack, B. N., Pearson, D., Griffiths, O., Luque, D., Harris, A. W., . . . & Le Pelley, M. E. (2017). Neurophysiological evidence of efference copies to inner speech. Elife, 6.Find this resource:

Wilkinson, S. (2014). Accounting for the phenomenology and varieties of auditory verbal hallucination within a predictive processing framework. Consciousness and Cognition, 30, 142–155.Find this resource:

Woodruff, P. W., Wright, I. C., Bullmore, E. T., Brammer, M., Howard, R. J., Williams, S. C., . . . & Murray, R. M. (1997). Auditory hallucinations and the temporal cortical response to speech in schizophrenia: A functional magnetic resonance imaging study. American Journal of Psychiatry, 154, 1676–1682.Find this resource:

Woodward, N. D., Karbasforoushan, H., & Heckers, S. (2012). Thalamocortical dysconnectivity in schizophrenia. American Journal of Psychiatry, 169, 1092–1099.Find this resource:

Youn, T., Park, H. J., Kim, J. J., Kim, M. S., & Kwon, J. S. (2003). Altered hemispheric asymmetry and positive symptoms in schizophrenia: equivalent current dipole of auditory mismatch negativity. Schizophrenia Research, 59, 253–260.Find this resource:

Zmigrod, L., Garrison, J. R., Carr, J., & Simons, J. S. (2016). The neural mechanisms of hallucinations: A quantitative meta-analysis of neuroimaging studies. Neuroscience & Biobehavioral Reviews, 69, 113–123.Find this resource: